ML20126E074

From kanterella
Jump to navigation Jump to search
Licensee Contractor and Vendor Inspection Status Report. Quarterly Report,July-September 1992.(White Book)
ML20126E074
Person / Time
Issue date: 11/30/1992
From:
Office of Nuclear Reactor Regulation
To:
References
NUREG-0040, NUREG-0040-V16-N03, NUREG-40, NUREG-40-V16-N3, NUDOCS 9212280295
Download: ML20126E074 (142)


Text

.

l NUREG-0040 Vol.16, No. 3 Licensee Contractor anc Vendor Inspec1: ion Status Reaort Quarterly Report July-September 1992 U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation

,p"' " *%

[

g

gA2228!;8 '2"oo 0040IR.

.PDR-

AVAILABILITY NOTICE Availabihty of Reference Materials Cited in NRC Publications Most documents cited in NRC pubhcations will be availt.ble from one of the following sources:

1.

The NRC Public Document Room, 2120 L Street, NW., Lower Lovel, Washington, DC 20555 2.

The Superintendent of Documents U.S. Government Printing Office, P.O. Box 37082, Washirigton, DC 20013-7082 3.

The National Technical Information Sorvico. Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publica-tions, it is not intended to be exhaustiva.

Referenced documents available for inspection ano copying for a fee from the NRC Pubhc Document Room include NRC correspondence and internal NRC memoranda: NRC bulletins, circulars, information noticos, inspection and investigation noticos; licensoo event reports; vendor reports and correspondence; Commission paper =' and applicant and licensos docu-monts and correspondenco.

Tho following documents in the NUREG soi,es aie available for purchase from the GPO Sales Program: formal NRC staff and contractor reports, N3C-spensorod conforonco procood-ings, international agreement reports, grant publications, and NRC booklets and brochuros.

Also available are regulatory guidos, NRC rogu!ations in the Code of Federal Regulations, and Nuclear Regulatory Commission Issuances.

Documents available from the National Technical Information Servico includo NUREG-series reports and technical reports prepared by other Federal agencios and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.

Documents available from public and special technical hbraries includo all open hterature hems, such as books, journal articles, and transactions. Federal Register notices Federal and State logislation, and congrossional reports can usually be obtained from these libraries.

Documents such es thesos, dissertations, foreign reports and translations, and non-NRC conference proceedings are available for purchase from the organization sponsoring the pubhcation cited.

Single copios of NRC draft reports are available troo, to the extent of supply, upon written request to the Office of Administration, Distribution and Mail Services Section, U.S. Nuclear Regulatory Commission, Washington, DC 20555.

Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 NorfolK Avenue, Bethesde., Maryland, for use by the public. Codes and standards are usually copyrighted and may be purchased from the originating orgari,2ation or, if they are American National Standards, from the American National Standards institute,1430 Broadway Now York, NY 10018.

q

- NUREG-0040 Vol.:16,iNo. 3 L l

)

Licensee Contractor 1o and Vendor Inspection Status Report Quarterly Report July-September 1992

. Manuscript Completed: October 1992 Date Published: November 1992 h-Division of Reactor Inspection and Licensee Performance Ollice of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission

-Washington, DC 20555

+

T r*

u l

y v

+e4 m

..,. ~

~

p ABSTRACT-

,q This' periodical covers-the results of inspections performed;by

-the NRC's Vendor Inspection Branch.that have-been distributed-to:~

+

-the inspected organization during the period-from. July =1992

~through September 1992.

s e

i s

- t i

l, iii

TABLE OF CONTENTS PAGE Abstract iii Preface vii Index ix Inspection Reports 1

Selected Bulletins and Information Notices Concerning Adequacy of Vendor Audits and Quality of Vendor Products 120 Correspondence Relnted To Vendor Issues 121 i

r p

V

PREFACE A fundamental _ promise of the Nuclear Regulatory Commission (NRC) licensing and inspection program is that licensees are responsible for the proper construction and safe and officient operation of their nuclear power plants.

The total government-industry system for the inspection of commercial 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 compliance with requirements prescribed by the NRC's rules and regulations (Title 10 Code of Federal Regulations).

The NRC performs an overview of the commercial nuclear industry by inspection to determine whether its requirements _are being met by licensees and their contractors, while the major inspection effort is performed by the industry-within the framework of ongoing quality verification programs.

The licensee is responsible for developing and maintaining a detailed quality assurance (QA) plan with implementing procedures; pursuant to 10 CFR 50.

Through a system of planned and periodic audits and inspections, the licensee is responsible for assuring that suppliers, contractors and vendors also have suitable and appropriato quality programs that meet NRC requirements, guides, codes and standards.

The Vendor Inspection Branch (VIB) reviews and inspects nuclear-steam system suppliers (NSSSs), architect engineering (AE) firms, suppliers of_ products and services, independent testing.

laboratories performing equipment qualification tests, and holders of NRC licenses (construction permit holders and operating licenses) in vendor-related areas.

These inspections are performed to assure'that the root causes of reported vendor-related problems are determined and appropriate corrective actions are devolcoed.

The inspections also review the vendors' conformance with applicable NRC and industry quality requirements, the adequacy of licensees' oversight of their vendors, and that adequate-interfaces exist-between licensees and vendors.

The VIB inspection emphasis is placed on the quality and suitability of vendor products,-licensee-vendor interface, environmental qualification of equipment, and review of equipment problems found during operation and their corrective action.

When nonconformances with NRC requirements-and regulations are-found, the inspected organization is required to take appropriate corrective action and to institute preventive measures to-preclude recurrence.

When generic implications are identified, NRC assures that affected licensees are informed through vendor reporting or by URC generic correspondence such as information-notices and bulletins.

This periodical (White Book) is-published quarterly and contains copies of all vendor inspection reports issued during the calendar quarter for which it is published.

Each vendor vii

innpoetion report lists the nuclear facilities to which the results are applicable thereby informing licensees and vendors of In addition,-the affected Regional Offices potential problems.

notified of any significant problem areas that may require are special attention.

The White Book also contains a list of selected bulletinu and information notices involving vendor issues.

Copies of other pertinent correspondence involving vendor inrues are also included in this White Book issue.

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

viii

INDEX FACILITl REPORT 11UnllEB MU Copes-Vulcan, Inc.

99900080/92-01 1

Lake City, Pennsylv 41a Elgar Corporation 99900871/92-01 20 San Diego, California Florida Power Corporation 05000302/92-201 36 Crystal River, Florida GilB Industrial Battery Company 99901251/92-01 76 Lombard, Illinois Columbus, liebraska Westinghouse Electric Corporation 99900404/92-01 86 Nuclear and Advanced Technology Division Pittsburgh, Pennsylvania Wyle Laboratories 99900902/92-01 101 liuntsville, Alabama iX

4 1

J

- INSPECTION REPORTS h

o

-s '.

a

}

,)

m--

_.x--

1L Rf Cg*%

  1. gk

.t' '

UNITED STATES 3

C.

~2 NUCLEAR REGULATORY COMMISSION

[

WASHINGTON, D C 20%5 o

/

SEP 3 019M Docket No. 99900080 Mr. F. R. Robertson, President Copes-Vulcan, Inc.

P. O. Box 577 Lake City, Pennsylvania 16423-0577

Dear Mr. Robertson:

SUBJECT:

NOTICE OF VIOLATION AND NOTICE OF NONCONFORMANCE (NRC INSPECTION REPORT N0. 99900080/92-01)

This letter adda.sses the inspection of your facility at Lake City, Pennsylvania conducted by Mr. L. L. Campbell and Mr. C.-K. Battige of this office on August 31 through September 4,1992, and the discussions of their findings with memtsers of your staff at the conclusion of the inspection. The performance based inspection was conducted to evaluate Copes-Vulcan, Inc.'s (CV's) quality program and its implementation in selected areas such as (1) design processes and interfaces, (2) purchased material and services, (3) control of work processes, (4) inspection and tests, and (5) CV's commercial grade dedication program.

Areas examined during the NRC inspection and our. findings are discussed in the inspection report (Lnclosure 3). This inspection consisted of an examination of procedures and representative records, interviews with personnel, and observations by the inspectors.

During the inspection it was found that the implementation of your quality assurance (QA) program failed to meet certain NRC requirements. Although CV was implementing the essential elements of the dedication process, CV's QA manual and implementing procedures do not contain adequate requirements and interfaces to ensure that all items purchased as commercial grade items (CGIs) for use in safety-related applications are adequately dedicated as basic components.

As a result of this program deficiency, the NRC inspectors observed during the inspection some CGIs dedicated without adequately verifying critical characteristics, such as hardness and thread size, specified by CV design engineering and without conducting audits, required by-the CV QA manual, of suppliers controlling critical characteristics. This inspection also identified instances in which CV failed to follow procedure requirements for activities such as the processing of 10 CFR Part 21 evaluations, revising work procedures, and processing customer purchase orders. The specific findings and references to the pertinent requirements are identified in the enclosures of this letter.

I

Mr. F. R. Robertson The inspection also identified that certain of your activities appeared to be in violation of NRC requirements, as specified in the enclosed Notice of Violation (Enclosure 1). The violation identified that CV's procedures did not address the requirements of Section 21.21, " Notification," of Title 10 of the Code of Federal Regulations (10 CFR) as revised and effective on October 29,1991 (e.g., the 60 day period for evaluating potential defects and failures or filing an interim report was not addressed).

You are required to respond to this letter and should follow the instructions specified in the enclosed Notice of Violation when preparing your response, in your response, you should document the specific actions taken and any additional actions you plan to prevent recurrence. After reviewing your response to this Notice, including your proposed corrective actions and the results of future inspections, the NRC will determine whether further NRC enforcement action is necessary to ensure compliance with NRC regulatory requirements.

Further, please provide us within 30 days from the date of this letter a written statement in accordance with the instructions specified in the enclosed Notice of Nonconformance. We will consider extending the response time if you can show good cause for us to do so.

The responses requested by this letter and the enclosed Notices are not subject to the clearance procedures of the Office of Management and Budget as required by the Paperwork Reduction Act of 1980, Public Law No.96-511.

In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosures will be placed in the NRC Public Document Room.

If you have any questions concerning this inspection, we will be pleased to discuss them with you.

Sincerely, q

l l

n *

' ;? L ky }w V

~

y Leif J. Hofrho)m, Chief Vendor Inspection Branch Division of Reactor Inspection and Safeguards Office of Nuclear Reactor Regulation

Enclosures:

1.

Notice of Violation 2.

Notice of Nonconformance 3.

Inspection Report No. 99900080/92-01 2

l ENCLOSURE 1 NOTICE Of V10LAT10N Cope s-Vulc an. Inc.

Docket No.:

99900080/92-01 Lake City. Pennsylvania During an NRC inspection conducted August 31 through September 4, 1992, a violation of NRC requirements was identified.

In accordance with " General Statement of Policy and Procedure for NRC Enforcement Actions," of 10 CFR Part 2, Appc, dix C (1992), the violation is listed below:

Section 21.21, " Notification of Failure to Comply or Existence of Defect and its Evaluation," of 10 CFR requires, in part, that each corporation subject to the regulations adopt appropriate procedures for either evaluating deviations and failures to comply, or informing the licensee or purchaser of the deviation or failure to comply.

Contrary to the above requirements, CV had not revised its procedures, required by 10 CFR 21.21, to address the substantive revisions to 10 CFR Part 21 that became effective on October 29, 1991 (99900080/92-01-01).

The NRC has classified this violation as Severity Level IV (Supplement Vil).

Pursuant to the provisions of 10 CFR 2.201, CV is hereby required to submit a written statement or explanation to the U. S. Nuclear Regulatory Commission, AITN: Document Control Desk, Washington, D.C. 20555, with a copy to the Chief, Vendor Inspection Branch, Division of Reactor inspection and Safeguards, Of fice of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting this Notice of Violation. This reply should be clearly marked as a " Reply to Notice of Violation" and should include for each

~

violation (1) the reason for the violation, or if contested, the basis for disputing the violation, (2) the corrective steps that have been taken and the results achieved, (3) the corrective steps that will be taken to avoid further violations, and (4) the date when full compliance will be achieved. Where good cause is shown, consideration will be given to extending the response time.

Dated at Rockville, Maryl.and this day of

, ~, 1992 l

3

N011CE OF NONCONFORMANCE Copes-Vulcan, Inc.

nocket No.: 99900080/92-01 Lake City, Pennsylvania During an inspection conducted on August 31 through September 4, 1992, it appears that certain of your activities were not conducted in accordance with NRC requirements.

A.

Criterion 11', " Quality Assurance Program," of Appendix B to 10 CFR Part 50 requires that activities affecting quality be accomplished in accordance with a quality assurance program which shall be documented by written policies, procedures and instructions and that activities affecting quality shall be accomplished under suitably utrolled conditions which include the use of appropriate equipment including identifying the need for special controls, processes, test equipment, tools and skills to attain the required quality, and for verification of quality by inspection and test.

In addition, Criterion !!!, " Design Control," and Criterion VII, " Control of Purchased Material, Equipment, and Services," of Appendix B to 10 CFR Part 50 require that for items intended for use in safety-related applications, the important design, mater;al, and performance characteristics be identified, acceptance criteria be established, and reasonable assurance be provided that the items conform to the acceptance criteria.

Contrary to the above, the CV quality assurance (QA) manual and implementing procedures did not provide sufficient requirenents and interfaces to ensure that commercial grade items (CGis) dedicated as basic components would be adequately verified to be capable of performing their safety functions. Although the CV QA manual requires that CV design engineering identify critical characteristics and features of CGis to be verified during the dedication process, the quality assurance program did not identify written procedures, instructions, or guidance for (1) determining the critical characteristics of a CGI, (2) identifying the methods to verify that the critical characteristics are acceptable, and (3) the control and use of,'

and processing exceptions to, the Master Form CV-1981, " Commercial Grade Safety-Related items," data base used to identify methods for accepting selected valve part critical characteristics on shop work orders.

As a result of these program daficiencies, the dedication of the yoke to bonnct bolting material for G control valves being supplied to Virginia Power and processed by CV Shop Order No. 2097320, dated July 22, 1992, did not include verification of all the critical characteristics identified by CV design engineering for load bearing fasteners.

Additionally, a replacement internal valve plug spring being supplied to Baltimore Gas and Electric Company (BG&E) and processed by CV Shop Order 4

No.-1196213,'dcted July 23, 1992, eas dedicated eithout verifying the.

spring constant (99900080/92-01-02).

B.

Criterton V, _"Insttuctions~, Procedures, and Drawings," of Appendix B to 10 CFR Part 50 requires, in part, that activities affecting quality be accomplished in accordance with instructions, procedures, or drawings.

The following quality related activities were not accomplished in-accordance with the applicable CV procedure requirements (99900080/92-01-03):

1.

Section 3, " Documents Used On More Than One Job," of CV Procedurc No. l-6.06, " Document Control," dated July 15, 1992, requires tiat the review and approval of procedures shall be denoted on the revision page of the procedure, and that the' distribution of revised procedures shall be performed by the issuing department using CV Form 705158, " Document Submittal," or Form 705168,

" Verification of Print Receipt," and requires that obsolete procedures be collected and destroyed by the issuing department.

Contrary to the above, Procedure No. 50-5.9.9 was revised and placed in Shop Order No. 1196213 package for.the dedication of a spring without receiving all the required reviews and approvals and being distributed and controlled as required by Procedure No. 1-6.06.

It is also noted that Paragraphs No. 7.2'and 7.3 of Revision No. 15 to' Procedure No. 50-5.9.9, which provided test, analysis, and identification requirements for material standards, were inadvertently omitted from Revision No. 16 to Procedure No. 50-5.9.9.

2.

Paragraph 5.9 of Procedure No. 1-6.03, " Contract Management,"

requires that the contract engineer prepare procedures and instructions that are required _to support a customer purchase order (PO) that are in addition to the; requirements of the CV QA manual. BG&E PO 746776X, processed in accordance with CV Shop Order No. 1196213 requires that if the springs are dedicated as basic components a dedication program meeting the requirements of EPRI NP-5652, " Guideline for the Utilization of Commercial Grade items in Nuclear Safety Related Applications (NCIG-07)," and NRC Generic letter (GL) 89-02, " Actions to Improve the Detetion of Counterfeit and Fraudulently Marked Products," dated March 21, 1989, be.used.

Contrary to the above, the dedication of the springs being supplied to BG&E was performed to the existing _CV dedication program, which does not address all the requirements. of EPRI-5252 and NRC GL-89-02, and procedures were not prepared to meet the dedication requirements of the BG&E purchase order as required by Paragraph 5.9 of Procedure No. 1-6.03. 5

3.

Section 5, " Action," and Section 6, " Evaluation," of Procedure No. 1-6.26, " Identification, Evaluation, and Notification Requirements Per-10 CFR Part 21," dated August 16, 1991, requires, in part, that an Evaluation Committee and Panel meet and investigate potential deviations and defects and that the proceedings be documented on Form QC-37, "10 CFR 21 Evaluation Report."- This form identifies the discovered condition and committee and panel members evaluating the condition, and requires an evaluation report be prepared with a format in accordance with Form QC-38, "10 CFR 21, NRC Notification Report."

Contrary to the above, the CV 10 CFR Part 21 package for evaluatin,g incorrect bolting material for yoke to bonnet on CV control valves supplied to Texas Utilities Electric's Comanche Peak Electric Station did not include the Form QC-38 evaluation format and Form QC-37 did not indicate that the CV president and vice president of engineering had participated in the evaluation as required by Procedure No. 1-6.26.

Please provide a written statement or explanation to the V. S. Nuclear Regulatory Commission, ATTN: Document Control Desk, Washington, D.C. 20555, with a copy to the Chief, Vendor Inspection Branch, Division of Reactor Inspection and Safeguards, Offise of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting this Notice of Nonconformance.

This reply should be clearly marked as a " Reply to a Notice of Nonconformance" and should include for each nonconformance:

(1) a description of steps that have been or will be taken to correct these items; (2) a description of steps that have been or will be taken to prevent recurrence; and (3) the dates your corrective actions and preventive measures were or will be completed.

Dated at_Rockville, Maryland, this

~

day of 5, rr-

, 1992.

I a 6

ORGANIZATION:

_ Copes-Vulcan, Inc., Lake City, Pennsylvania-REPORT NO.:

99900080/92 !

CORRESPONDENCE ADDRESS:

Mr. F. R. Robertson, President Copes-Vulcan, Inc.

P.O. Box 577 Lake City, PA 16423-0577 ORGANIZATIONAL CONTACT:

Mr. J. R. Scarpelli NUCLEAR INDUSTRY ACTIVITY:

Manufactures and supplies valves and valve parts for commercial nuclear power plants.

INSPECTION CONDUCTE0:

August 31 - September 4, 1992 4W Y#f?2 L. L/tampbell, Team Leader Date Reactive Inspection Section No. 1 4

Vendor Inspection Branch (VIB)

OTHER INSPECTORS:

C. K. Batti f

APPROVED:

-o NWP 9 -22 9 7_.

Uldis Potapovs, Chief i

Date Reactive Inspection Section No. 1 Vendor Inspection Branch INSPECTION BASES:

10 CFR Part 21 and Part 50, Appendix B INSPECTION SCOPE:

To review and evaluate the Copes-Vulcan, Inc.

(CV) quality assurance (QA) program and its implementation in selected areas such as (1) design processes and interfaces,-(2) purchased material and services, (3) inspection and test, and (4) CV's commercial grade dedication program.

PLANT SITE APPLICABILITY:

Calvert Cliffs (50-317/318)

Comanche Peak-(50-445/446)

Diablo Canyon (50-275/323)

H. B. Robinson (50-261)

Surry (50-280/281)

Trojan (50-344)

Other plants using'CV valves 7

j

l 1

INSPECTION

SUMMARY

1.1 Violations Contrary to the requirements of Title 10 of the Code of Federal Reaulations (10 CFR) Part 21, Copes-Vulcan, Inc.'s (CV's) Procedure No.1-6.26,

" Identification, Evaluation, and Notification Requirements Per 10 CFR Part 21," dated August 16, 1991, failed to implement the requirements of the October 29, 1991, revision of Section 21.21(a) of 10 CFR Part 21 (Violation 99900080/92-01-01, see Section 3.2 of this report).

1.2 Nonconformances Contrary to Criteria II, III, and VII of Appendix B to 10 CFR Part 50, the CV quality assurance (QA) manual and implementing procedures did not provide sufficient requirements or interfaces to ensure that commercial grade items (CGls) would be adequately verified as being capable of performing their safety functions.

Contrary to Criterion V of Appendix B to 10 CFR Part 50 and to the requirements specified in CV procedures, the NRC inspectors identified three instances where CV failed to follow procedure requirements (Nonconformance 99900080/92-01-03, see Sections 3.2.1 and 3.4.3.2 of this report).

2 STATUS OF PREVIOUS INSPECTION FINDINGS 2.1 Violation 99900080/88-01. Item 88-01-01 (Closed)

Violation 88-01-01 stated, in part., that contrary to Section 21.21 of 10 CFR Part 21, CV failed to notify all of its commercial nuclear customers of incorrect center-of-gravity and weight information that was provided to them on valve assembly drawings so that the customers could notify appropriate licensees or perform an evaluation.

Following the NRC inspection in November 1988, CV issued letters to its nuclear customers in December 1988, informing them that CV had determined that the valve weight and/or conter-of-gravity referenced on some valve assembly drawings was incorrect.

2.2 Nonconformance 99900080/88-01-02 (Closed)

Nonconformance 88-01-02 stated, in part, that contrary to Criterion III,

" Design Control," of Appendix B to 10 CFR Part 50, CV f ailed to have its valve assembly center-of-gravity and weight data independently verified for technical adequacy.

The NRC inspectors randomly selected and reviewed the following calculations performed by CV and verified that they had been independently verified as being technically adequate:

Referenced Shop Order (50) Mo. 7710-69609-1-1, 12 inch cast valve,

  • 1 8

y pressure class 300#, with a D-100-160 actuator, was calculated on June 21, 1990 and independently checked on June 25, 1990.

SO No. 8920-97290, 3 inch cast valve, pressure class 150#, with a D-100-100 actuator, was calculated on January 29, 1990, and independently checked on February 1, 1990.

S0 No. 8920-97290, 1 inch cast valve, pressure class 150#, with a 0-100-60 actuator, was calculated on March 15, 1990, and independently checked on March 18, 1990.

S0 No. 9110-63584, 6 inch cast valve, pressure class 900#, with a 0-100-160 DA actuator, was calculated December 31, 1991, and independently checked on February 11, 1992.

3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 Entrance and Exit Meetinos In the entrance meeting on August 31, 1992, the U. S. Nuclear Regulatory Commission (NRC) inspectors discussed the scope of the inspection and established interfaces with CV management. During the exit meeting on September 4, 1992, the NRC inspectors discussed their findings and concerns with CV's management and staff.

3.2 10 CFR Part 21 The NRC inspectors determined that CV has maintained the required 10 CFR Part 21 postings and a procedure for implementing Part 21 requirements, Procedure No.1-6.26, " Identification, Evaluation and Notification Requirements Per 10 CFR Part 21," dated August 16, 1991. CV informed the NRC inspectors that they did not have copies of NRC Information Notice 91-76, 1

"10 CFR PARTS 21 AND 50.55(e) FINAL RULES," dated November 26, 1991, nor the revised 10 CFR Parts 21 and 50 that became effective October 29, 1991, but a Nuclear Utilities Procurement Issues Council newsletter had made them aware of the October 1991 changes. CV, however, failed to revise Procedure No. 1-6.26 to address the revised requirements to 10 CFR Part 21 such as the 60 day limit for completing an evaluation, or filing an interim report, and 2he proper location for contacting the NRC (the NRC Operations Center, Bethesda, MD).

CV made a verbal commitment to the NRC inspectors to revise Procedure No.1-6.26 to conform with the new requirements of 10 CFR Part 21 (Violation 99900080/91-01-01).

3.2.1 Imolementation of 10 CFR Part 21 Procedures The NRC inspectors reviewed CV's 10 CFR Part 21 evaluation for the J

5/8 inch x l-1/8 inch long x 11 UNC socket head cap screws that were reported to the NRC by Texas Utilities Electric Company (TV) in 10 CFR-Part 21 Report No. P21R-92-002, dated February 13, 1992, for the Comanche Peak Steam Electr i Station (CPSES). TV indicated that unlike the material specified in CV's drawings, ASTM A193, Grade B6, the installed material was not ferromagnetic, 1 9

and that testing of fasteners removed from valves in CPSES, Unit 2, identified fastener material composition and properties associated with ASTM F837, type XM7 material, it is noted that the CV valves having the improper fastener material were supplied to TU and other NRC licensees by the Westinghouse Electric Company.

In a letter to CV, dated November 26, 1991, Westinghouse E17ctric Company, Energy System Business Unit (WESBU) originally indicated the fasteners at CPSES were ASTM A193, Grade B8, material (type 304 stainless steel, nonmagnetic). WESBU later identified ;he f astener material used as ASTM F837, type XM7, Unified Numbering System (UNS) S30430, which has a composition very similar to type 304 stainless steel (UNS S30400) with the addition of 3.00 to 4.00 percent copper.

The NRC inspectors questioned CV concerning the WESBU material classification of the fasteners, which were installed on valves built in the late 1960's or 1970's, since ASTM F837, a standard for socket head cap screws, did not exist until 1988. CV presented a chemical analysis on the actual bolts removed from service, performed by the Materials Research Division of Modern Industries, Inc., Erie, Pennsylvania, dated January 31, 1991, which indicated the chemical composition matched that specified for XM7 material, a material standard that existed during the 1960's and 1970's. CV indicated that the fasteners were not built to ASTM F837: rather, the material supplier had inadvertently supplied XM7 instead of ASTM A193, Grade B6, fasteners.

The specified yoke to bonnet fastener material, ASTM A193, Grade B6, is a ferritic (and therefore ferromagnetic) stainless steel with a specified yield strength of 85 ksi. Alloy XM7 material is an austenitic stainless steel (nonmagnetic) with a yield of approximately 59 ksi. CV performed an evaluation of the yoke to bonnet connection under a seismic loading of 3.0 g horizontal and 2.0 g vertical to the largest actuator (D-160) used on the subject valves. The calculation results irdicated the yoke to bonnet bolts would experience stresses less than 30 ksi.

CV determined that the worst case valve operability acceptance criteria were met. The NRC inspectors reviewed CV's calculations and found them to be adequate.

It is noted that CV's analysis did not address the possible effects of a change in natural frequency of the valves from yielding of the fasteners and preloads due to complexities of modeling the yoke to bonnet joint. The NRC reviewed documents at CV which indicateu that WESBU made an evaluation that considered these effects and found the fasteners to be adequate. The WESBU evaluation was not available to the NRC inspectors for review, however, the NRC inspectors reviewed WESBU Letter ET-NSI-MFSL-92-175, " Customer Notification for Resolution of P1-92-002 Regarding incorrect Bolting Material for Yoke to Bonnet on Copes-Vulcan Control Velves," dated March 16, 1992.

This letter recommends that " Licensees evaluate the safety significance of this issue if modifications such as valve restraints / supports or operator modifications have been made to the valves beyond th ' original design conditions," and also states "Due to the time period of manufacture of these valves, when records were not required to be maintained for commercial grade parts, Westinghouse further recommends replacing all Copes-Vulcan control valve yoke to bonnet bolts as soon as practical utilizing the material as listed in the bills-of-material on the Copes-Vulcan specific outline drawings." 10

The NRC inspectors also reviewed the implementation requirements of Procedure No. 1-6.26, which requires, in part, that an Evaluation Committee and Panel meet and investigate potential deviations and defects and that the proceedings be documented on Form QC-37, "10 CFR 21 Evaluation Report," which identifies the discovered condition and committee and panel members evaluating the condition, and that an evaluation report be prepared with a format in accordance with Form QC-38, "10 CFR 21, NRC Notification Report." Contrary to the above, the CV 10 CFR Part 21 package for evaluating incorrect belting material for yoke to bonnet on CV control valves supplied to the CPSES did not include the Form QC-38 evaluation format and form QC-37 did not indicate that the CV president and vice president of engineering had participated in the evaluation as required by Procedure No.1-6.26.

Although the NRC inspectors reviewed CV's technical evaluation and believed it to have been adequate, this is an instance in which procedure requirements were not followed (Nonconformance 99900080/92-01-03).

3.3 CV's Actions Relatives to Licensee Event Reports (LERs) and Part 21 Reports The NRC inspectors reviewed CV's actions relative to LERs and 10 CFR Part 21 reports which identified deficiencies asscciated with valves and valve parts supplied by CV to the Comanche Peak Steam Electric Station, the H.B. Robinson Steam Electric Plant, and the Trojan Nuclear Plant.

The most significant of these reports identified that incorrect valve yoke to bonnet fasteners with a yield strength of approximately one half of the material specified on the valve drawing had been installed in CV control valves.

3.3.1 LER 261/91-001-00. "Inonerable Pressurizer Power Operated Relief Valves." dated February }8.1991 LER 91-001-00 was initiated by Carolina Power & Light Company (CP&L) when CV notified CP&L that certain items, supplied by CV and installed in the H. 8. Robinson Steam Electric Plant's pressurizer power operated relief valve (PORV) operators, were not qualified as safety-related parts as required by CP&L's P0. CV informed the NRC inspectors that CP&L had called CV's Lake City, Pennsylvania facility on a Saturday, and reached someone in accounting who just happened to be in.

The CV accountant informed CP&L that he couldn't help them with their verbal request.

Later that Saturday, CP&L called CV's Lake City, Pennsylvania Part Service Department and placed a verbal order, which according to CV, was to provide valve operator parts consisting of a diaphragm base, diaphragm cover, bolts, and nuts for each of the PORVs, and that the parts should be certified as being the correct part number. CV informed CP&L that the parts were available at CV's Charlotte, North Carolina plant and that CP&L could pick up the parts with a Certificate of Conformance (CoC) for part number only. CV emphasized to the NRC inspectors that this CP&L verbal order was not considered a nuclear safety-related order and that CV personnel involved in the telephone calls with CP&L were not aware of any nuclear safety-related aspects of the parts supplied.

CP&L obtained the PORV parts and CoC from CV's Charlotte, North Carolina plant on January 5, 1991. CP&L accepted the parts and CoC, which certified that the 11

~

parts were the correct part numbers, and met the requirements of the CP?d. P0.

The CoC was dated January 1, 1991, and, according to CV, the CoC was certified to the verbal requirements (part number only) understood by CV's Part Service representative.

The CV supplied parts were installed and the PORVs returned to servke January 12, 1991. On January 24, 1991, three weeks after the verbM P0 to CV, CV received a written P0 from CP&L for the PORV parts which invokee the requirements of 10 CFR Part 21 and Appendix B to 10 CFR Part 50.

Upcn receipt and review of the PO by CV's Sales department, CV call CP&L and inforr*d them that CP&L's written P0 did not match their verbal order.and that the-perts obtained from the CV Charlotte, North Carolina plant were not produced under the QA program required by the written CP&L P0.

Based on a review of LER 91-001-00, discussions with CP&L personnel prior to the conduct of this inspection in which CP&L indicated-that CV was informed verbally that Appendix B to 10 CFR Part 50 applied, and discussions with CV 1

personnel during the-inspection, 9e NRC inspectors concluded that the resulting miscommunication bet w n CP&L and CV was due in part to (1) the verbal order being taken by CV personnel who were not that familiar with or responsible for receiving safety-related P0s, and therefore may not have fully understood CP&L's verbal P0, (2) CP&L's three week delay in finalizing the written P0 and formally contacting CV's sales department and (3) CP&L not following up on their verbal P0 the following Monday with CV's Sales department.

It is noted the CV was unaware of the LER ft:ad by CP&L concerning the PORV replacement parts.

3.3.2 LfR 344/92-002-00. "Inadeauate Disc Nut Lockina Device Desian Results in Residual Heat Removal Pumo Discharae Check Valve Failure." dates!

February 27. 1992 LER 92-002-00 was initiated by the Portland General Electric Company (PG&E) when the disc nut lockwires on both Residual Heat Removal (RHR) pump discharge check valves were found broken. On December 20, 1991, while the Trojan Nuclear Plant was in Mode 5 (Cold Shutdown) during the 1991 refueling outage,-

routine in-service testing of the RHR pumps was being done. While the "A" train pump was running by itself, operators noticed that the indicated flow was lower and the electrical current drawn by the notor was higher-than those of the "B" train pump.

In addition, the "B" train pump rotated backwards while the "A" train pump was operating. On January 21, 1992, the "B" train RHR pump discharge check valve was inspected, and it was-found that the nut, washer and lock wire-that fasten the disc to the swing arm were missing. On January 28, 1992, the "A" train RHR pump discharge check valve was inspected and the nut, washer and lock wire were in place, but the nut was not tight and thn lock wire was broken. The cause was determined by PG&E to be inadequate design of the disc nut locking device which allowed the nut to rotate between the limits of the small diameter lockwire. Motion of the nut induced low stress, high cycle fatigue on the lockwire that resulted in its failure. The-valves were repaired by torquing the disc retaining nut and replacing the lockwire with a 1/8 inch cotter pin, PG&E indicated that the check valve.

12

inspection program wSil be + w 15ed to req & 9 documenting the type'and condition of the fastener locking.4eviceand s ability to limit rotation'of-the fastener.

-CV informed the NRC inspectors that PG&E had called them concerning the i

lockwire in fV supplied check valves and had discussed the feasibility of using a pin to secure the disc nut. CV indicated that they receive several calls each week'regarding CV supplied valves and very often the inquirers do not discuss the particular circumstances associated with the inquiry. CV was unaware that PG&E had submitted LER 92-002-00 to the NRC and that PG&E considered the lockwire failure to be significant and the result of an inadequate design.

CV informed the NRC inspectors that since they are now aware of the reported event in PG&E's LER 92-002-00, a 10 CFR Part 21 evaluation of the reported condition will be performed in accordance with CV Procedure No.1-6.26.

3.3.3 10 CFR Part 21 Report No. P21R-92-002. " Comanche Peak Steam Electric Station (CPSES) Docket No. 50-445 and 50-446. Yoke to Bonnet Material.

Copes-Vulcan Valves." dated February-13. 1992 On November 11, 1991, during CPSES, Unit 2, construction activities it was discovered that the 5/8 inch x l-1/8 inch long yoke to bonnet socket head cap screw material installed in CV valves was not the material specified on the CV~

drawings, ASTM A 193, Grade B6, but was a material having properties similar to ASTM F837, type XM7, material. Section 3.2.1 of this report discusses the results of the NRC' inspector's review of CV's 10 CFR Part 21 evaluation package for TV's 10 CFR Part 21 Report No. P21R-92-002.

It is noted that although TV-determined that no substantial safety hazard exists at CPSES due to the incorrect yoke to bonnet fastener material installed in safety-related CV control valves, TV indicated that the condition was reported to the NRC because it could affect other licensees.

3.3.4 LER 275/90-017-01. " Reactor Trin Resultino from Failed Ooen Pressurizer Sprity_1glve Due to Incorrect Screw Installation." dated Ocfober 23. 1991 LER 90-017-01 was initiated at Diablo Canyon Power Plant (DCPP), Unit 1, when a pressurizer spray valve, RCS-1-PCV-4558, failed 'open when-its feedback linkage disconnected as the proper locking desice had not been installed. As-a' result, the reactor coolant system (RCS) (.coled down at a rate greater than allowed by the Technical Specifications.

The licensee, Pacific Gas and-Electric Company (PG&E), indicated that the most significant contributor to the overcooling after a reactor trip on low pressurizer pressure was the failure of the.mn safety-related 40 percent steam dump valve, SDV PCV-1, supplied by CV. Further, PG&E indicated this valve stuck open due to the

-failure of the pilot valve stem, which allowed steam flow to bend the-valve main stem and caused the main plug to be held' partially open..The licensee also indicated that the main valve plug stuck'to its seat due to "microwelding," with the root cause an inadequate design of trim components by the manufacturer, CV. 13

- ~. -

~

The NRC inspector _s discussed this nonsafetyJrelated valve f ailure-with-CV engineering and were informed that CV had originally supplied the steam dump.

valves.with _a singleiangle seat design,Thich had a_ relatively large area of' contact <- At_ the request of PG&E, the valve trim was modified to a dual angle seat contact, whereby -a-single line of contact is formed-between the. trim:and seat. CV engineering indicated this was~done presunably to provide a leak-tight seal for the valve..llowever, with -the material used, type. 410' stainless -

.iteel, galling may occur in service, which may lead to;the valve sticking.-

PG5E's steam dump valves were.again: modified by CV to a design-similar totthe-original, where a-single line of contact is' eliminated by use of a single:

angle contact seat.

The NRC inspectors were informed by CV that the current-design utilizes-type 420 stainless steel, which has less tendency to gall, ~ stick, or'. _"microweld."

CV indicated that a possible reason for valve sticking was related to installation / maintenance of certain valves which are installed upside-downL in the plant.

In this situation, mechanics assembling the-valve or performing maintenance may inadvertently cause the trim and seat to develop scratches or gall, which may lead to sticking at a later time. CV indicated-that they-work" closely with licensees to assure proper operation of their valves. As a '

result of this latest modification, no further problems have been encountered associated with the-40 percent steam dump valves at DCPP.

3.4-CV Commen_lai GrdLItem Oediradon Proats 3.4.1 Methodology CV presently procures CGIs to be dedicated for safety-ralated applicationL from suppliers who have not been audited or surveyed and then performs a standard receipt inspection when the_ CGis-are received, normally-consisting of( a review of documentation, marking z.nd identification, and damage checks. Oncez the CGI has been. accepted, it is placed in stock or_ sent to the _ shop for processing in' accordance with the. dedication requirements identified in the shop-work order package. CV's dos.ign engineering department;has developed'a listing of:

various valve parts, which are not part off the ASME -Section III Code boundary, and identified each part's critical. characteristics 'and features-that need to '

be verified. CV design: engineering has~ prepared _ a data base, ~ referred to as -

the Master Form CV-1981, " Commercial Grade-Sa' ty-Related : Items," which? lists characteristics and features to be verified by the.specified generic valve part (i.e., fasteners, screw, stem). A CV record' technician then translates dedication requirements from the Master form CV-1981 into the shop work order-

. package. These dedication activities are performed by CV or CV~ approved subsuppliers in.:accordance with the requirements of Appendix B to-

- 10 CFR Part150 with110 CFR Part:21 being invoked on CV subsuppliers.

Parts n

whose critical characteristics and features are acceptable lare considered dedicated.

3.4.2 hlity Assurance Manual ' and -Imolementino _ procedures The NRC' inspectors reviewed the follo' wing documents and discussed their content' with CV in order to evaluate the process used for dedicating items

- procured as CGIs and supplied to CV's nuclear customers as basic components:

7-7 14 y

e-r'am m

'ae e

-oww ai

i

~

Procedure No. 1-6.03, iContract Management," dated July 15, 1992-

=

Procedure No.t 1-6.04, "Qu'ality Control-Inspection Point Plan, Document-Control Sheet & Shop _ Order,"'--dated July _15, 1992 y

c o

Procedure No. 1-_6.05, " Purchasing Control," dated August 16,-1991 a

Procedure No. 1-6.05.01, " Vendor Control," dated July 15, 1992 i

-=

Procedure No. 1-6.06, " Document Control," dated July 15, 1992 Procedure No. 1-6.07, " Work Order Router Sheet," dated July 15, 1992 Procedure No. 1-6.25, " Design ~ Control," dated July 15, 1992 Procedure No. 1-6.26, " Identification, Lyaluation and Notification-Requirements Per 10 CFR Part 21," dated August 16, 1991 Although Procedure No.1-6.26 requires that CV design engineering identify-critical characteristics and features of_ CGIs to be verified during the dedication process, there are no CV procedures that provide requirements'or guidance for (1) determining the critical characteristics of a CGI, (2) identifying the methods to verify that the critical characteristics are acceptable, (3) the control and use of and processing exceptions-to the Master:

Form CV-1981 data base used to identify methods for accepting-selected valve -

part critical characteristics, and (4) incorporating the form CV-1981 requirements into shop orders.

As a result of these program deficiencies, the dedication of the yoke to-bonnet bolting material for CV control valves being supplied to Virginia Power and processed by CV Shop Order-No. 2097320 did not include verification of all the critical characteristics identified by CV design engineering for load bearing fasteners.

It appeared to the NRC inspectors-thatidue to inadequate procedure guidance, personnel incorrectly selected the verification of critical characteristics for (non load-bearing) screws from the Master form CV-1981. The characteristics for'" screws" do not include strength, which would be applicable 'for the load bearing yoke to bonnet-fasteners.

Additionally, a replacement internal valve plug spring being supplied to BG&E l-and processed by CV Shop Order No. 1197320 was dedicated without' verifying the L

spring constant.

It is noted that although CV performed an audit in 1989.of the spring manufacturer, Duer Spring and: Manufacturing Co., under their Navy.

Nuclear Audit Program and informed the-NRC inspectors that they did evaluate the spring manufacturer's: control of -the spring constant, CV's QA.manualedoes.

not permit the use of this-type of an audit to control critical characteristics. _ Therefore, the acceptability of the spring constant was not verified in accordance with CV's 10 CFR Part:50, Appendix 0,.QA program requirements (see Nonconformance 99900080/92-01-02 and Sections 3.4.3.1 and 3.4.3.2 of this report).

L 15 m

w..

3.4.3 _ Review o_f_CV Dedica11on Pat _a_ge3 k

During the conduct of the inspection, the NRC inspectors. observed the dedication of 'an internal valve plug spring being supplied to BG&E and a-control valve assembly being supplied to Virginia Power, with various parts being dedicated, including the 5/8 inch yoke to bonnet fasteners discussed in TV's 10 CFR Part 21 Report No. P21R-92-002 (see Section 3.3.3_ of this report).

1he NRC inspectors also reviewed the documents such as the customer's PO to CV, selected audits of CV's subsuppliers, subsupplier furnished documentation, shop work order packages, and CV procedures for the valve plug spring and the yoke to bonnet fasteners. The results of these observations and reviews are presented in the following paragraphs.

3.4.3.1 CV Shop Order 2097320. 3 Inch. M0 lb Class. Control Valve Assembly Virginia Power issued P0 No. CNT375135, dated February 24, 1992, for the purchase of four, 3 inch, 150 lb. class, air operated head pressure control valves used to throttle the flow of brackish service water to the control room, and emergency switchgear room chillers at the Surry Power Station. The Virginia Power P0 invoked the requirements of 10 CFR Part 21 and Appendix B to 10 CFR Part 50, however, based on documents presented to the NRC inspectors, unlike the BG&E P0, there were no requirements for CV to comply with EPRI NP-5652 or NRC Gl. 89-02.

Following the assembly of these valves the NRC inspectors and a Virginia Power QA specialist observed the seat leakage test for one of the control valves, S/N 9920-97320-3-2 (Virginia Power-Valve Tag No. 1-SW-PC-101E). With approximately 45 lb.-force of air to the operator, the observed seat leakage was 0.6 cc per minute, and was within the 200 cc per minute acceptance criteria. During the conduct of the seat leak test the NRC inspectors noted that the control valve was of a configuration which used 5/8 inch x l-1/8 inch x 11 UNC yoke to bonnet socket head cap screws that were identical to those fasteners identified as being the incorrect material on TU 10 CFR Part 21 Report No. 21R-92-002 (see Section 3.2.1 of this report).

The NRC inspectors reviewed the activities performed by CV to dedicate the yoke to bonnet fasteners and determined that as a result of CV's dedication program deficiencies discussed in Section 3.4 of this report, the dedication-of the yoke to bonnet fasteners was inadequate. Based on discussions with CV personnel involved in the preparation of the shop order package for the-control valve assembly, critical characteristics for a.non load bearing screw were-identified as the applicable critical characteristics for the load bearing yoke to bonnet fasteners. This error, identified by the NRC-inspectors, appeared.to be caused, in part, because the-5/8 inch load bearing-yoke to bonnet fasteners, item 12 on.CV-Drawing No. D-339652, Revision 1, are identified as socket head _ cap screws. 'CV personnel incorrectly selected the verification of critical characteristics for non load bearing screws from the Master Form CV-1981, and not the critical characteristics such as hardness,__

J being visually free from cracks, and thread size listed for fasteners as being applicable.for the load bearing yoke to bonnet fasteners. Following the discussions on the use of the incorrect critical characteristics, CV QA personnel indicated that the identified condition would be evaluated and necessary corrective action taken. At the conclusion of the NRC inspection,

_g.

16

CV had_not completed its evaluation of this discrepancy to determine the acceptability of the installed fasteners (see Nonconformance 99900080/92 02).

The NRC inspectors reviewed the control and use of the Master form CV-1981 and determined that there were no procedure requirements for its preparation and use, and for translating form CV-1981 requirements into shop work order packages. The NRC inspectors also determined that procedures for preparing and processing the documents in the shop work order package did not address the use of form CV-1981 (see Nonconformance 99900080/92-01-02).

The NRC inspectors also reviewed the valve bonnet material's physical and chemical analysis performed by Material Research Division of Modern Industries, Inc., an audited and approved 10 CFR Part 50, Appendix B, subsupplier. The analysis indicated that the bonnet material was ASTM-A-182, Grade F316, and met the material requirements specified on CV Drawing No. D-339652, Revision 1.

3.4.3.2 Internal Valve Plua Sprina BG&E issued P0 74677GX, dated July 7, 1992, to CV for the purchase of five springs for valves on CV Drawing No. B-143178, item 13, Part No. 185847, alloy _

A-286, for the Model D-100-100 control valve.Section V, " Quality Requirements," of the BG&E P0 invoked the requirements of Appendix B to 10 CFR Part 50 with a provision that if.CV procures items from subsuppliers not-audited to the requirements _of Appendix B.to 10 CFR Part 50 (or equivalent),

CV's QA program and implementing procedures shall address the_ control of these items and shall meet the intent of EPRI NP-5652 and NRC GL 89-02.

Based on the review of the CV QA manual and implementing procedures, the NRC inspectors determined that CV's dedication program did not meet the requirements of Appendix B to 10 CFR Part 50 or Section V of BG&E's P0 because the CV implementing procedures did not provide sufficieni. requirements and interfaces to ensure that commercial grade items (CGIs) dedicated as basic components would be adequately verified to be capable of performing their safety functions as discussed in Section 3.4.2 of this report-(see Nonconformance 99900080/92-01-02 and Section 3.4 of this report).

The NRC inspectors also determined that CV Contract Management. personnel l

failed to prepare necessary procedures and instructions to meet the l

requirements of Section V of the BG&E P0. Paragraph 5.9 of Procedure No. 1-6.03 requires that the contract engineer prepare procedures and instructions that are required to support a customer P0 that are in addition to the requirements of the CV QA manual. BG&E P0 No. 746776X, processed in accordance with CV Shop Order No. 1196213, requires that if the s) rings are dedicated as basic components that a dedication program meeting tie requirements of EPRI NP-5652 and NRC GL 89-02 be used. The dedication of the springs being supplied to BG&E was performed to.the existing CV dedication-program, which does not address all the requirements of EPRI-5652 and NRC GL 89-02, such as the use of commercial grade surveys, source inspections, or acceptable supplier / item performance records to verify critical characteristics. Also, procedures were not: prepared to meet the dedication,

17

requirements of the BG&E PO'as required by Paragraph 5.9 of Procedure No.1-6.03-(see Nonconformance 99900080/92-01-03 and Section 3.4 of this report).

The CV shop order package for the spring identified the following activities to be performed to verify that the spring's critical characteristics were present-(1) verify parallelism, (2) verify part number, (3) verify _ alignment with ends, (4) visually inspect,- (5) verify wire diameter and length,-and _(6).

alloy identity.

The NRC inspectors reviewed CV-Design Engineering Drawing No. 5-185947, " Spring, Tandem Trim, Second Generation," Revision 0, with CV design engineering and discussed _a referenced note on the spring drawing i

indicating the spring rate-was equal-to 180 pounds / inch. CV informed the NRC inspectors that characteristics such as the spring constant,-solution treatment and a.ging listed on the spring drawing are controlled by the spring manufacturer. _It is noted that although CV performed an audit in 1989 of the.

spring manufacturer, Duer Spring and Manufacturing Co;, under their Navy Nuclear Audit Program and informed the NRC inspectors, they did evaluate the -

spring manuf acturer's control of the spring constant.

CV's QA manual does not permit the use of this type of an-audit to control a critical characteristics (see Nonconformance 99900080/92-01-02 and Section 3.4 of this report).

The NRC inspectors-reviewed Procedure No. 50-5.9.9, " Alloy Identity Procedure," Revision 16, dated October-- 15, 1988, and observed the alloy identification test performed on the spring. CV uses an Acromag Model 1101 thermoelectric metal tester to determine the generic metallurgical group (i.e., austenitic stainless steel, ferritic martensitic stainless steel, cobalt based alloys, etc.) of a part.

In order to properly identify the-spring material alloy, the Quality Control (QC)-supervisor needed to establish-a reference specimen. CV Drawing No S-185947 specified the spring be made of Allegheny Ludlum alloy A-286 or equal material per ASTM A638, Grade 630. CV had a test specimen _of MIL-S-24550A (UNS K66286) material readily available-with an-independent 1aboratory analysis which indicated that'the-chemical composition of the test specimen was very similar to that specified for ASTM -

A638, Grade 630, material. CV concluded that this specimen could be used as a reference standard for alloy identification of the spring.. Serial number "B-20" and the material type were engraved on this test specimen,1and handwritten on the copy _ of Supplement I to Procedure No; 50-5.9.9 used by the QC inspector performing the testing; The-supplement included a listing.of referenced standards, their alloy identification, serial numbers and scale reading. The alloy identification test of the spring indicated it was made-of the same generic alloy as the reference standard.

While the NRC inspectors concurred with-the technical evaluation performed to certify the suitability of the reference standard, the NRC inspectors determined that the QC supervisor's update of Supplement I to Procedure-No. 50-5.9.9.was not performed in accordaoce with CV-procedure requirements.

Section 3, of CV Procedure No.1-6.06, requires that the review and approval of procedures shall be denoted on the' revision page of the procedure, and that the distribution of revised procedures shall be performed by the issuing department Lusing CV Form -705158, " Document Submittal," or Form 705168, i

" Verification of Print Receipt," and requires that obsolete procedures. be t

collected and destroyed be the issuing department.

Procedure No. 50-5.9.9 was revised and placed in the shop order package and used for the dcdication of l i l

18

the spring without receiving all the requ! red reviews and. approvals and being.

distributed and ' controlled as required by Procedure th). : 1-6.06., It is~also

~

- noted that Paragraphs No. 7.2 land' 7.3 :of Revision No, '15 'to Procedure No.

50-5.9.9, which provided test,- analysis, and identification requirements for material. standards, were inadvertently omitted from Revision No.16 to

' Procedure No. 50-5.9.9 (see Nonconformance 99900080/92-01-03).

4 PERSONNEL CONTACTED

+

James R. Scarpelli, Director of Quality Assurance

+

Nard K. Himes, Supervisor of Quality Control

+

Michael T. Rathers, Manager of Marketing & Sales Nuclear' Valves

+

Paul M. Peoples, Manager of Marketing and Sales, Aftermarket

+

Tim Kunkle, Manager of Product Design & Development Robert L. fetterman, Chief Applications Engineer Lou Mayers, Quality Control Inspector Carrie Casler, Record Technician-Cheri Urda, Senior Sales Representative Thomas Dzikowski, Quality Control Inspector Greg Soltys, Industrial Engineer Specialist (Virginia Power)

Attended the entrance meeting on August 31, 1992

+ Attended the exit meeting on September 4, 1992 4 19

g >O M 4t/

j 8

f*%

i

+

.+

C UNITED STATES

.. /I'@J'i NUCLEAR REGULATORY COMMISSION y,,g w Asmaon. o c. ma o

g' *... ' p jut 2 31W Docket No.: 99900871/92-01 Mr. K. Kilpatrick, President Elgar Corporation 9250 Brcwn Deer Ibad San Diego, California 92121

Dear Mr. Kilpstrick:

SUIUlrr: IKirICE OF VIOIATIOt1 AND IKTTICE OF NotK0tHDWANCE (NRC INSPDCTION REFORT No. 99900871/92-01)

'1his letter adiresses the U. S. Nuclear Regulatory Cbmmission (NRC) inspection conductcd by K. R. Haidu, J. J. Petrosino, aM R. K. Frahm Jr. of this office on June 16 throtyh June 19, 1992, of your facility at San Diego, California, ard the discussions of our findirgs with you and other members of your staff at the conclusion of the inspection.

'Ihe inspection team evaluated defects reported to NRC staff by the Texas Utilities Electric Corporation (TUEC) in mrtain electrical inverters manufactured and supplied by your cmpurf to TUDC's Oranche Peak Steam Electric Station (CPSES). 'lhe team also evaluated the implementation of Elgar's corrective actions taken in response to conocrns expreved 'in our August 7,1990, letter regarding a previous NRC inspection at Elgar. "Ihe specific areas examincd durirq the inspection and our findings are dientM in the enclosed report. 'Ihe inspectors examined procedures and representative reconis, interviewed personnel, and made observations.

The inspection team found that certain of your activities appeared to violate imC requirements, as specified in the enclosed Notice of Violation (Notice).

'Ihe violations are of concern because Elgar's inadequate 10 CFR Part 21 procedures allowed Elgar to provide an NRC licensee with certain safety-related components that deviated fran the licensee's procurement documents.

Although Elgar knew of the deviation, Elgar failed to recognize the need to either evaluate the deviation or infom the licensee. You are required to respond to this letter and should follow the inst. ructions specified in the encloccd Notice when preparing your response.

'Ihe inspectors also found that the implerentation of your QA program failed to meet certain imC rcquirerents. 'Ihe specific findirqs and references to the pertinent requirements are listed in the enclocures. Please provide us within 30 days of this letter a written statement in amordance with the instructions specifiod in the enclosed Notim of Nonconforrance.

'Ihe responses rcquested by this lette are not subject to the clearance proccdures of the office of Kvagement and Budget as required by the Paperwork Reduction Act of 1980, FL 96-511.

)

1 20

K. Kilpatrick In accordarce with Section 2.790 of Title 10 of.tle Cbde of Federal Regulations (10 CFR 2.790) of the NRC's " Rules of Practice," a ocpy of this letter ard the enclosed inspection report sill be placed in the IRC Public Docurent Roca.

Sincerely, 7?

/ Iaif J.

brrhobu, Chief Vendor Inspection Branch Division of Reactor Inspection and Safeguards Office of Nuclear Reactor Regulation

Enclosures:

1.

Notice of Violation 2.

Notice of Nonconformance 3.

Inspection Report 99900871/92-01 cc:

Mr. T. Roth, QA Manager Elgar Corporation 9250 Brown Deer Road San Diego, Califamia 92121 i

i 21

4 DICIDSURE 1 IK7I' ICE OF VIOIATIOff Elgar Corporation Docket 16.2 99900371 San Dicgo, California Ikport 16. : 92-01 Durirg a U. S.11uclear Ikgulatory Ccmission (11RC) ircpoction corducted frcan June 16 throu3h June 19,1992, the inspection team identified the followirn i

violations of IIRC requirements. In accordance w3th the " General Statement of Policy and Proceduro for !!RC D1forcement Actions," Appctdix C to Part 2 of j

Title 10 of the Cbdo of Federal Regulations (10 CFR)- (1992), the violations j

aro listed belcw.

Section 21.21, "llotification of failure to ocatply or existence of a defect and its evaluaticri," of Title 10 of the Code of Fodcral Regulations (10 CFR 21.21), states, in part, that each irdividual, corporation, partrarship, or other entity subject to the regulations in this put must adopt appropriato proccdures to evaluato deviations and failures to ocmply, in all cases, witnin 60 days of discovery.

1.

Cbntrary to 10 CIR 21.21, Elgar failed to inoAporate the 60-day maximum tinn limit for evaluatiry deviations, ard otPer.new rcquirements that are specified in the current revision of 10 QE Part 21, of July 31, 1991, in Elgar Proceduro 10005-01, " Evaluation aid M ortirq of Defects ard 11onccxtpliance Pursuant to 10 CFR Part 2? <" of Au:pwt 30, 1991, which Elgar has adopted to implement 10 CIR Part E.

(Violaticai 92-01-01)

This is a Severity invol V violaticri (Supplement VII).

2.

Contrary to 10 CIR 21.21, Elgar failed to adopt ard 12tplc aIpropriate su uiures to ensure that deviations were ova,

.4 or passed on to customers if Elgar did not have the capability to perform the evaluation. Consequently, Elgar failed to evaluata a deviation rcgardire five Itcldcd caso circuit brnakers (MCCBs) that it supplied to Texas Utilities Electric Corporation (7UDC) with a certificate of conformnce (CoC) irdicatirg that Elgar had ensured that the h323s were new ard unused and that the MCCBs could be traced to the original manufacturer. Elgar furnished 7UEC with such a CoC even thcaxJh Elgar had received a letter frm the MCCB supplier informing Elgar that it=

does not retain records to certify the origin of specific %crants, except when a special order was placed. Ilowever, Elgar did not place any special requirements in its purchase order with the supplier requirire the nocded traceability. -(Violation 92-01-02)

'Ihis is a Severity Invol IV violation (Supplement VII)..

2 22

Purtannt to the proviolons of 10 CMt 2.201,1:lgar Corporation is hereby required to subnit a written statencnt or explanation to the U. S.

11aclear IvrJulatory Ctannission, ATIN: Document Control Deck, Warhington, D. C. 20555, with a copy to Mr. Icif J. llorrholm, Chief, Vcnior Ingxction 11rart:h, Division of Imactor Inspection and Safeguania, Offloo of fluclear Practor Irgulation, within 30 days of the dato of the letter trancmittinJ thin 110tico of Violation (11otico). 'Ihis reply thould be clearly markod as a "Peply to a flotico of Violation" ard should incitdo, for Violation 92-01-02:

(1) the reason for the violation, or, if contestau, the basis for dicputinj the violation; (2) the correctivo stopc. that have boca taken arxl the results achieved; (3) the corrective steps that will be taken to avoid further violations; and (4) the date when full ocamliarxxi will be achiavod. We will consider extenliny the regonno timo if you can rAcu good cause for us to do so.

Dated at Rockvil o, luryland j

thio 73 day of 1

1992. i 23 l

DJCIDSURE 2 IDI' ICE OF IJOf100fiFOIMAIJCE Elgar Corporation Docket llo.: 99900871 San Diego, California Report lio.t 92-01 Durity a U. S. lluclear Regulatorf Omnission (!mC) inspection conducted frca June 16 through June 19, 1992, the inspection team determined that certain activities wre not conducted in accortlance with NRC requirements that were contractually irqposed by purt:hase artiers frm nuclear power plant licensees.

The imC has classified those items as renconformnces to the requirements of Title 10 of the Code of Federal Rmulat1QDs, Part 50 (10 GR Part 50),

Arpendix B, " Quality Assuranco Criteria for liuclear Power Plants ard niel Reprocessirg Plants."

1.

Criterion I, "OrtJanization," of Apporrilx B to 10 CFR Part 50, states, in part, that the percons ard ortJanizations performirg quality arsurance (QA) functions shall have sufficient authority and ortJanizational freedm to find quality proble:ts; to initiato, recarnend, or provide solutions; and to verify implementation of solutions. Such persons and orgvtizations performing QA functions shall report to a mnagement level haviry this required authority ard ortJanizatiom1 freedm, includlig sufficient irdepeMence from cast and schodulo shen opposed to safety considerations. 7he QA prograra my have other characteristics if persons and ortJanizations assigned the QA furctions have this rcquired authority and ortJanizatiom1 frmdm.

Contrary to the above, Elgar failed to ensure that its QA Manager had adoquate irdependence frca cost and schedule when opposed to safety considerations.

In addition to performiry his regular QA responsibilities, the Elgar QA Manager was also desi';mtad as and performed the functions of Elgar's (12stcmcr Service Kimger and its Field Servico Kanager.

(flonoonformnoe 92-01-03) 2.

Criterion IV, "Prccurement th'ent Control," of Appendix B to 10 CFR Part 50,' states, in part, that reasures shall be established to ensure that applicable regulatory requirements, design bases, ard other requirements necessar'f to ensure adequate quality are irv;1uded or referenced in the documents for procuring material.

Contrary to the above, Elgar failed to ensure that it specified Texas Utilities Electric Corporation's ('IUDC's) requirement, to supply new and unused molded case circuit breakers (MCCBs) that wre tracoable to the origiml equipmnt mnufacturer, in its pirchase order (FO) to the supplier.

(lionconformnce 92-01-04) 3.

Critorion VII, :"Cbntrol-of Purchased Material, RIuiptent, c!.2 Services,"

of AppeMix B to 10 GR Part 50, states, in part, that measures shall be established to ensure that purchased material, cquipmnt, and services, 24

khother purdiased directly frm the mnufacturer or purdnsod through contractors ard subcontractors, conform to the procurement dments.

Dccumentary evidence that the mterial erd cquipnent confom to the procurement requirunents shall be available and be sufficient for determinirq the specific rcquirements inot by the purchased storial ard equipment.

Contrary to the above, Elgar certified to WDC that tho safety-related MOCBs conformed to all IV requirements for two fos even though Elgar did not have documented evidence to irdicate that the cirulit breakers were new ard supplied frun the origimi circuit breaker mnufacturer.

(11onconformnoo 92-01-05) 4.

Criterion III, " Design Control," of Apperdix B to 10 CFR Part 50, states, in part, that measures shall be astablishod for selecting materials, parts, equipnent, and processes that are essential to the safety-related functions of the structures, system, and ccmponents ard for reviewirg each of these itcens for suitability of application.

Contrary to the above, Elgar dodicated ard supplied rcplacement MCCBa to WDC as safety-relatcd without ensurirg that the MOCBs would moet Elgar's primary design function (critical duracteristic) for the MCCD's electrical inverter application. The MOCB's primary design function for Elgar's electrical inverter application is to continuously carry the inccnirn electrical voltage ard rated current without interruption.

(Nonconformanoo 92-01-06)

Please provide a written statement or explamtion to the U.S. Nuclear Rrgulatory Ocumission, ATIN: W e nt Control Desk, Washington, D.C. 20555 with a copy to Mr. Leif J. Norrholm, clief, Verdor Incpoetion Branch, Division of Reactor Inspection aid Safeguards, Offico of Nuclear Reactor Regulation, within 30 days of the data of the letter transmittirq thin Notice of Noncon-formance. This reply should be clearly mrked as a " Reply to a Notico of Nonconformance," and should includo for cadt nonconformnce: _(1) a descrip-tion of steps that have been or will be taken to ca u st these items, (2) a description of steps that have been or will be taken to prevent recurrence, ard (3) the dates your corrective actions ard preventative measures here or will be completcd.

D'ited at Rockvil e Marylard this J3 day of 1992.

4 M

e 25 a

DiC10GURE 3 U. S. IRCEAR RIDUIA10tY RIMIP4101 Of7 ICE OP 1RCEAR RI7CIOR RTEUIATIQi DIVISIQi OF RI%CIUt IllSPECTIQi NO SATTEUAIOS OIGNTIZATIQi:

Elgar Corporation San Diogo, California RIIORT 10.!

99900871/92-01 OTCNIIZATIQlAL Mr. Ray:nord D. Daniel, Director coffrAct:

Quality ard Rollability (619) 458-0206 ODRRIEIQDDICE 9250 Brown Door Road ADDRESS:

San Diogo, California 92121 NUCLI7J1 DIDUSIRY Mmufactures aM cupplies safety-rtlated electrical ACTIVITt:

inverters, uninterniptibio }xuer supplies (UPS), and replacement ocuponents.

INSPIrrICt1 Juno 16-19, 1992 C0tIDUCITD:

LEAD INSiu;ws:

t7 2*/12 K. R. Haidu, 7bam Icador Ibte Special Projects Sectial (SPS)

Vendor Inspection Branch (VIB)

Orl1[D1 DISPIPIORS:

Joseph J. Petrosino, VIB RcmId K. Frahm Jr., VID

~/Mv!f4 APP 10VED BY:

G. C. CValins, 011cf, SPS, VIB

'Date Division of Reactor Ins}xction ard Safeguards office of Nuc1 car Reactor Rayulation INSPD'f1Gi BASES:

10 CFR Part 21 aM 10 CFR Part 50, AppeJdiX B INSPIrrIQi SCOPE:

'Iho NRC inspection team evaluated defects reported by the 7bxas Utilities Elex:tric Corporation in electrical inverters manufactured by Elgar ard supplied to the Camncho Peak Steam Electric Station Unit 2 and reviewed the status of Elgar's corrective action on previous inspection firdings.

PIRTP SITE Numerous APPLICABILJTY: 26

i 1 INSPIrrI0tl StH RRY 1.1 Violations 1.1.1 Otxitrary to Section 21.21 of Title 10 of the Oodo of Foderal Regulations (10 CFR Part 21), Elgar failed to incorporato the required tine limits and other ncu requirements specified in the current ravision of 10 CFR Part 21, of July 31, 1991, in its Proceduro 10005-01. Elgar had not revised Procoduro 10005-01, " Evaluation ard Reportirn of Defects and Honocxtpliarre Purcuant to 10 CIR Part 21," of August 30, 1991, that it had adcpted to inplement its 10 CFR Part 21 program.

(Violation 92-01-01) 1.1.2 Contrary to 10 CFR 21.21, Elgar failed to adopt ard inplement appropriato procedurcs to ensure that it evaluated deviations or informod customers of them if Elgar did not have the capability to perfom the evaluation. Consequently, Elgar failed to evaluata a deviation regardirn f1vo nolded caso circuit breakers (HOCBs) that it st4 plied to Texas Utilities Electric Corporation (WDC) with a certificato of conformance (CoC) indicating that they were nr.v and unused MCCBs and that Elgar could verify the traceability of the MCCDs to the original equipnent manufacturer. Elgar issued the Coc even though its staff had previously rocsived a letter frun the supplier of the IKX2s in which the supplier stated that it could not certify the origin of the MCCDs.

(Violation 92-01-02) 1.2 lionconformrres 1.2.1 Contrary to Critorion I, "OrtJanization," of Appardix B to Title 10, Codo of roderal Regulations, Part 50 (10 CFR Part 50), the Elgar M }bnager was also designated as Elgar's Custmer Service Manager ard its Field Servico Ibnager ard perfomcd the functions of both these positions.

(Nonconfomanco 92-01-03) 1.2.2 Contrary to Criterion IV of Apperdix B to 10 CFR Part 50, Elgar failed to specify the requirement to supply only new circuit breakers traordalo to the original equi;nent mnufacturer in its purthaco order (PO) to the supplier. (Nonconformanoo 92-01-04) 1.2.3 Contrary to Criterion VII of Apperdix B to 10 CFR Part 50, Elgar certified that MOCBs confomcd to all purchase order requirements without proper bases and docunentation to substantiato these claims.

(Nonconformance 92-01-05) 1.2.4 Contrary to Criterion III of AppeJdix B to 10 CFR Part 50, Elgar failed to tako adequato measures to verify that MCCBs would perform their safety-related function durity the dodication process.

(Nonconfomanco 92-01-06) 1.3 Unresolved Itemq Elgar nust perform an evaluation to detemino if a defect exista regarding the deviation frcxn the two 'IUDC IOs, or inform WDC within 5 work days if Elgar detemines it does not have the capability to perform the evaltation.

(Unresolycd Item 92-01-07) !

27

2 !II'NIUS OF IRIN100S IllSPirr1Cil PIllD11CS 2.1 ylglntion 99900871/90-01-01 (CloggQ. Elgar ind failed to adopt prcoxlurca to implcent 10 QR Part 21.

Durity thin inspoction, the imC inopoctors revicutd the correctivo actions taken by Elgar to crsure tint it ind adopted proocxtures to adoquately implement 10 CFR Part 21, which inpitments Sectica 206 of the EncIvy Roon;anization Act of 1974. Thetem revicscd Elgar's corrective actions taken to addrean thin violation and ovaluated the adcquacy of Elgar's procedaro rcgardirn the ncv revisions to 10 CFR part 21 inauod on July 31, 1991.

'Iho team revicuri Elgar Proceduro 10005-01, "Evaltntion ard Reportirg of Defocts and 11onacoplirincns Purctant to 10 OH Part 21," August 30, 1991. The team fourd tint even though Elgar revised its 10 CFR Part 21 procoduro, it failtd to incorporato the July 31, 1991, revision of 10 CFR Part 21 into Procoduro 10005-01. Conocquently, the team clocal this violation, lut established Violation 92-01-01 ard will folicM tho correctivo action an prt of this ryM violation.

2.2 Violation.39900871/90-01-02 (Onen). Elgar ind failtd to adogtntoly evaltato deviations listed on Elgar ergirmrity chango retices (IX:lla) for its safety-related electrical inverters ard uninterruptibio power supplies (UPs).

Elgar nico failcd to inform its custamorn of theco deviations. Elgar had fourd numerous deviations in responno to the 10C staff's inspections at the Elgar f acility in 1988 and 1990, iut failtd either to evaluate or to infonn its custcurn of timo doviations so that its custcmcrs could ovaltute them in acconlanco with 10 CIR Part 21.

Durity thin 1992 inspoction, the 10C incpoctora r(questcd a coloctext sanple of.the notification lottern Elgar had tranantitted to its cuatcarra, int Elgar could not prtdoco the rMwntation in a timely manner. ' Ibis violation will remain open pendirn review durity a futuro irepoction.

2.3 HQDQgn[grunce 99900871/90-91-03 -(ClgngR. - Contrary to Criterion II of Amerrlix B to 10 ("E Part 50, five of the six enployees who typically performed activitica affectity the cplity of nuclear safety-related systems arti exponenta lad not bocn irdcctrirnted in the Elgar quality assuranco program. 'Iba itepoction team examincd the correctivo action taken by Elgar ard datomityx1 it to be adequato.

2.4 HgDQQDf9EM!1go 99900871/90-01-04_ _ (Ooon).. Contrary to Criterion III of Appendix B to 10 CIR Part 50, the wirity termination locations were not -

corroctly translated frm the olcetrical schomtic for a r.afety-relatcd UPS supplied to 1Ulr's Omarche Peak Steam Electric Station (CPSES). h team did not review thin mtter in sufficient detail to determino the adcrpey of Elgar's corttetivo action. 'Ihis renconformnoo will runnin open.

2.5 Ugnggr1[pnMIYa193pp871/90-01-05 = (OTal. Elgar failcd to take adagtnto correctivo action to address nonconformances in its design review process.

'Iho team did not revitM Elgar's corrective actions to deterntino their adcquacy. 'Ihis renconformra will remin open. 28 r-e.

e

,m, e, v m.

-w mm-w-r--+

--e--

~

em.

2.6 floroonformvy3e 999_00871/90-01-06 (Cice.od). Elgar failed to establish namires to encure that its OA program corrects otrapanent-relatal deficierries fourd 17f Elgar's field service reprecentatives. %e t eam reviewed this mtter ard fcurd it to be adequately acktrensed.

2.7 lignoonformroe 999008.71/90-01-07 (Closed). Elgar neither establishod an inturrul atx11t schedule nor rerformd any atdits of hs established QA program frcan June 1988 until March 1990. We incpoction tavra reviewed the corrective action taken lyf Elgar and determinod it to be satisfactory.

2.8 Rven Item 99900871/90-01-08 /Clo w &.

Elgar limital its design change reviews to only printcd circuit boards (ICBs) ard had not considered it necessary to review design changes to other ocmponents, such as transforcers, chokes, host sinks, ard silicon-controlled rectifiers (SCRs). %c team reviewed a sanple of II21s ard cancitt'ai that Elgar is adcquately reviewing all design charges.

2.9 Cben Item 99900871/90-01-09 (Open). We I E found a nonconformance regardirg cracked solder joints in PCBs supplied 17/ Elgar to 'IUDC. W e team did not review this matter in sufficient detail to determine if Elgar had i

satisfactorily evaluated tie issue.

3 INSPDC.TIOtl FDOINGS MO CTIMIR COEDTIS 3.1 Entrance and Exit Meetirrrs Durirg the liRC cntrance unetiin on June 16, 1992, the inspectors outlined the scope of the irnpoction to the Elgar staff. During the NRC exit meeting performed on June 19, 1992, the team loader surnarized the team's findirgs to the Elgar mnagenent ard staff.

3.2 Review of 10 CFR Part 21 Prorrram We imC team reviewcd Elgar's 30 CFR Part 21 program to deterinine if Elgar was cmplyiry with the requirenents of 10 CFR Part 21. W e team reviewed Elgar's Procedure 10005-01, " Evaluation and Reportirn of Defects ard Nonaccpliance Pursuant to 10 CFR Part 21," August 30, 1991. Elgar established Procedure 10005-01 to implement the provisions of 10 CFR Part 21.

We 10C inspectors reviewed Elgar's program ard procedure for implementirq 10 CFR Part 21.

We team fourd that Elgar had not revised Procedure 10005-01 to incorporate the July 31, 1991, revision to 10 CFR Part 21. Eis revision included provisions for time limit requiruments for evaluatirg deviations and informirg 100 licensees. In this revision,- Elgar also changed requirments for the capability of suppliers to perform evaluations ard changed the requirements for records. We team found that Elgar had failed to irwr>te any of the 10 CFR Part 21 charges into Procedure 10005-01.

(Violation 92-01-01)

We team also fourd that Elgar had fallod to evaluate a deviation to a safety-related procurement document rcgardirg MC.CBs that Elgar supplied to 'IUDC's 29

CPSES. TUlr required Elgar to ensure traceability to the origimi equipnent mnufacturer (ODi) of the HTDs. However, Elgar certified that the MTDs met TUDC's 10 requirments even tJoxjh the MCCB supplier had previously informed Elgar that the MOCBs were tot traceable to the 004. We team concludod that Elgar Procedure 10005-01 was not adequate to ensure that deviations to procurment documents were evaluated or that custwors were notified.

(Violation 92-01-02) 3.3 Irdereidence of OA thrnggr We team obccrvcd that the Elgar @ Manager appeared to be periodically performity tasks that were relatxd to custacrs' inoaming rcquests for infonction or acr.ponents; conscquently, the team asked the m }bmger if he was performirg tanks other than thoce that would to relattd to the quality assuranm departrent. We @ Manager explained to the team that Elgar lacked two r.aragers and that he was fillirg these two positicns in addition to performity his gular duties as m Itunger. 20 m Manager was actirg as Elgar's Field 5 vice Manager (FGi), ard its Customer Service Mmager (CSM).

We inspectors determined that the m }bnager was responsible for coordinatirq all of Elgar's field service requests ard queries for its safety-related and ocutcrcial inverters ard UPS units. The inspectors also detennined that the field service department consisted of tha QA Manager, who acted as the Fai, ard a fcM other Elgar enployees kho performcd the actual field service work for cumercial field service requests. Elgar used a few specializcd consultants for all of its ime licensee field service work, ard this was also coonlimttd by the OA Manager. Werefore, the QA Manager appeared to be performirg nest of Elgar's administrative ard FQi duties.

'the imC inspectors found the number of Elgar enployees in the custmer service onJanization was similar to that for the field service onJanization. For exanple, the custaner service department consisted of a custacr Service Administrator ard the CSM. Se Administrator reported to the CSM; however, the QA tunager, actiry as Cui, made most of the decisions and responses to the custmer's questions and cer: plaints.

Criterion I, "onJanization," of Arpendix B " Quality Assurance criteria for fluclear power Plants and Fuel Reprocessirg Plants," to 10 CFR Part 50 requires, in part, that the persons and organizations performirg quality assurance (QA) functions shall have sufficient authority and organizational freedcn to fird quality problems, incitdlig independence frun cost ard schedule when oppoced to safety considerations.

%e team concluded that the @ Manager could rot easily maintain his onJanizational irdeperdence or freedczn from cost ard schedule when opposed to safety-related considerations.

(Nonconformnoe 92-01-03) 3.4 Ctranche Peak Steam Electric Station (CPSES) Unit 2 Issues On May 13, 1992, personnel frun the %xas Utilities Electric Corporation (7UDC), the licensee for the CPSES, infomed the IEC that they observed several defects in 103-1-132 type uninterruptible power supplies (UPS) 30

supplied by Elgar. 1ho insgetico team revicM correspondence exclmrged between 7UIr and Elgar, interviewed various Elgar personnel and determined the follcuirg:

'IUlr iscued 10 CP-0445 to Elgar to furnish four each of the 25WA itdel 253-1-108 ard 10WA Model 103-1-132 UPSs to moet Gibts aryl Hill, Incorporated, Specification 2323-IS-9, Revision 1 of October 28, 1977.

'IUDC roccived all the units and stored them in its warehouse.

hhen operations perrarel prepared to start up CPSIS Unit 2, they encountercd problems in startJrg the UPSs ard roted several conponents missirg.

In a letter of llovember 25,1991, 'IUDC issued a contract to Elgar to provido only technical assistance to the service contractor in trouble shootirg the UNs. Elgar otocrved theco activities but did rot particlp te in them.

On February 11, 1992, after determiniry that ta:hnical assistance did not provido the required results, 'IUDC issued a FO to Elgar for onsite services includirn electrical, QA ard quality control services. Elgar personnel otrrpletcd servicing the UPSs ard made them operational by April 24, 1992.

'IUIr reported sovan defects in Elgar inverters in Unit 2.

'Ibe followiry paragraphs contain the results of the review of each reported defect.

Failures in inverter control circuit identified as "J7".

1UDC comtissioned the UPSs after a prolorged storage period durirg which they were doenergized. In such a ecse, high in-rush currents chargirg the capcitors in the pulso width nodulatirq (ITM) board identified as "J7" caused repeated fuse failures. Elgar corducted a failure analysis on the ITN ard determined that the tolerances in capacitors C101, ard C102 ard integratcd circuit Z103 camponents were inadoquate. She inadequate tolerances permitted pulse widths outside the original design intent and did rot allow sufficient time for the capacitor prochargo.

'Ihis could have causcd the inverter fuse to fail. Elgar issued engineerirg change notico (Dai) 8162 to chargo the tolerances of the throo cnuponents. Elgar also notificd other custavJs and roccnmendcd the replacement of all "J7" boards delivered before }hrch 1992.

Missirn upgrade on circuit board.

Elgar had made several revisions to the circuits frce the time it had surplied UPSs to CPSES. Elgar service personnel inspected the UPSs,and replaced all the missirg parts ard obsolete cards, to brirs each UPS to conformarce with the latest drawings. 31

- _ =.

Incorrectly wourd transformer Elgar noted that the coils in the transfonter on the gato drive cituilt leard had been incorrectly wound. The drawiry requires the coils to be wound concentrically with nylon wrap insulation, but the ca mon practico of the !bgnetics Department was to wird multiple secondaries bifilar if the turn ratio and inductarce were the same. This practice my inluce noiso in the output from the cocondary.

On Hardi 19, 1992, Elgar iscuod an intenal " Quality Alert" reminding the production staff to build the UPS in acconiance with the print, and to wind all multiple scoondary win 11ngs separately. 1he " Quality Alert" also advised inspectors to perfom a capacitanco chock to verify all transformr secondary coils are wcurd concentrically.

Missirn parts on PCB "J2" e

Durirg a service trip to CPSIS, Elgar service personnel obactved that the IG "J2" was missing a resistor identified as "R144".

Elgar engineering personnel determined that IIII 1899 had not been implementtd to chargo the IG "J2" drawirn to add R144. Elgar revised the drawing ard added R144 to the IG.

Elgar notified other affected customers, Poor crinp connections, incorrectly cupplied parts, ard wiring errors e

hhile scIvicity the UPSs, Elgar service personnel oboctYod poor crimp connections, incorrect parts, ard wirirn errors that differed frun Elgar's stardattis connections. Elgar service personnel infonned the NRC inspectors that they had to remvo inprcperly crimpcd connections mde to conductors in the inverters ard properly crinp them with appropriato crinpiry tools to meet the Elgar workmanship stardards. Por instance, a Molex pin has to be cripped with a Molex c.rinpirg tool. 1his tool is not readily available at construction sites. Elgar service pcInonnel corrected all the workmanship errors.

1ho inspection team concitdod that Elgar had adequately evaluated defects putulant to 10 CPR Part 21 ard had advised affected custczas where appropriate to reviso or upgrade their Elgar equipment. The inspectors obcerved Elgar craftsmen followiry the establirted proccdures using appropriate tools to perfonn various activitics.

3.5 hise Failure in Ptuer Line Oculitioner at Nine Mile Point In January 1990, personnel from the Niagara Mohawk Iber Corporation, the licencoe for the Nine Milo Point Nuclear Power Station Units 1 and 2, obocrved a blown fuso in the 103-1-176 type power line conditioner (PLC) mnufactured by Elgar.1hc PLC roccives 575 Vac, converts it to 120 Vac with a 3-percent tolerarm, ard supplies this voltage to the static switd) of a 103-1-176 type.

Elgar UPS as an alternate to the normal (inverter) output voltage. 1he Plc output voltage is used only when the inverter output voltage of the UPS exoecds the permissible tolerance.

~7-32

~

In this configuration, a momentary loss of input pcwcr (80 millisecords) to the PIE caused the loss of power to the SCR control logic b2 fore interrupting the power to the SCR gate drives. Such a situation caused all the SCRs to be

" gated-on" by the 1cgic default sigrals.

7b oorrect this problem, Elgar llaplemented an ergineering change to connect the pcwer supplies by diode to both the Sca control logic and the SCR gate drive so that both would remain switched on or would remain switchcd off at the same levels. Elgar added a monitor to irdicate the fuso condition.

3.6 Procurement and Dedication of Cumercial Grad 9 Items In its activity with the nuclear irdustry, Elgar primarily provides spare aM replacement parts for the UPSs ard PIru it had manufactured. Elgar procures all these cmponents as ocxanercial grade with no tedinical or quality requirements imposed on the Ios to its suppliers. When Elgar is required by PO to supply the utilities with safety-related ccanpanents, it performs a dedication prccess in accordance with its applicable acceptance test procedure (ATP).

The inspection team reviewcd several data packages for various conponents to evaluate the adoquacy of Elgar's prcx:urement and dedication programs. The inspectors did not fird any concerns after reviewing data packages for capacitors aM SCRs, but found several irronsisten::les with two packages for MCCBs as described below.

The NRC inspectors reviewed two data packages for MCGs recently supplied to 7UDC. The first package was PO S 0014347 7S2 (which corresponds to Elgar sales order (SO) 53814), Revision 2, October 3,1991 which documented the order for three Elge model 852-U53-TE circuit breakers, 480 Vac, 50 anp, 3 pole, with molded case ard auxiliary switch. The second was PO S 0014343 6S2 (SO f^316), Revision 1, October 23,1991, - for two adiitional Elgar model 852-U53-TE circuit breakers. The original POs were dated July 1 and July 3,1991, respectively. 7UDC stated on both IOs that "All breakers supplied shall be new not repaired, reconditioned, or used aM are to be supplied frcxu the l

circuit breaker manufacturer" and "the supplier shall caplete ard sign a certification dment which irdicates that the requirements of this purchase order have been met."

l:

Elgar's QA Manager reviewed the PO ard ocr?_ract requirements and documented.

)

his review on Form 1001040-36 of July 15, 1991. He noted in the " Additional Requirements /Octuments" section to " order frtra factory direct - need C of C stating parts are new & not refurbished." However, Elgar issued PO 156607 to GE Supply Company (GESCO) of San Diego on August 7,1991, for five commercial grade circuit breakers (Elgar mcdel 852-U53-TE, GE model niunber TED134050) ard failed to specify the requirement that the breakers be new ard be from the original circuit breaker ranufacturer.

(Nonconfornance 92-01-04) u l

Elgar received the breakers ard subsequently dedicated them in accordance with ATP T52-XXX-XX, Revision A, on September 9, 1991. Elgar prepared two mrtificates of confomance (CoC), one for each IO, on September 11 ard 25, 1991, to certify to 7UEr that the breakers not all the applicable PO ' 1 1

33 l

requirements. Elgar stated on cadi CbC that "the mterial arri cmponents furniched herewith are in confonnance with the applicable requirements, specifications, ard drawirgs listed in the purchase onler." funce, Elgar's certification implied that the MCCBs were new and were frcan tne origimi mnufacturer without proper bases to substantiate these claiIns.

(Nonconformnce 92-01-05)

Elgar also received a letter frm GESCD of March 7,1989, which documents a telephone conversation concerning GESOD's in@ility to trace the origin of the MCEBs supplied to Elgar. The letter states, in part, that "we do not retain records to certify the origin of specific products, execpt khere a special onder was placed so that a documntary chain of sale was created." Even though Elgar knew frm this letter that GESCO could not trace its existing stock back to the original cquipnent mnufacturer, Elgar did not evaluate the deviation or forvani the infonration to the custmer. The team concluded that Elgar failed to adopt ard imple.nent appropriate procedures to ensure that deviations were evaluated or were forwarded to customers if Elgar did not have the capability to perform the evaluation.

(Violation'92-01-02 was cited in this area in Section 3.2)

The NRC inspection team notified Elgar that on June 18, 1992, the team discovered a deviation from the IUDC Fos as detailed above. Therefore, pursuant to 10 CFR Part 21, Elgar must initiate an evaluation to detennine if a defect exists, or inform the affected licensees within 5 work days if Elgar determines it does not have the capability to perform the evaluation.

(Unresolved Item 92-01-07)

The NRC inspectors also reviewed the ATP used by Elgar to dodicate the ccumercial grade circuit breakers as safety-related. On the ATP Elgar verified the diJnensic.ns and markings, insulation resistance, termimi disconnect and continuity, and auxiliary contact. However, Elgar did not verify on the ATP that the breakers could continuously carry the inoming electrical voltage and rated current without interruption for a specified period of time. The primry function and most critical diaracteristic of the -

MCCBs is to provide an uninterruptible supply of electric power. Elgar agreed with this =mmnt and informed the NRC inspectors that it was reviewing each of its ATPs to evaluate its adequacy in ensuring the capability of'the basic cxanponent to perfonn its safety-related functions.

(Nonconformance 92-01-06) 1he team expressed a concern that Elgar's dedication m for circuit breakers does not adequately address a verification that the breakers are traceable to the original equipment manufacturer, arri does not call'for performing a detailed visual examimtion of the breakers to check for indication of fraud. Elgar does rot perform this examination at receipt inspection or as part of the ATP. Hcwever, the examination is rw-ary to minimize the possibility of substandard circuit breakers being used in safety-related applications in nuclear power plants. In NRC Bulletin 88-10, "Nonconfonning Molded-Case Circuit Breakers," and its supplement, the staff gave guidance on detectiry fraudulent breakers. 34

= _ = _.

4 PERSCtMEL 00!TIACIID Elciar Cuweration

+

K R. Kilpatrick, President

+*

R. Daniel, Director of Quality Assurance

+*

R. Garrett, Vice President, Operations

+*

T. Roth, Quality Assurance Manager

+*

J. 'Iuritto, Quality Assurance Specialist Ultimate Powr Solutions

+*

R. Parrish, Engineering Manager l

l Attended Entrance Meeting on June 16, 1992

+ Atterded Dcit Mocting on June 19, 1992.

l.

l l

t t,

35 Lt'

panow

'g UNITED STATES n

3

'J i

NUCLEAR REGULATORY COMMISSION

  • d$ff n

wAsmuotow, o c. m,w s'..~.cf

.+

July 2. 1992 Docket No. 50-302 Mr. Percy M. Board, Jr.

Senior Vice President, Nuclear Operations florida Power Corporation ATTN: Manager, Nuclear Operations Licensing Post Office Box 219-NA-21 Crystal River, florida 32629

Dear Mr. Beard:

SUDJECT:

INSPECTION Of THE PROCUREMENT AND COMMERCIAL GRADE DEDICATION PROGRAMS AT THE CRYSTAL RIVER NUCLEAR PLANT, UNIT 3 (REPORT NO. 50-302/92-201)

This letter transmits the report of the inspection conducted March 23-27, 1992, at the Crystal River Nuclear Plant, Unit 3, by R. P. McIntyre, S. D. Alexander, J. J. Petrosino, and W. C. Gleaves of the U.S. Nuclear Regulatory Commission's Vendor Inspection Branch and by F. Jape and H. Themas

)

of NRC Region II. The inspection was related to activities at the plant site authorized by NRC license DPR-72. At the conclusion of the inspection, we-discussed our findings with you and the members of your staff identified in Section 5 of the enclosed inspection report.

The inspection was conducted to review the implementation of the florida Power Corporation (FPC) programs for the procurement and dedication of commercial grade items (CGis) used in safety-related applications at Crystal River.and also to review the corrective actions for certain installed CGis that were found to be of unverified quality in the 1989 NRC procurement inspection report (Report 50-302/89-200) and in the subsequent Notice of Violation. The results of the inspection indicate that FPC failed to properly dedicate certain CG!s procured for use in safety-related applications. Consequently, some CGis of indeterminate quality were installed or available for

-installation in safety-related plant systems. The specific deficiencies contributing to this condition included failure of Nuclear Procurement Engineering to adequately identify safety functions of the parent component and/or the piece part within the parent component; failure to identify or' consider failure modes and the effect they could have on the component and the surrounding area; and failure to identify characteristics for which assurance would need to be obtained, in addition, the Procurement and Material Quality Assurance staf? failed to verify or obtain other assurance for characteristics that were identified on your functional analysis / critical characteristics review (FACCR) form. Finally, engineering involvement to confirm the technical adequacy of the assurance activities carried out by the procurement organization was weak.

36

Mr. Percy H. Beard, Jr.,

for the seven unsuitable CGIs listed in the December 1,1989, Notice of Violation and Proposed Imposition of Civil Penalty (which was later withdrawn), the team determined that FPC had written justifications for continued operation for the CGis in the seven examples, but had not completed J

the corrective actions la all cases. This untimely disposition of corrective actiens for certain CGIs, over nearly 3 years, is of concern.

.l lhe inspection findings presented to your representatives during the exit meeting on March 27, 1992, and discussed in this letter and the enclosed report, are considered deficiencies in your commercial grade procurement and dedication activities and will be referred to the NRC Region 11 office for any appropriate enforcement action. However, to address the items of indeterminate quality found during this inspection, you are requested to make an assessment of the safety implications that these deficiencies could have and take appropriate corrective actions based on your review of the information contained in this report.

The team considered as a procurement program strength TPC's policy of shifting to the purchase of Appendix B items when available, thus reducing the number of items which need to be purchased commercial grade and dedicated for use in safety-related applications.

The team also considered as a program strength the upgrading of the Crystal River receipt inspection testing capabilities to perform more extensive commercial grade dedication-testing and acceptance activities.

In accordance with 10 CFR 2.790(a), a copy of this letter and the enclosure will be placed in the NRC Public Document Room.

Should you have any questions concerning this inspection,. we will be pleased t

to discuss them with you. Thank you for your cooperation in this inspection.

Sincerely, n

arg et Division of Reactor-PPdjects I/II Office of Nuclear Reactor Regulation

Enclosure:

Inspection Report 50-302/92-201 cc w/ enclosure:

See next page 37 Mr. Percy H. Beard Crystal River Nuclear Plant Florida Power Corporation Unit 3 CC:

Mr. A. H. Stephens Mr. Robert G. Have, Director General Counsel Emergency Management Florida Power Corporation Department of Commuisity Affairs HAC-ASD 2740 Centerview Drive P. O. Box 14042 Tallahassee, Florida 32399-2100 St. Petersburg, Florida 33733 Hr. P. f. McKee, Director Chairman Nuclear Plant Operations Board of County Commissioners florida Power Corporation Citrus County i

P. O. Jox 219-NA-2C 110 North Apopka Avenue Crystal River, Florida 32629 inverness, Florida 32650 Mr. Robert B. Borsum Mr. Rolf C. Widell, Director B&W Nuclear Technologies Nuclear Operations Site Support 1700 Rockville Pike, Suite 525 Florida Power Corporation Rockville, Maryland 20852-P. O. Box 219-NA-21 Crystal River, Florida 32629 Regional Administrator, Region II U.S. Nuclear Regulatory Commission Mr. Jacob Daniel Nash 101 Marietta Street, N.W., Suite 2900 Of fice of Radiation Control Atlanta, Georgia 30323 Department of Health and Rehabilitative Services Administrator 1317 Winewood Boulevard Department of Environmental Regulation Tallahassee, Florida 32399-0700 Power Plant Siting Section State of Florida Attorney General 2600 Blair Stone Road Department of Legal Affairs Tallahassee, Florida 32301 The Capitol Tallahassee, Florida 32304 Senior Resident inspector Crystal River Unit 3 Mr. Gary Boldt U. S. Nuclear Regulatory Commission Vice President-Nuclear Production 6745 N. Tallahassee Road Florida Power Corporation Crystal River, Florida 32629 P.O. Box 219-5A-2C Crystal River, florida 32629 38

1 1

U.S.. NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION DIVISION OF REACTOR INSPECTION AND SAFEGUARDS 3

Report No.:

50-302/92-201 Docket No.:

50-302 License No.:

DPR-72 Licensee:

Florida Power Corporation P. O. Box 219-NA-21 Crystal River, Florida-32629 Facility Name:

Crystal River Nuclear Plant, Unit 3 Inspection at:

Crystal River, Florida Inspection Conducted:

March 23 through 27, 1992 i

Prepared by:

Richard P. McIntyre, Team Leader Date

~

Vendor Inspection Branch (VIB)

Inspection Team:

S.D. Alexander, EQ and Test Engineer, VIB J.J. Petrosino, Reactor Engineer, VIB W.C. Gleaves,- Reactor Engineer, VIB t

M. Thomas, Rea, r Engineer, Region II F.91 ape,,y tio Chie,, Region II:

l-I ch L,

. iI L

i Reviewed by:

Vendor l/ lkrrholm, cme ~f '

Leif J Date Inspector Branch Division of Reactor Inspection

't and Safeguards Of e f Nuclear React gulation-

-[

Approved by:

E

%d 0

flfrian K. Grimes, Direc' tor -

Date_

~

Division of Reactor Inspection and Safeguards j

Office of Nuclear Reactor Regulation i

i t

39-

. - ~..

j

i TABLE Of CONTENT $

EA92 EXECUTIVE

SUMMARY

..................................................... i s

1 INTRODUCTION.................................................... 1 3

i 2

COMMERCI AL GRADE DEDICATION PROGRAM REVIEW.....................

2 j

2.1 Procedures Review.......................................... 2 2.2 Commercial Grade Supplier Surveys..........................

7 2.3 Source Inspections.........................................

9 2.3.1-Ebosco Contracted Source inspections...............

10

2. 4 Pa r t s Cl a s s i fic a t i on.......................................

10 --

2.5 Material Upgrades.........................................

13 2.6 Follow-up on NRC Procurement Inspection Report (50-302/89-200). Corrective Actions..................

15

2. 7 T rend ing o f Suppl iers......................................

17

2. 8 Rece i pt i n s pec t i on.. -.......................................

18 3

DED I C AT I ON PAC KAG E R E V I E d.......................................

18 4

PROCUREMENT AND DEDICATION TRAINING..............................- 28 5

EXIT MEETING..........,..........................................

30.

40

EXECUTIVE

SUMMARY

From March 23 through March 27, 1992, representatives of the U.S. Nuclear Regulatory r.ommission's (NRC's) Vendor Inspection Branch (VIB) and Region 11 conducted an inspection of Florida Power Corporation's (FPC's) activities related to the procurement and dedication of commercial grade items (CG!s)it 3 used in safety-related applications at the Crystal River Nuclear Plant, Un (CR3). The inspection team reviewed FPC's procurement and ded3ation program to assess its compliance with the quality assurance (QA) requirements of Appendix B to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR Part 50).

On August 24, 1990, the NRC staff forwarded to the Commission SECb 90-304, "HUMARC Initiatives on Procurement," in which the staff reported the status of the Nuclear Management and Resources Council's (NUMARC's) initiatives on general procurement practices.

Procurement initiatives as described in NUMARC 90-13, " Nuclear Procurement Program improvements," dated October 1990, comitted licensees to assess their procurement programs and take specific action to strengthen inadequate programs. The industry initiative on the dedication of C0ls, which was to be accomplished by January 1, 1990, stated that licensee programs should meet the intent of the guidance provided in the Electric Power Research Institute (EPRI) Final Report NP-5652, " Guideline for the Utilization of Commercial Grade items in Nuclear Safety Related Applications (NCIG-07)," dated June 1988. The staff also stated in SECY-90-304 that it would conduct assessments at selected sites to review the licensees' implementation of improved procurement and commercial grade dedication programs, assess improvements made in the areas covered by the NUMARC initiatives, and report the results of those assessments to the Commission. From February to July 1991, the NRC's Vendor inspection Branch conducted eight assessments of selected licensees to determine the current status of activities to improve the procurement programs related to industry initiatives and NRC requirements. On September 16, 1991, the NRC staff forwarded to the Comm15sion SECY-91-291, " Status of HRC's Procurement Assessments and Resumption of Programmatic Inspection Activity,"_in which the staff reported on the results of its assessments and noted that it was resuming inspection and enforcement activities.

The NRC conducted this inspection, the third since completing the eight assessments, to review FPC's procurement and dedication program and its implementation since January 1, 1990 (the effective date of the NUMARC initiative on dedication of CGis), and also to review the corrective actions for certain installed CGis that were found to be of unverified quality in the-1989 VIB procurement inspection report (Report 50-302/89-200) and in the subsequent Notice of Violation. The inspection focused on a review of procedures and representative records (including approximately 40 procurement and dedication packages for mechanical and electrical CGis); interviews with FPC staff, including senior management and CR3 site personnel; and observations by the inspection team members. The inspection team also held meetings with FPC's management to discuss relevant aspects of commercial grade dedication and to discuss areas requiring additional information. The inspection team findings were discussed with FPC's representatives and senior management at the exit meeting held March 27, 1992. The inspection team

-i-41

identified three deficiencies which are summarized below.

DdhhmLR=l01&i The inspection team identified a number of examples where f PC either installed CGis in safety-related plant applications or had identified them as available for installation into safety-related applications at CR3 without adequate review for suitability of application of these materials, parts and equipment, that were essential to the safety-related functions of the structures, systems and components.

fPC failed to adequately determine the acceptability of using CCis in safety-related applications which resulted in the use or warehousing of CGis of indeterminate quality.

Examples of this practice included:

A fuel injector adapter nozzle for the emergency dietel generators (EDGs) procured under PO f6/0284K.

F our sets of thrust bearings for decay heat removal pumps procured under PO f670378V.

Check valve parts for swing check valves used in the raw water system procured under PO f6704047K.

A centrifugal pump and motor assembly for EDG standby jacket water cooling pumps procured under PO f84?798V.

DdkhMLM1101-02 The inspection team identified several weaknesses in the generic procurement program and in the licensee's actions to implement it. These weaknesses contributed to the specific examples of deficient CGI dedication described in Deficiency 92-201-01.

The most significant programmatic weakness resulted from the division of dedication responsibilities between the procurement engineering (PE) staff and the procurement and material quality assurance (PQA) staff. This division of responsibilities resulted in numerous instances in which the licensee inadequately transferred critical characteristics from the functional analysis / critical characteristics review form (FACCR) to the receipt inspection plan (RIP) and source inspection plan (SIP) and ultimately resulted in the dedication of CGis that did not demonstrate suitability for safety-related applications.

The program included no requirements for feedback to PE or for the PE staff to review the PQA staff specification of verification methods and acceptance criteria as a means to obtain assurance for the characteristics listed on the FACCR.

The team also determined that a lack of procedural guidance often led to the inappropriate or incomplete identification of FACCR attributes including the name and functional description of the CGl's parent component; the parent component's safety function; the piece part's (CGI) safety function; the failure modes and effects adverse to safety; and characteristics necessary to ensure performance of the safety functions or prevent the effects of the identified failure modes.

These f ailures, coupled with the inadequate transfer of characteristics into means to obtain assurance by PQA, ultimately 42

i i

resulted in CR3 using or warehousing CGls of indeterminate quality, Another weakness in the CR3 dedication process concerned the performance of source inspections by outside contractors and the use of their documented reports as part of the dedication process.

In numerous instances the i

inspection team found that the source inspection report or vendor quality assurance report provided by the contractor, and accepted by fPC, did not contain adequate evidence to support assurance for the characteristics listed on the SIP.

In many cases inadequate guidance was provided to the source inspectors for documenting objective evidence and ensuring that SIP attributes were properly or adequately established.

In most cases, the source inspection reports did not support the source inspector's review to determine that no design, material, or manufacturing changes had occurred that would affect the form, fit, or function, and material traceability for a specific CGl.

The source inspection activities should have ensured that documentation received from the vendor, such as certificates of conformance (CoCs) and certified material test reports (CHIRs), was acceptable and meaningful. Additionally, the licensee had not implemented procedures to require the PQA or PE staff to review source inspection activities (usually documented on the SIP by the source inspector) and source inspection reports for technical adequacy and accuracy.

Dgfiriency 92-201-03 1he team determined that fPC's corrective actions were inadequate concerning the procurements which formed the basis of the NRC's December 1,1989, Notice of Violation and Proposed imposition of Civil Penalty. On April 27, 1990, the NRC withdrew the Notice of Violation and Proposed Civil Penalty "without-reaching the merit," that is, the notice was withdrawn for regulatory reasons unrelated to the merits of the findings for which the violations were cited.

However, the NRC expects licensees to take timely and appropriate corrective action to rectify deficiencies, regardless of NRC enforcement actions.

The inspection team reviewed the licensee's disposition of the seven examples

-and found that subsequent to the inspection, FPC had documented justifications for continued operation (JCOs) for the CGis in the seven examples, but had not completed the corrective action in all cases. During this inspection the team determined that the corrective action for the following examples had not.been completed:

Example 2 - Material Qualification form (MQF)- 1433-89.

FPC had not yet replaced the ASCO solenoid valves for main steam valve 148. These valves were scheduled to be replaced in the April 1992 outage.

Examples 4 and 5 - MQF 1332-88 and MQF 1301-87.

FPC had purchased-replacement molded case circuit breakers (HCCDs) for-the inservice ITE MCCDs, but had not yet installed them. FPC had not been able to resolve the differences between the MCCBs currently in service and the replacement MCCBs. Therefore, the original JC0 still remains in effect.

-lii-43 I

Example 6 - HQF 972-85. The Johnson Controls solenoid valve remains installed in the air damper system under the original JCO. FPC stated that the system undergoes monthly. periodic testing-The untimely completion of the corrective actions in nearly 3 years is considered by the staff to be of concern. The MCCB issue will be reviewed during a future inspection, i

1 1

-iv-44

i 1

INTRODUCTION During this inspection, the team reviewed the florida Power Corporation (FPC) program and its implementation for the procurement of commercial grade items (CGis) used in safety-related applications at the Crystal River Nuclear Plant, Unit 3 (CR3).

The team also reviewed the TPC program and its implementation for determination or verification of suitability of those CGIs for their intended or approved safety-related applications, a process referred to as

" dedication.

Part 21 of Title 10 of the Code of federal Regulations (10 CFR Part 21) defines dedication as the point at which an item or service becomes a_" basic component," that is, essentially an item (or service) with safety-related functions. However, the 10 0FR Part 21 definition of CGI (Section i

21.3(a)(4)(a-1)), distinguishes CGis from items procured as basic components.

The regulation, then, allows the procurement of items that are to become basic components that meet the definition of CGis without invoking 10 CFR Part 21 in the procurement documents.

When CGIs are procured for safety-related service, their procurement and dedication constitute activities affecting quality, and, therefore, _ these activities must be controlled in accordance with the requirements of Appendix B to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR Part 50).

In particular, Criterion 111. " Design Control," and Criterion Vil,

" Control of Purchased Material, Equipment, and Services," of Appendix B are most pertinent to the procurement and dedication of CGis. Therefore, the team reviewed the FPC program governing these activities and the implementation of that program for compliance with these and other applicable Appendix B criteria and with the requirements of 10 CFR Part 21.

Additionally, the NRC has provided further guidance to the requirements of-Appendix B as they pertain to the procurement and dedication of CGis in NRC Generic Letter (GL) 89-02, " Actions to improve the Detection:of Counterfeit and fraudulently Marketed Products," dated March 21, 1989, and GL 91-05,

" Licensee Commercial-Grade Procurement and Dedication Programs, dated April 9, 1991. Therefore, the FPC CGI procurement and dedication >rogram and its implementation were also evaluated for consistency with tie guidance and NRC staff positions promulgated'in these generic letters.

finally, with respect to procurement in general, includ_ing procurement and dedication of CGis, FPC has committed to various industry standards and other publications (as endorsed or conditionally endorsed by NRC regulatory guides

[RGs), NUREGs,~and generic-letters [GLs]), as stated in the FPC QA program description, as contained or referenced in the FPC Final / Updated Safety Analysis Report for CR3, and as expressed for the industry by.the Nuclear Man--

. +

agement and Resources Council (NUMARC) in the NUMARC initiative.on the dedica-tion of CGis as part of NUMARC 90-13,_ " Nuclear Procurement Program Improvements." In particular, FPC, like other nuclear utilities, was committed to have established a program for procurement and dedication of CGis i

consistent with Electric Power Research Institute (EPRI) Final Report NP-5652, L

" Guideline for the Utilization of Commercial Grade items in Nuclear Safety Related Applications (NCIG-07)," on or before January 1, 1990.- The acceptance L

4,

~i 45

.u

methods described in HP-5652 were conditionally endorsed by the NRC in GL 89-02 and the NRC Staff positions on several dedication issues were later clarified in GL 91-05. IPC stated that it had implemented this commitment as of October 1989.

2 COMMERCIAL GRADE DEDICATION PROGRAM REVIEW 2.I f tpcthnL laticw i

IPC described and prescribed its program for procuring and dedicating CGis for safety-related applications at CR3 in three groups of procedures:

the Nuclear Protunment and Storage Manual (NP&SM), the Nuclear Engineering Procedures (NEPs), and the instructions and other directives of the Procurement Quality Assurance (PQA) and Material Quality Control (MQC) organizations. The PQA, MQC, and Nuclear Procurement Engineering (PE) organizations use these groups of procedures to perform the various tasks involved in procuring and dedicating CGis at CR3. lho NP&SM includes general guidance and all the pertinent FPC policies and general procedures for procuring and dedicating CGis at CR3.

lhe NEPs and the instructions and other directives of the PQA and MQC organizations provide more guidance on subjects related to procurement.

The principal NP&SM sections relevant to this inspection were Section 2.0,

" Classification of items and Services," Section 3.0, " General Requirements for Procurement Documents," Section 5.0, " General Purchase Requirements of Safety Related (SR) Items and Services" (sic), Section 6.0, " Safety Related Procurement Methods," Section 7.0, " Evaluation and Control of Supplier Performance," and Section 8.0, " Receipt of Shipments and Receiving inspection." The team reviewed the currently effective revision of the NP&SM, Revision 7 of October 31, 1989. FPC issued this NP&SM revision to implement the NUMARC CGI dedication initiative and to adopt the dedication methodology of EPRI NP-5652, and this was the revision in effect for the individual dedication documents selected for review.

The two main NEPs of interest were NEP 215, " Configuration item Data Control,"

and NEP 254, " Plant Equipment Equivalency Replacement Evaluation." The team reviewed the it. test revision of NEP 215 (Revision 10, dated December 31, 1991), which contained additional guidance on the safety classification of parts. The staff reviewed the latest revision of NEP 254 (Revision 5, December 31, 1991), which governed the process of like-for-like or equivalency determinations for replacement items.

The inspection team reviewed Sections 2.0, 3.0, 5.0, and 6.0 of the NP&SM and found it generally consistent with the provisions of EPRI NP-5652. However, these sections included only limited guidance on the principles and considerations to observe while obtaining critical characteristics by deriving them from safety functions and other requirements for verifying the suitability of an application. For example, the licensee must consider failure modes adverse to safety. To supplement this guidance, the licensee included examples of safety functions, critical characteristics, and failure 46

modes with some standard sets to be applied to certain types of components in generic dedications.

NP&SM Section 6.0 contained a general description of FPC's four procurement methods.

In this section FPC stated that the FPC specification method and the FPC verification method were the only methods authorized for procuring FPC's

" nuclear grade" items, which are basic components requiring purchase orders that include 10 CFR Part 21 applicability statements. However, FPC stated that CGis could be procured using any of FPC's four methods as discussed below. The inspection team noted as a strength that all of the FPC procurement methods involved making an engineering determination of critical characteristics using the functional analysis / critical characteristics review (FACCR).

Section 6.1 described FPC's specification procurement method ("S" procurement) requirements in detail.

Section 6.1.4 covered the use of specifications or

" mini specifications" to procure CGis, which can be used only when the component need meet no design requirements unique to nuclear facilities.

FPC used this method for more complex items, for specifying non-nuclear unique options, and for specifying special test requirements that were to supplement the manufacturer's product information.

The supplier must be on the approved nuclear supplier list (ANSL), and acceptance could be by any combination of inspections and tests, surveys, or source inspections as determined by the PQA organization.

NP&SM Section 6.2 described FPC's verification procurement method in detail.

FPC uses this method to procure items for safety-related service from suppliers who, for various reasons, may not be on the ANSL.

FPC used the verification method for items for which it must specify in the procurement documents that 10 CFR Part 21 applies and for items defined as CGls. FPC would perform this verification method for components covered by 10 CFR Part 21 if the supplier's QA program lacked certain elements or was not properly implemented to the extent that would require FPC to supplement it with portions of FPC's own QA program activities including one or more of the following:

source surveillance or inspections and special receipt inspections and tests.

FPC also used this verification method in procuring CGls similarly to EPRI CGI acceptance methods 1 and 3.

Section 6.2.2 prescribed the portions of the process for which the PQA organization was responsible.

These actlons included preparing a critical characteristic verification plan, although its use was not evident in most of the dedication packages reviewed. Most often, PQA personnel apparently used a receipt inspection plan (RIP) instead of a verification plan. FPC authorized this practice in a note in the procedure that applied to a specific set of circumstances. However, FPC personnel apparently loosely interpreted this note to allow the practice as a rule rather than as the exception.

If the circumstances warranted preparing a source inspection plan (SIP), the PQA organization would perform this task.

The licensee would document on the RIP the fact that the source inspector completed the " statement of conformance" authorizing shipment of the CGI after completion of the source inspection.

However, neither the PQA nor the PE organization was required to review the RIP, the SIP, or their results. 47 1

NP&SM Section 6.3 prescribed the procedures to follow for FPC's catalog procurement method ("K" procurement).

The procedure allowed for performing "K* procurement for CGis (for SR applications) when the manuf acturer's published product description, including the catalog, instruction manual (s),

and drawing (s), is an adequate product specification for the applicable technical requirements, lhe procedure also called for the supplier of K-procurement items to be an " approved" CGI supplier (listed on the ANSL) for the CGI to be supplied, llowever, the procedure did not specify the FPC

" class" ("A,"

"B," or "C") of supplier, described as the ANSL level of approval, as discussed below. However, this method was not analogous to any particular EPRI acceptance method; rather, it could require (in addition to standard receipt inspection as a minimum) one or more of the following:

a commercial grade survey of the supplier (EPRI method 2), and source inspection (s) (EPRI method 3), and special tests and inspections (EPRI method

1) af ter receipt (including post installation), depending on the ANSL approval level or class of the CGI supplier. A particular strength of the catalog method was that it required a formal engineering evaluation of the adequacy of the catalog or other product information for use in specifying the technical requirements of the item, lhe catalog method also required that FPC document the evaluation on a safety-related catalog evaluation form (Attachment 68 to the procedure). This form required the PO to specify a supplier certificate of conformance (CoC) to the catalog (or other) specifications, and provided for including in the P0 the source inspection requirements, if specified by the PQA organization. Another strength of this met!.:,d, as described in the procedure, was the use (as specified by pE on the catalog evaluation form) of a configuration certificate for replacement items, Attachment 60.

The supplier would document and certify on this form any changes to design, material, manuf acture, or interchangeability (or to propose alternative with changes described) since a previous procurement referenced on the form.

However, most of the dedication files reviewed, to which this process would be applicable, included a standard engineering letter instead of Attachment 6C as the means of requiring design, material, and process change history and ef fect on, seismic fragility, sensitivity or other aspects of the component.

NP&SM Section 6.4 described FPC's commodity procurement method ("C"

~

procurement) in detail. This was the prescribed method for procuring generic material (or services) that were supposed to be widely produced for general industry use according to nationally recognized standards. The licensee listed rolled stainless steel plate conforming to "ASIM A480-75" as the example. The procedure had the general requirement that such commodities be ordered with manufacturer's markings or be in sealed, marked containers.

In the procedure, the licensee did not authorize pressure boundary materials that must meet American Society of Mechanical Engineers (ASME) Code Section III and Specification B31.7 ("S" only).

The licensee also did not authorize certain other items such as items not marked by the manufacturer or not of a standard mill size and thus requiring ANSL-listed suppliers. A strength of this procedure was that the licensee required a formal documented engineering evaluation using the commodity evaluation sheet, Attachment 60 (with detailed instructions on Attachment 60-1). However, this section lacked adequate guidance on acceptance methods, particularly those for verifying material. 48

Tiie team noted the following categories of programmatic weaknesses:

(1)

The procedures did not provide sufficient guidance to aid in performing a technical evaluation.

For example, they did not include detailed guidance for documenting the name and functional description of the CGl's parent component and did not include a requirement or provision for this purpose on the FACCR form. Rather, the FACCR only provided for entering the tag number of the parent component.

The procedures also did not include detailed guidance on the considerations for determining i

the safety function of the parent component, rather giving only simplistic examples.

FPC claimed that procurement engineers do not need this guidance because they have all necessary design information available in the configuration management information system (CMIS) and have all necessary information on procurement available in the fully integrated materials information system (FIMIS). However, the team observed numerous examples in which the safety function (s) of the parent component were inappropriately specified, including listing parent-safety functions that are too remote from the function of the replacement piece part or too general to be used in analyzing the part's safety function (s) and failure modes, or its contribution to (or degradation of) the performance of the parent's safety function. The team concluded that this weakness contributed to the numerous examples observed of incomplete identification (or inappropriate identification) of part safety functions and failure modes that could affect the ability of the parent or surrounding safety-related equipment to perform its function.

l (2)

The procedures did not include detailed, specific guidance on the principles, process, methodology, and considerations to be used in deriving the critical characteristics of the CGI from its safety functions, failure modes, or any other safety application suitability requirements, including those -for seismicity and sometimes those for I

environmental qualification. The procedures included only simplistic l

examples and standard applications. The team observed numerous examples of inadequate or incomplete specification of critical characteristics in FACCRs. Some listed were necessary for dedication but inconsistent with the identified safety functions and failure modes.

In addition, some that would be essential to the performance of the identified safety functions or necessary to prevent degradation or identified failure modes were inappropriate or omitted altogether.

The team noted a significant strength in the FPC program in that critical characteristics were defined in NP&SM 5.0.3, Note 2, as attributes " essential to form, fit, and functional performance" that

" provide assurance that the item will perform its safety function." The definition in Note 2 was consistent with the NRC's position expressed in Generic letter 91-05. However, weaknesses in the program prevented FPC from using this concept of critical characteristics in practice.

(3)

The division of responsibilities between the PE and PQA organizations was the most fundamental weakness in the FPC dedication program.

PE was responsible for the first part of the technical evaluation process:

5-49

.. - ~.

. _. ~..

4 determining:stfety functions, ' failure modes and effects', and critical-characteristics.

The. PQA_ staff had _the engineering line responsibility of translating the critical characteristics into verification methods..

and acceptance criteria. However, FPC had not 'establishec' requirement L

for PQA responding to engineering or for the engineering sJf to review-E the verification methods and acceptance criteria specified by the-PQA-organization for-the identified critical-characteristics even though the engineering staff has design responsibili.ty and acces'; to design:

information. Furthermore,-FPC did;nct require other parts of the quality assurance organization _ to review this information ~ for consistency with engineering and quality-requirements.

The inappropriate division of responsibility, the lack of response and review, and the often inadequate derivation and specification of-critical characteristics on the FACCRs, resulted in the numerous examples of verification methods and acceptance criteria that the team found to be inadequately developed from the critical characteristics.

The team found that many verification plans (VPs), source inspection plans (SIPS), and receipt inspection plans (RIPS) written by P0A were one or more of _the following:

incomplete, inappropriate, or lacking specificity necessary for meaningful execution.

The procedures of NP&SM Section 3.0.;rst established the division of-dedication responsibilities with other guidance on PQA's role in verifying the critical characteristics described in Section 7.5..

Howe,er, the team noted that the good definition.of critical-characteristics in Note 2 of NP&SM Paragraph 5.0.2 was-severely: weakened by the next statement in that note:

" Selected Critical Characteristics identified in Sectica C cf Attachment SA [the FACCR form prepared.by __PE) may take a different form than a critical characteristic for acceptance developed by Procurement Quality Assurance in accordance with1

Sectior,

'.5."

Moreover, the example given indicated that PQA might specify " additional critical characteristics for acceptance" of markings-and hardness M Nuclear Engineering (PF] had specified shear and-tensile str mgth. d ductility. The_ note further stated that-markings-and material ! w cen could-provide reasonable assurance that-the material specitjod f s the material _ received.

Unfortunately, the explanatory statements and-the example in Note 2 had-P two undesirable effects on the program and on its implementation:

(1) they did not support..or clarify the stated definition of-critical L

. characteristics, but rather negated the dedication principle of specifying critical characteristics that are derived from safety functions or for prevention of failure modes adverse.to safety, (2) formally establi ~ 'd the notion that PQA could interpret the! critical characteristics

,ted by PE as it deemed appropriate, which often ~

t p

resulted-in the i mropriate translation.of PE's~ requirements.

FPC's procedures _c not_ include adequate guidance on including seismic-and EQ performance spabilities as critical characteristics to ensure L

that they will be pr.perly verified where required. Seismic L

qualification was addressed most frequently either through the plant'

. S 50 7

y w-

-e

equipment equivalen y replacement evaluation (PEERE) process or by a

requiring vendor certifications in purchase orders (P0s) using attached, standard " engineering letters," such as letters 1082, 1137, 1176, and 1187. Such letters usually asked the vendor to certify that the design, material, or manufacturing process had not been changed in any manner that would af fect the form, fit, function, or " structural integrity" (and later, that would affect seismic perfcrmance or fragility or sensitivity) since some previous date listed or referenced previous procurement. However, the team found that because the procedures did not include specific detailed instructions for determining and verifying this fact, the vendor certifications to this effect were often not adequately verified or validated by FPC (or most frequently, by FPC's contractor source inspectors).

Thus, these suppliers were not usually-surveyed (including evaluating design-controls) to supplement the source inspections.

This resulted in the numerous examples in which the team found that FPC had not established an adequate basis for demonstrating that replacement parts were equivalent (like-for-like) or were seismically (or sometimes environmentally) qualified.

The team identified as Deficiency 91-201-02 the weaknesses found in the FPC procurement and commercial grade dedication program and in the implementing procedures for this program.

C.t

[gmmercial Grade Supplier Surveys sction 7.0 of the NP&SM contained general requirements for evaluating ipplier nd maintaining the CR3 ANSL.

The team reviewed the current

! vision,f NP&SM 7.0, Revision 4, of September 30, 1989.

Section 7.2

' dressed e>aluation of commercial grade vendors.

IPC classified CGI vendors

. cording to the degree to which their documented commercial quality control programs were effectively implemented and to which their activities controlled the critical characteristics applicable to their scope of supply. Tr e classification also reflected the type and degree of verification required, Each class of vendor carried its own requirements for ANSL a-oroval/ restriction codes. The procedure required that FPC accept CGI components procured from a Class A CGI vendor if FPC had verified by survey that the vendor maintained adequate control of the critical characteristics applicable to its scope of supply. The procedure specified that this survey must includ, as a minimum, a CR3 standard receipt inspection, and a review of a CoC attesting that the item / service provided was processed under the specific FPC-ap,; roved ccmmercial quality program identified by manual [ title and number], revision, and date.

The requirements for Class B vendors were similar except that FPC had only conditionally approved the vendor's quali.ty program and must verify separately (after receipt) the critical characteristics not adequately controlled by the vendor because of missing program elements or identified weaknesses.

FPC classified as Class C those CGI vendors for whom it had no survey rsport (or acceptable report).

Prospective CGI vendors were to be initially classified "C."

The team found as a strength of the process of controlling procurement from commercial grade vendors the verification requirements in Section 7.5 for the three classes of CGI vendors.

This section included more detailed requirements for VPs, RIPS, and SIPS. Ilowever, a fundamental weakness was that the procedures did not 51

. - ~ - - -

require-these' plans.to b' reviewed or approved by engineering..'and did not e

require final review of_ the results for technical acceptability.

Section 7.2.5 addressed commercial grade surveys. performed by organizations other than FPC. The general-guidance. in-Section 7.2, was similar to. that' for Appendix B suppliers:

use these where available and acceptable.

The requirements for reviewing such survey reports were appropriate to determine acceptability commensurate with the requirements for FPC-conducted surveys, covered in Section 7.2.6.

However, a weakness common to both sections was the stated practice of.having PQA convert the critical characteristics established by engineering into vendor QA/QC program controls deemed by PQA to be necessary to control the critical characteristics. _These procedures did not require FPC to verify at the vendor's facility that the critical characteristics for the type of items actual _1y being procured are specifically and effectively controlled.

FPC approved CGI vendors triennially, with annual program reviews. The team noted as a strength the formal requirement to review each vendor's performance annually based on " receiving and source inspections, surveillance reports, NRC reports, etc.;" although the-procedure did not specify dedication testing, inservice testing and surveillance / inspections, and inservice failure data from FPC and the industry. Nevertheless, the team was concerned that arbitrary triennial surveys with annual program updates and performance reviews may not be adequate coverage depending on several factors, including',

but not necessarily limited to (1) the complexity of the CGI(s) in question; (2) the frequency and size of purchases; (3) the critical characteristics to be verified by survey and the extent to which those are relied upon to support dedication; (4) the strength of the supplier's controls on design, materials, manufacturing processes, and subsuppliers of parts and services; and (5) the strength of the supplier's commitment.or obligation either to not make changes-in certain products, or at least to inform the customer of any-changes made.

The procedures did not provide for increasing or decreasing the frequency of-the survey on the basis of these factors. _ The procedures also did-not address the survey of distributors where applicable in. addition to the manufacturer in; accordance with the GL 89-02 comments on the use of EPRI method 2.

Section 7.6 addressed the conduct of source inspections, requiring inspectors,-

FPC or contractor, to be qualified in accordance with'American National Standards Institute (ANSI). Standard N45.2.6.

However, the procedures did not require FPC-to verify the inspector's technical knowledge or experience in the-areas to be inspected, and did not require field surveillance of the performance of contractor-provided source inspectors., Although1 the-procedures-required written inspection reports, they did not require either PQA or PE to-review the results of the source inspectio_n for either quality or technical accuracy, except'where there was a; formal supplier exception, deviation-a request, or nonconformance report generated.

To assess the_ effectiveness of the implementation of FPC's commercial grade survey program in support of dedication (though very. limited), the team also l

reviewed two completed survey re' ports for some of the individual-dedication

. packages reviewed. The two commercial grade surveys-reviewed were performed 52

to verify specific critical characteristics 'and appeared to be adequately performed and documented.

2.3-12 tree insoections Tha team observed a-fundamental weakness with FPC's use of source inspections.

At CR3 FPC performed source inspections as an acceptance method analogous to EPRI~ method 3, " source verification." The team noted that, among the' dedications reviewed, the source inspection was the most commonly used method--

of verifying critical characteristics.

FPC performs source inspections for its Class B or C commercial suppliers.in conjunction withts verification ("V") and catalog ("K") methods of procurement.

FPC placed the procedure for conducting source inspections in NP&SM Section 7.5, " Establishing Source, Receipt, and Post-Installation Verification Requirements," and NP&SM Section 7.6, " Conduct of_ Source Inspections." The team reviewed the current revision of Section 7.5 (Revision 7, January 31,1990), and of Section 7.6 (Revision 5.1, January 30, 1990).

The team attributed the inadequacy of source inspection activities to.three principal causes:

(1)

FPC gave inadequate guidance to the source inspectors for documenting evidence and raw data -for verifying critical characteristics..The detailed instructions for the source inspection report called for_ a narrative summary of inspection activities but did not require that-the particular critical characteristics be listed and their method of assurance or results be documented.

Accordingly, the team found numerous examples in which' FPC's contractor source inspectors inadequately executed SIPS. These inspectors were employed by Bechtel and Ebasco Services and were presumably qualified, according to ANSI N45.2.6. :In some instances,:the team found clear indication that the~ contractor inspectors had _not verified SIP attributes.

In other cases, the verification war, not by the method specified, and in others the team found insufficient evidence to determine whether the-SIP attribute was properly verified.

l

'(2)-

There was a lack of review (and lack of requirement.for review) of the technical adequacy and accuracy of the source inspection reports.

Receipt inspectors-were. required simply to document-that the " statement of conformance" (a QA shipping release) was signed by the source; inspector.and received in the documentation package with the item.

h The team found that-NP&SM Paragraph 8.4.9 required that MQC transmit completed P0s including associated documentation to the purchasing clerk, as opposed to PQA. FPC explained that this requirement did not exclude PQA from the routing when ina procedure was written because, at-L that time, PQA was part of the sar..c organization as MQC. However,.this was no longer the case, and had not been for some time. -Therefore,- the procedure, as written, did not require PQA to receive the documents'from

-g_

53

HQC for review. An dQC representative stated that an interim change would be or was being prepared to reflect the desired routing and current organizational structure.

In addition, the team noted that PQA had no requirement or practice to review the source inspection reports for adequacy and accuracy, that is, for compliance with the SIP and sufficient documented evidence of critical characteristic verification.

(3)

The procedures did not adequately address review of documentation to establish the traceability of the CGls as received to their original equipment manufacturers (OLMs) which would be necessary to establish the validity and applicability of vendor controls and vendor supplied information or documentation tc the extent they are relied upon to support dedication or qualification.

2.3.1 MHgghninchLiqnce Insintion The team reviewed FPC's basis for accepting source inspection activities performed by Ebasco Services, Incorporated (Ebasco). The team reviewed numerous documents to evaluate FPC's acceptance of the source inspection activities and interviewed FPC staff to obtain a better understanding of FPC's methodology for acceptance of these services.

The documents reviewed included (1) the Florida Power and Light Company's (FPL's) Aueust 7, 1990, audit report of the establishment and implementation of its quality assurance program at Ebasco's Lyndhurst, New Jersey, facility; (2) Ebasco's source inspection reports (SIPS); (3) FPC's vendor quality assurance reports; (4) FPC contract No. N00337AD of December 14, 1990, with Ebasco; and (5) Ebasco's procedures and nuclear QA program manual No. ETR 1001, Revision 14, of June 15, 1990.

The team reviewed the FPL audit report and found that it was not a performance-based audit and did not address the adequacy or effectiveness of the contractor's source inspection activities.

Instead, the FPL report stated that certain program elements were reviewed such as " verified qualification records for six inspectors indicating that certifications were adequate." The team noted that during FPC's review of the FPL audit, it found that Ebasco had not properly conducted source inspections during a previous contract with fPC.

Consequently, FPC staff recommended to its management that it conditionally accept Ebasco's services based, in part, on an FPC review of Ebasco's SIRS at the completion of its sou ce inspections. The team noted that PQA had initialed and dated the SIPS initially submitted by Ebasco, indicating that FPC had reviewed the SIRS. However, as discussed earlier, the team found that FPC had failed to establish any procedures, instructions, or policies to implement and control a review of source inspection reports.

2.4 Parts _Cl m ificat_ign The process of assigning safety classifications to items and services was described and prescribed by Section 2.0 of the NP&SM, " Classification of items and Services," (Revision 6, September 30, 1989) with additional guidance in NEP 215.

Equipment and components, i.e., items with unique system and function-related identifiers called " tag numbers," classified as safety-related in accordance with NP&SM 2.0 and NEP 215, would be listed in the CR3 configuration management informatii n system (CMIS) computer database that 54

_.g_

serves CR3 for a1Q-list. - The Safety Classification Rcview ~(SCR)L checklist

= forms-(Attachment 2C toLNP&SH 2.0):for_this process were used.to-document,the.

~

justification for classifying. items-differently-than their parent equipment,.

. component, or system was classified in the CHIS;- thatvis, to classify as nonsafety-related (NSR) those items with no identifiable safety-functions,:and presumably with no_ credible failure modes adverse to: the prformance of-the parent components' saf ety functions. - FPC would then establish the QA controls _

applicable to a procurement on the basis of the SCR through use of the

' Classification ofiltems_(COI) form (Attachment _2A to NP&SM 2.0) or-Classif' cation of Services form (Attachment 2B).

in performing the SCR, FPC determines the safety classification of the item b'y determining-if the function it performs is within any of the standard defined safety-related functional categories:

(1) maintaining the integrity of the reactor coolant pressure boundary, (2) shutting dawn the reactor and-maintaining it in a safe shutdown. condition, and (3) mitigating the effects of design basis accidents and preventing the'offsite release of_ radioactivity in-excess of 10 CFR Part 100 guidelines.

After evalut. ting the item and documenting this evaluation on an SCR form, the CR3 procurement engineer determines the procurement classification and handling of an item to be procured by completing a COI checklist form. The information on the C01-form is organized as a " decision tree" flow chart with<

which the classifier first documents the results of the SCR; If the SCR:

- determination is that the application is safety-related, the classifier would -

indicate on the COI whether or not the item is available from a supplier with a 10 CFR,-Part 50, Appendix B, QA program, and a program for reporting defects and noncompilance pursuant to 10 CFR Part 21.

If the item is.available from such_ a supplier, the classifier would designate it-as " nuclear grade"- (an FPC representative explained that a nuclear grade component is not a CGI,;but a -

basic component as specified-by 10 CFR Part 21).

The checklist directs-the classifier to' designate a nuclear grade item by appending the procurement documents with Attachment Q.

Attachment Q indicates, among other things, that.10 CFR Part 21, applies to the-procurement, thus invoking it on the supplier. Attachment Q also imposes QA--

r program and documentation requirements such as requiring'the supplier to malrtain a QA program in accordance with 10 CFR Part 50, Appendix.B, and:

applicable ANSI standards.

If FPC approved the supplier's'QA program and the supplier is listed on the ANSL, the checklist directs that the " specification" procurement method-("D")-be used.

If the supplier's QA program has.not.been 4

fully approved by'FPC and the supplier is not; listed on the ANSL, the checklist directs that FPC use the " verification": procurement method (type "V"),fas described in the NP&SH (Sections 2.0, 5.0, and 6.0).

3 If the item is not 'available from an Appendix B supplier, then the checklist -

~

applies the three tests 1or conditions from 10 CFR 21.3(a)(4)(a-1)-in question l

' form to determine if the item can be considered a CGI.

If the item fails.any:

of theitests, it. is considered nuclear grade, and the classifier is directed

~

to the-Attachment Q block; but tf all three conditions are satif fled, then the-item can be' considered a CGI and the checklist -directs the clas tifier to append Attachment "C" to the procurement documents and to use methods "O," :

55-u m

a w

.w-.

.,,w-g w,.

g g

y-g y

E "V," K" (catalog method) or "C" (commodity method). Attachment C imposes certain nonnuclear technical and quality requirements on the supplier (without affecting the design-of the item itself). such as invoking the supplier's commercial quality program (referenced by title and number and by the FPC-approved version, revision, and date) in the manufacture and supply ~of the item and requiring certification of conformance to that effect as recommended in EPRI NP-5652.

In reviewing these procedures, the team found potential weaknesses in FPC's safety classification process. First, the procedures and SCR checklists did not require FPC to consider failure modes adverse to surrounding and adjacent safety-related equipment or systems, not just the parent component, equipment, or system.

Second, the structure of the SCR excluded from classification as safety-related those reactor coolant system (RCS) component parts that are not considered " pressure retaining" under the applicable provisions (Sections III and XI) of the American Society of Mechanical Engineers (ASME) Boller and Pressure Vessel Code (the ASME Code). However, these parts (such as small instrument lines, valves and other fittings, certain valve parts, gaskets, seals, 0-rings, strainers, and filter media) may contribute to retaining pressure or preventing leakage or may have failure modes and effects that are i

adverse to safety.

Such parts may also be subject to system material compatibility requirements or requirements for exclusion of contaminants such as leachable halides or mercury.

Third, Section 2.3.2 stated that all chemicals, including lubricants, even after being classified as available for safety-related service as defined in Section 2.3.1, were to be procured in accordance with Section 4.4 covering "X" procurement; that is, they were to be procured as NSR with special requirements. This paragraph implied that FPC was to observe NSR controls in l

procuring all lubricants used in safety-related and even EQ applications, even though the engineering organization might impose special requirements.

Finally, Section 2.4.1.3 stsed that 10 CFR Part 21 applied to nuclear grade purchases and that the specification method ("0") was to be used for suppliers with 10 CFR Part 50, Appendix B, QA programs. However, it then required that nuclear grade item procurement from other suppliers should use the-l-

verification ("V") method; in which case, FPC would assume the 10 CF_R Part 21 reporting responsibilities "in lieu of the supplier."

In view of the FPC definition of the term " nuclear grade," a basic component, not a CGI per 10 CFR 21.3, Paragraph 2.4.1.3 would allow, contrary to 10 CFR'21.31, for FPC to procure a basic component without stating in the procurement documents that 10 CFR Part 21 applied. FPC maintained that the CGI checklist, if followed inst'ad, would not allow this because it required Attachment "Q" in all e

nuclear grade item procurement.

FPC agreed that Paragraph 2.4.1.3 was at least inconsistent with the COI checklist, and agreed to consider revising it.

The team noted that while the paragraph allowing a. violation of 10 CFR 21.31 did not in itself constitute such a violation, it was nevertheless l

inconsistent with the intent of the regulation.

i

. i i

56 l-

,r-m e

e 4

2.5 ~ tijLtfrial Uporadu During the previous NRC procurement inspection at CR3 (NRC Inspection Report.

No. 50-302/89-200), the staff cited violations of 10 CFR Part 50, Appendix B, upon finding deficiencies'_in the " material qualification" process in which FPC would complete material qualification forms (HQFs) to document dedication of CGIs_at_CR3. Since that time FPC has made changes in-this process-to correct the previous deficiencies.

FPC's. current practice is to. perform " material upgrades" and is significantly limited in scope compared -to material qualifications.

In this_ practice, FPC completes a material upgrade form (MUF) to document that it has verified as suitable for safety-related service those items originally procured as NSR items without any intention of dedicating them, but has reclassified them as safety-related.

The team reviewed Revision 7, of NP&SM Section 9.1 (September 30,1989),in.

which FPC listed the material upgrade requirements. While the MUF process as-described in this section represented a_significant improvement over the former MQF process,_ its ineffectiveness was limited by some of the same general dedication program and implementation deficiencies as cited above.

The team reviewed several MUF record files and made the following findings:

(1)

MUF 0007-90 was for 2 Dresser-Ashcroft model 30EI60LO25 bi-metal thermometers purchased as NSR items from Epperson & Company, under FPC P0 02004731, of January 31, 1990. The thermometers were replacements for model 30AI60LO25, and were identical to it except that they did not have a removable gauge glass bezel.

FPC used these thermometers in the lube oil system for the reactor coolant makeup water pump, part of the core emergency cooling support systems (Tag Nos. MU-52, -54, and TI). An Epperson packing list was the only evidence of traceability back to the distributor, Epperson, and supposedly accompanied-the items, although FPC-did not conduct a formal receiving inspection until one month after receipt. The team found no other evidence of traceability to the OEM in the file.

According to the FACCR, pressure retention was the only safety function of the device._ FPC upgraded this CGI because its application had been reclassified as SR _(pressure boundary) when maintenance revealed that these thermometers were installed in the lube oil system pipes without thermal wells as had been previously thought. The FACCR listed the failure modes--as material degradation and deformation and the effects as leakage. Although-the instrument manufacturer recommended thermal wells, FPC made the upgrade primarily. because the original thermometers satisfactorily performed the safety function of retaining pressure and because-the process interface of the replacements was supposed to be identical to that of the original. On PEERE 105, Revision 2, FPC.

documented the suitability of the replacement item relying-heavily on the thermometer's supposed all-welded stainless steel construction.

Accordingly, the FACCR listed part number, certain dimensions, weight, and all-welded stainless steel construction as critical characteristics.

However, on the RIP, FPC listed stainless steel construction (but not "all welded") with several other attributes such as dimensions.and 57

_m weight under characteristic 4. " Verify component configuration and-dimensions agree with th'ose listed below and~on the attached PQA/ catalog sheets,"_for which the inspection basis and criteria were given simply as "PQA/ catalog," and the inspection method was simply " review." FPC-did not _ list the specific catalog or the specified part number _ listed on the RIP.. Furthermore, the only documented evidence of stainless steel construction on the completed RIP.was the words " magnet --OK" with a check mark that had been handwritten in next to the attribute. -FPC did not indicate which of the welded parts were checked (stem, threaded plug, or both)'or whether the item was found to be magnetic or not.

Thus, this check was of questionable value and was inconclusive.

If the instrument was not magnetic, it could have been made o_f a nonmagnetic, but incompatible, alloy such as copper-nickel which would be virtually indistinguishable visually (on a new part) from stainless steel.

If it--

was magnetic, it would most likely have been a magnetic stainless steel, but-it could also have been, though unlikely, a much less corrosion-resistant metal, such as carbon steel.

Therefore, if a catalog description and verbal assurance during a telephone conference were the only evidence of the vendor's control of the material, merely performing a magnetic check would not be adequate to verify the material properties of a pressure retaining fitting for which material degradation was a concern, presumably to verify long-term corrosion resistance for leak ~

tightness.

(2)

MUF 0013-90 documented the upgrade and acceptance of a mounting plate for a States sliding link terminal block.

FPC upgraded the NSR material upon visually comparing the thickness and al_loy separators of the replacement-metal plate with a sample plate that.had been procured as a SR item. The plate was allowed a thickness tolerance of +/-10 percent and an alloy separator tolerance of +/-25 percent because the plate was required to have only a low strength.

This methodology was probably adequate for the strength requirement of the application. However, the work order under which the terminal blocks were being replaced (when found corroded) indicated that the mounting screws of the installed terminal block were so badly corroded that they could not be removed (presumably implying other means were used, such as drilling). This-raised the concern that the original terminal blocks _and their mounting plates may not have been suitable for their application / environment._

However, the team found no evidence that FPC considered any. alternative tc the existing type of circuit connection-device, despite the evidence of significant corrosion apparently on both the terminal screws and strips which necessitated replacing the terminal block and cleaning

" debris" in the bottom of the connection box), and despite the statement on the FACCR that the blocks should be environmentally qualified.

(3)

MUF 0014-90 documented upgrade of Amphenol 31-219 double-female BNC coaxial cable connectors (Mil Type UG-914/U) purchased as NSR items from the EMSCO company _ under FPC P0 F890648A, of May 16, 1990, for use in environmentally qualified (in-containment) sensor cables for the valve acoustic flow monitor system.

The basis for traceability to the OEM was stated on the MUF as the P0 to L l 58

l EMSCO and the Amphenol part number, but the FACCR specified an Amphenol marking or packaging and military part number. According to the verification block on HUF 0014-90, the connector was marked with the corresponding military connector number, UG-914/U; although no mention was made of Amphenol packaging or marking.

The receiving inspection report listed only the Amphenol part number without any military part number or reference to OEM packaging. The team found no other documented evidence of traceability to the OEH.

The EQ report (T068-3TR-003) by the sensor system manufacturer, Technology for Energy Corporation (TEC) specified this part as an Amphenol type 31-219 connector, military part number "VG-914" [ sic] with "KEL-F 81" dielectric (relied upon for EQ temperature and radiation characteristics), characteristic impedance of 50 ohms nominal, brass metal parts, and silver or golt finish. The Amphenol catalog pages in the HUF file listed other specifications such as voltage rating (500 volts peak), dielectric withstand voltage (1500 volts rms), center contact resistance of 1.5 milliohms, outer contact resistance of 0.2 milliohms, braid-to-body resistance of 0.1 milliohms, insulation resistance of 5000 megohms minimum, and RF leakage and insertion loss specifications. The material specification for a female center contact was silver-plated beryllium-copper and other metal parts were to be of

" silver-finish" brass. The EQ report, in its discussion of the sensor cabling, indicated the importance of the cabling, which would be influenced by in-line connectors. However, the FACCR listed the critical characteristics (other than markings) as nominal dimensions,

" material appearance of body: plated (silver in color) brass,"

electrical coatinuity of center conductor, and insulation resistance of "5 megohms at MOO VOC" [ sic) instead of 5000 megohms.

FPC provided no technical justii cation for choosing these critical characteristics i

instead of the specifications in the catalog and EQ test report, and stated no justification for deviating from the acceptance criteria in the catalog and in the EQ report.

Furthermore, FPC did not adequately verify certain critical characteristics, such as the material composition, indicated in the FACCR and gave no technical basis to justify a less rigorous means of verification.

2.6 Follow-up on NRC Procurement Inspection Report (50-302/89-2001 Corrective Actions On August 14, 1989, the Vendor Inspection Branch issued Inspection Report 50-302/89-200 on the previous inspection of the procurement program at CR3.

On December 1,1989, the HRC issued FPC a Notice of Violation and Proposed Imposition of Civil Penalty, which included seven examples of CGIs that FPC had installed in safety-related applications at CR3 without adequately selecting the item or reviewing its suitability. On April 27, 1990, the NRC withdrew the Notice of Violation and Proposed Civil Penalty "without reaching the merit," that is, the notice was withdrawn for regulatory reasons unrelated to the merits of the findings for which the violations were cited. However, the NRC expects licensees to take timely and appropriate corrective action to rectify these deficiencies, regardless of NRC enforcement actions. The team examined the seven examples and found their status to be as follows: l 59

-(l)~

Material Qualification Form (MQF) 1436-89. FPC replaced two ASCO solenoid valves for CAV-6-SV and CAV-7-SV (chemical addition and containment isolation) with nonnuclear grade valves and documented this replacement on Modification Approval Record-(MAR). 89-01-016-01.

FPC had dedicated these. valves merely by visually inspecting them, comparing their part numbers with those in the ASCO Catalog and verifying coil continuity.

FPC corrected the deficiency by replacing thsse two valves-with nuclear grade ASCO valves and documenting this replacement on PEERE 0044.

(2)

MQF 1433-89.

FPC had replaced four ASCO solenoid valves for MSVs 130 and 148 with ASCO valves with different electrical coils. FPC had dedicated these valves merely by visually inspecting them and comparing their part numbers with those in the ASCO commercial grade catalog.

FPC has since purchased replacement nuclear grade valves and obtained a :

certificate of compliance.

FPC replaced the valves for MSV 130 and scheduled to replace those for MSV 148 during the April 30, 1992, outage. The untimely replacement of MSV 148 valves is a concern.

(3)

MQF 1413-88.

FPC purchased Agastat time delay relays to' replace an original relay as a like component.

FPC dedicated the item by visually checking the physical dimensions and verifying the information.on the namepl ate. This item was originally thought to perform a safety-related function and hence the above dedication was considered inadequate.

However, upon reviewing the matter and discussing it with licensee personnel, the team found that the relays in this application did not have a safety-related. function and, therefore, required no further action by FPC.

(4,5) MQF 1332-88'and MQF 1301-87. FPC purchased ITE molded. case circuit breakers (MCCBs) as CGIs and dedicated these items merely by visually inspecting them to verify their dimensions and part numbers and -to check for physical damage. Although-the licensee contended that the MCCBs had been dedicated according to its procedures at the time, the team found the pertinent provisions of procedures at the time required certification of critical characteristics. - However, upon reviewing the dedication and installation documents, the team found that FPC had not tested the MCCBs to verify the critical characteristics of tripping on overloads and faults, as stated in the FACCR. FPC-had purchased but not yet installed replacement MCCBs that differ in certain respects from the l-originals. The replacements had a short circuit interrupting rating at 480 Vac whereas the' originals had a short circuit interrupting rating at

(

600 Vac, which FPC relied upon to demonstrate su hability of application.

The licensee prepared a corrective' action plan to resolve the matter.

L The licensee had not resolved the issue at the time of this inspection.

l When the NRC first-found this issue,:the licensee prepared a justification for continued operation (JCO) that remains in effect. The-P NRC will review this matter during a future inspection. The failure to-close out the corrective action plan in nearly 3 years is a concern.

! l 60

(6)

MQF 792-85. FPC used a Johnson Controls' solenoid valve, model V11HA--

'100, to replace a Johnson Controls Model V-24-2.

The solenoid valve 1s-

~

used for-AH-366-SV which controls supply air that opens.and' closes air handling. damper D-60.- Exhaust damper D-60 opens when the emergency diesel generator (EDG) supply dampers open and the ventilation fans start.1 FPC purchased the item as a like-for-like replacement and dedicated it by visually inspecting it to determine equivalency.

FPC performs-a monthly periodic test of the system. FPC purchased the item directly from Johnson Controls to establish traceability to the OEM with no middle supplier. FPC prepared a JC0 that still remains in-effect.

FPC considers the components operable and suitable for continued use.

The untimely corrective action is a concern.

(7)

P0 F9038125V.

FPC issued this purchase order for an anode used to protect service water piping against corrosion. FPC did not obtain assurance with respect to the material hardness or the physical and-chemical properties ' listed on the certified material test report provided by the-manufacturer. FPC did not establish traceability from-the material manufacturer through the vendor. The team reviewed this item and found that it is does not perform a safety-related function.

It is installed in safety-related piping to extend the service life of the piping system. Therefore no further action is planned. The item remains in service.

The team found the status of examples 1, 3 and 7 to be satisfactory. Examples 2, 4, 5 and 6 still require further action by FPC. The licensee has had corrective actions in place, but has not resolved all the issues in a timely manner. This failure to take timely corrective actions is a concern and is identified as Deficiency 92-201-03.

2.7 Trendina of Sucoliers The team reviewed FPC's process fu tracking and trending deficiencies in the performance of suppliers for CR3. FPC performs this activity as required by Quality Programs Administrative Procedure 27, Revision 10, "Nonconformance-Tracking / Trending System." The Supervisor, Quality. Systems prepares the reprt, and the Director, Quality Programs issues the report for review and action.- The report is directed to the Senior Vice President, Nuclear.

Operations, with copies to the appropriate managers.

The report contains three sections related to supplier nonconformances. The; first section describes the status of nonconformances issued, both closed and open. This section shows the backlog and age of open items. Most items are closed in less than 3' months. At present, four are more than 6 months old, and none are over a year. 'The items tracked and trended come from vendor audits, requests' for corrective action, vendor audit problem reports, and quality material problem reports. The second -section deals with supplier -

nonconformances by cause codes, such as unacceptable hardware, inadequate documentation and purchase order violations-The team reviewed these data:and found 46 percent of the nonconformances to be hardware related. Unacceptable documentation accounts for approximately 18 percent. The third section is an 61

analysis of the nonconformances of vendors that had 10 or more problem reports or nonconformances during the previous 18 months. Thus it reveals vendors who may warrant attention.

Each issue of the trend report indicates those vendors who were added, retained or '(by deduction) removed from the list.

The program appears to be successful, in that the report provides managers with information on supplier and vendor performance. However, the trending program does not include any information provided by outside organizations such as the Institute of Nuclear Power Operations (INPO) and the NRC, and other licensees through licensee event reports (LERs).

2.8 Receipt inspecti.gn The inspection team reviewed sections of the FPC receipt inspection program for commercial grade and basic components that were intended for safety-related use, including those sections on receipt inspection activities. The team observed FPC's " Stores" and quality control (QC) personnel performing receipt inspection activities, discussed the-inspections with CR3 Stores and QC staff, and reviewed portions of CR3's receiving inspection program. The team based its review and observations, in part, on the requirements of Appendix B of 10 CFR Part 50, the American National Standards Institute (ANSI) standards, the American Society of Mechanical Engineers (ASME standards, and certain industry guidance, such as the Electric Power Research Institute's (EPRI) Nuclear Procedure No. (NP) 6629, " Guidelines for the Procurement and Receipt of Items for Nuclear Power Plants (NCIG-15)." The inspection team found no concerns in this area.

3 DEDICATION PACKAGE REVIEW To assist the NRC in reviewing individual dedications, FPC prepared, at the NRC's request, a number of files of dedication records compiled from diverse records, but each pertaining-to one dedication, as selected by the team from a review of the lists of CR3 dedication files. FPC organized the review packages by the following disciplines:. electrical, instrumentation, mechanical, and materials (including lubricants).

FPC also provided the associated commercial audit or commercial grade survey reports in separate files. The team reviewed the records for the selected dedications including purchase requisitions (prs), requisition review sheets, catalog procurement review forms, P0s and attachments (engineering letters and copies of Attachment C), invoices, FACCR forms, RIPS, receiving inspection reports, SIPS and reports, and maintenance work requests (WRs). The following examples are-i items that FPC purchased as commercial grade and either installed or made

(

available for installation in safety-related plant applications without i

performing an adequate review for suitability for service.

(1)

FPC issued PO F670284K, dated April 3, 1990, to Coltec Industries for an adapter nozzle for the fuel in.jection nozzle on the EDGs. FPC installed the part in the EDG under modification approval record (MAR) 88-01-12-01. This MAR documents one of the modifications implemented to increase the rating of the EDGs. While reviewing the dedication package

[

for item M-1 the inspection team made the following observations:~ L 62

On March 23, 1990, PE prepared and approved the FACCR for this item, listing the safety function for the nozzle adapter as " fuel system pressure boundary." The failure modes specified were fracture and thread shear.

PE further stated on the FACCR that the effects of the failure could_ result in a loss of fuel or in fuel leakage to the EDG which could cause the engine to run roughly.

The only critical characteristics listed on the FACCR were (1) the vendor part number and (2) the dimensions and configuration. The-critical characteristics were not adequate to address the failure modes stated for the part. Since fracture and *% read shear were listed as the failure modes, material properties are important critical characteristics which should have been verified, but were not.

PE also did not specify which dimensions needed to be verified, such' as length and diameter.

The source inspection requirements for this part were listed in the body of the P0. The IP requirements stated that the source inspector was to verify the dimensions to the vendor-supplied drawing and to prepare and sign the FPC statement of conformance. While reviewing the FPC statement of conformance prepared by the source inspector on March 23, 1990, the inspection team noted that the source inspector marked "N/A" in the drawing number space. Therefore, the team questioned whether the dimensions had been verified to the vendor drawing. The statement of conformance was the only documentation provided by the source inspector and did not describe the scope of activity that he performed.

In summary, FPC did not adequately describe the safety function for dedication of the adapter nozzle and did not state all of the effects of the part's failure. The critical characteristics stated on the FACCR were not adequate to' address the failure modes identified for the part.

The FACCR did not specify the dimensions that needed to be verified.

Material was not identified as a critical-characteristic and thus was not verified. The statement of conformance did not clearly indicate whether or not the source inspector verified the dimensions to the vendor drawing.

(2)

This item was a liner between the adapter check valve and the cylinder for-the EDG air start check valve. FPC procured this item as a catalog item under the same PO as was issued in item M-1.

FPC installed this part under MAR 88-01-12-01.

PE prepared and approved the FACCR for this item on March 23, 1990. The FACCR stated that the safety-function of the parent component (EDG) was to provide emergency power during-accident conditions. The EDG air start system appeared to be the more appropriate parent component based on the safety function of the part -

stated on the FACCR. The EDG appeared to be too far removed from the specific part in order to have relevant safety functions and failure modes. The FACCR stated that the safety function of the part was to contain air _for the air starting system.

It appears that-the more appropriate safety function for the'part was that it provided a flow path in the' starting air system. The failure modes specified were fracture and thread shear. The FACCR further stated that the failure 63

could result in a loss of air during starting.

However, FPC did not list the fact that the failure could also result =in the EDG failing to start-because of~a loss of air.

The critical characteristics listed nn the FACCR were (1) vendor part number and (2) dimensions and configuration.

The critical characteristics were not adequate-to address the failure modes stated-for the part. _ As discussed for item (1) above, material and hardness were important critical characteristics that should have been verified, and the FACCR did not specify the dimensions that needed to be verified.

The source inspection requirements for this part were the same as those listed for item (1). The weaknesses found by the inspection team for item (1) also apply to this part.

In st.mmary, the more appropriate parent component was not stated. The critical characteristics stated on the FACCR were not adequate to address the failure modes identified for the part. The FACCR did not specify the dimensions that needed to be verified. Material was not identified as a critical characteristic and thus was not verified. The statement of conformance did not clearly indicate whether or not the source inspector verified the dimensions to the vendors drawings.

(3)

FPC issued PO F670378V, dated May 23, 1990, to Miller Bearing Inc. for

.l four sets of thrust bearings for the decay-heat removal pumps.

FPC determined after completion of this inspection that the installed bearing set is currently installed in a warehoused _ decay heat removal pump, and not in service.

The FACCR dated May 12, 1990, listed the critical characteristics of the thrust bearings for the-decay heat removal pump as configuration, dimensions, and manufacturer bearing number. However, attachment SE, of NP&SM Section 5, lists typical critical characteristics of a-thrust bearing as configuration, dimensions, model number, load rating, and material. The team concluded-that the FACCR critical characteristics were not specific enough to adequately determine the suitability of the item for use in its safety-related application.

The team also reviewed the RIP dated May 22,-1990. FPC uses the

. critical characteristics listed on the FACCR when developing the criteria to be listed on the RIP. The RIP listed the following items-to

-be verified: dimensions listed,on the. catalog sheet, damage and workmanship, and the manufacturer's part number.. However. the RIP did not-_ include configuration, which was included in the FACCR as a critical

. characteristic for this item, and which-was not -adequately ve'rified.

FPC did not verify the material type and load rating,-which, although important characteristics of a bearing, were not listed in either document.-

The RIP indicated that only two of the four sets should be sampled. The receiving inspection report indicated that two of the four. bearing sets were sampled per-MIL-STD-1050. Sampling is not appropriate in this L l

64

I situation since FPC had not surveyed the vendor and had no basis for the assumption of lot homogeneity, since Miller Bearings, Inc. was neither an approved distributor for MRC Bearings, nor an approved conwnercial grade supplier.

(4)

FPC issued PO F842352K, dated May 12, 1990, to Dresser Pump Division of Dresser Industries, Incorporated, for an impeller for a Worthington pump model 6-HND-134. FPC installed the pump in the building spray system.

The inspection team made the following observations while reviewing this dedication package.

The FACCR for this item was prepared on March 30, 1990, and approved March 31, 1990, by PE. The RIP and SIP were prepared by PQA on March 29, 1990, and approved by PQA on March 30, 1990, which was before the FACCR was approvad. This did not appear to be in accordance with Section 7.5 of the FP&SM, which states that PQA will develop an RIP and SIP using the critical characteristics provided on the FACCR.

The critical characteristics specified on the FACCR included configuration (double opposing), various dimensions, material (ASTM A-296), number of vanes, and documentation as specified in letter 838,.

The inspection teart noted that FPC listed the number of vanes as a critical characteristic, however, specified no value as being acceptable. The aoove critical characteristics were not adequately translated to the SIP. The SIP specified that the source inspector perform random dimensional inspections to verify conformance with vendor's provided drawings, bill of material, and shop order.

The SIP further specified that the number of vanes and the at-shaft inside diameter (both ends) be recorded. The critical characteristics verified and documented on the SIP were not adequate to ensure that the pump impeller would perform its intended safety function.

The completed SIP, associated inspection report and the supplier's provided documentation stated that the material was supplied per material specification ASTM A-744.

The inspection team observed that this material specification differed from that in the FACCR and SIP.

The source inspector did not document the material deviation, and PQA did not note the deviation while reviewing the documentation provided by the source inspector.

Both PE and PQA stated that the source inspector was not authorized to accept the change to the material specification.

The source inspector did not record the number of vanes as required by the SIP. There was no documentation to indicate whether the number of vanes were verified during receipt inspection.

Engineering Letter 838 required that the impeller be subjected to a liquid penetrant or magnetic particle test in accordance with ASME Section V.

However, the SIP did not require that the adequacy of the nondestructive testing be verified during performance. PQA personnel stated that their interpretation of this requirement was that FPC must verify the adequacy of the nondestructive testing by reviewing the documentation provided by the supplier. l l

65

PE and PQA personnel informed the inspection team that problem report SEPR-92-0002 was written March 17, 1992, to document that the SIP had not incorporated all of the specific critical characteristics stated on the FACCR.

The licensee found the problem while reviewing the completed dedication ' package, after the inspection team had identified the package for review during this inspection. A revised RIP was written and performed on March 17, 1992, which verified and recorded the number of vanes.

PQA personnel stated that all of the dimensions stated on the FACCR could not be verified at the site.

FPC'placed the impeller on QC hold while deciding whether or not to send it back to the supplier who has the proper facilities to verify the dimensions that could not be verified at the site.

PQA personnel also stated that material specification ASTM A-296 was discontinued in 1980 and was replaced by_

specifications A-743 and A-744.

(5)

FPC issued PC F842722K, dated September 20, 1990, to Coltec Industries for a lower impeller shaft key for the blower on the Colt EDG. The FACCR was dated May 5, 1990. Attachment 5F to NP&SM Section 5 lists the following critical characteristics typically verified for a shear key; configuration, dimensions,' hardness, and material. The FACCR, Section C, lists the following product characteristics critical to assure safety function:

vendor part number, configuration and dimensions.

The previously mentioned critical characteristics in the FACCR were not believed to be sufficient to assure that the item was suitable to perform its intended safety function. The FPC design engineer responsible for writing the FACCR stated that the critical characteristics suggested by FPC procedures and not included in the FACCR, were adequately verified on the basis of a recent vendor survey, and therefore were not included in his list of critical characteristics.

The RIP dated September 20, 1990, identified that the following characteristics needed to be verified:

that the material received agreed with the P0 description, that the material was free of shipping damage, that FPC received the documentation and that the source inspection was completed. The RIP was found to be deficient because it did not adequately specify the critical characteristics listed in the FACCR; part number and configuration and dimensions.

The step, " Verify that the material received agrees with the purchase order description,"

is an example of characteristics and acceptance criteria listed in the RIP that the team found to be too broad in character to ensure that the item specified in the P0 was suitable for service.

L The SIP dated September 13, 1990, was written to verify the item I

critical characteristics by the following methods: review part number and configuration, perform or witness dimensional inspection, and review the vendor's documentation for material to be supplied and determine acceptance per the material requirements. The SIP failed to list necessary methods for verification so that the source inspector could determine that the item was acceptable. There was no indication that the material was adequately verified. Additionally, the SIP was prepared and approved by the same PQA engineer and PE did not confirm that the characteristics listed as critical would be-properly assured. 66

(6)

FPC issued P0 F844359C, dated May 13, 1991, to Consolidated Power Supply, Birmingham, Alabama, for 32 square feet of ASTM A240,1/2 inch lype 316 stainless steel material. The FACCR dated April 15,1991, stated that the critical characteristics were material-(ASTM A240, Type 316).and-thickness (1/2 inch plus or minus 0.06 inches). The FACCR did not have any basis of purchase stated, however, it did state that the Type 316 plate would be used for structural non-pressure retaining applications, and its typical failure modes could be buckling _ or elastic failure, with a potential effect of a loss of structural integrity. The team noted that, other than a CoC, neither the FACCR nor the RIP identified or required certification from the material manufacturer, such as by a CMTR, or verification and documentation of traceability-such as by a heat number. The team also noted that-the RIP required the CR3 QC inspector to verify the material " marking" by using an

" inspection method" with " mechanical inspection. equipment," and to

" check material hardness,"

The hardness verification acceptance showed one hardness entry (78 HRB);

however, it could not be determined from reviewing the RIP whether the incoming material was tested in only one location, or whether the 78 HRB was the average. The team also reviewed a quality control issue (QCI) form, No. 144476, dated January 14, 1992, and noted that it transferred-or released several sizes of ASTM A240 plate, one of which was four square feet of 1/2 inch thick ASTM A240, type 316 material. The QCI line item for the 1/2 inch plate identified FPC P0 F844359C and heat / serial No. 849798. Based upon the records in package H-21, the team could not correlate the heat / serial number stated on the QCI to the actual material received on the RIP. Heat / serial No. 849798 was not found by the team in any other documents contained in the package other than the QCI.

(7)

FPC issued P0 F670407K, Revision 1, dated March 11, 1991,.to Anchor Darling Valve Company (AD), Williamsport, Pennsylvania, for AD check valve parts including a disc seat for 24 inch,150 pound (1b.) rated, swing check valves used-in the-raw water system (RW), valve numbers RWV-35, 36, and 38.

Revision 1 of this P0 required FPC~ source-inspection. Since the initial P0 did not contain any source inspection requirements, AD shipped the order to FPC without-the benefit of the performance of any FPC source inspection activities. Therefore, revision 1 of the PO was actually written to require the disc seat to be shipped back to AD, so that FPC could conduct source inspection activities. The critical characteristics listed on the FACCR were:

dimensions, configuration, and ASTM B127-4400 material. The FACCR basis of purchase was "like original: AD drawing W8422083 2/B." The RIP stated in part to verify completion of source inspection by a review of FPC SIP and verify that the material received agrees with PO description. The FPC-SIP, dated June 5, 1990, stated, in part, " verify the following critical characteristics;.. dimensional conformance by performing or witnessing... configuration is per vendor supplied documents [and) verify by review-of records that material is B127-4400."

This SIP was approved by the same PQA person who prepared it. The SIP copy attached to the Ebasco source inspection report contained the 67

source inspector's results, including "the material was verified by review of the CoC stating the material to be B127-4400, this was found acceptable."

The team noted during their review of this package that the critical characteristics, basis for purchase, and material were not adequately verified. Also, a like-for-like evaluation for equivalency back to the original P0 and drawing revision was not performed.. The team also noted that verifying commercial grade material only by a review of a CoC is inadequate.

(8)

FPC issued PO F8450350, dated October 28, 1991, to Epperson & Company for a Worcester Controls Corporation 3/4 inch,- 3-way ball-valve with actuator used as a main steam block valve during'a station blackout.

1his valve is a pressure boundary component for supplying air to the main steam valve actuators (411-414). The critical characteristics of the valve included materials of construction (brass body and stainless steel ball and stem of valve), part number, configuration, an(

accessories as specified in the P0; and the dimensions and thtead sizes as specified in the vendor's drawing.

FPC was to perform a source inspection to verify the critical characteristics listed on the SIP.

Completion of the SIP was to be verified during receipt inspection as identified on the receipt inspection plan datad September 18, 1991.

The SIP stated " Verify by review of documentation that material of construction is brass body, stainless steel ball and stem. Material traceability must be maintained from documentation to item." The SIP also staud that " Items supplied must be identical with those supplied as original equipment.

No changer, in the design, material, manufacturing or interchangeability is permitted without approval by FPC prior to shipment _ pcr letter (1046A)."

Before conducting the source inspot.lon, Ebasco (the contractor who performed the source inspection), learned that Worcester Controls had moved its manufacturirg facility from Massachusetts to North Carolina and had changed the design drawing number and part number for the valve.

Ebasco requested that, in lieu of a review of exact traceability of-material to specific heat or CMTR number, they conduct a general review of the vendor's material control process to ensure that a system is in.

place for maintaining material control while manufacturing valves. This I

change in the SIP guidance was accepted by FPC. The Ebasco Services ll Vendor Quality Assurance Report and the SIP used by_the source inspector for documenting the source inspection activities performed on December 12, 1991, did not contain adequate documented objective-evidence for the activities supposedly performed. Both documents only stated.that a review of the vendor's material. control program and certificate of conformance were reviewed for material verification and stated that "all items are as previously supplied" to verify that no design material or manufacturing changes had occurred. Neither the SIP (with inspector notes) or the final report provided any objective evidence as to what was reviewed during the source inspection at i

Worcester Controls. FPC did not specify which activities should be

! 68

performed to verify the vendor's material control program and ul_timately the material was not verified.

- (9)

FPC issued PO-F844057V,-dated February __25, 1991,'to Interscan Corporation' for a Sulfur Dioxide Gas-Sensor Cell for-control room a

atmosphere toxic-gas monitoring and alarm system.

The_ descriptions of the parent system and part safety functions were incomplete, indicating pressure retention only. No_ functional performance was required to be-verified, yet the item was designated Class-lE during purchase requisition review and on Engineering Letter 1176.- The part failure modes were incomplete; considered " leakage" (of. connection to system),

but not leakage of sensor water (which occurred shortly after first '_

sensor was installed requiring its immediate replacement).

The critical characteristics were inconsistent with the safety functions as _ listed _

Retention-of-sensing liquid and composition of the liquid were omitted. _

j The verification methods and acceptance criteria were inappropriate and i

incomplete.

The vendor was required by Engineering Letter 1176 1

(attached to the P0) to provide a CoC to the effect that the part had j

undergone no changes in design, material, or manufacturing process that would impact seismic capability since a previous PO had been issued in 1988 (original). Other critical characteristics were to be verified by comparison to original part.

Visual verification of configuration was reasonable, but visual for dimensions was questionable, and material by appearance was inadequate. Although not listed as a critical characteristic, its function was to be verified by calibration after -

installation.

Some critical characteristics were to be verified by l

source inspection. However, the SIP did not list weight, markings or l

what to verify by functional test; nor was adequate guidance given on I

how to verify dimensions, what dimensions, or what tolerance was acceptable. The RIP also did not list weight or markings.

The validity of the CoC was to be verified during a source inspection by Ebasco.

However, the source inspector documented reviewing only two drawings, a 1989 sensor QA dimension drawing, and a 1981 "LD monitor-layout" drawing. The S1 report was apparently not reviewed for adequacy by PQA or PE.

(10)

FPC issued P0 F842336V, dated May 4, 1990, to Multiamp lesting-Service, Inc., for Multtamp-terminal-blocks for various Class-lE applications.

The FACCR indicated " harsh" environment, but listed as restrictions were: no HELB/LOCA and Zone 30.

It was not clear what environmental qualifications were required. The parent and part safety functions, failure modes, and critical characteristics were incomplete or inappropriate, as were the verification methods and acceptance criteria.

Both _ insulation resistance and dielectric strength were listed with inconsistent acceptance criteria.

(11) FPC issued P0 F842798V, dated November 1,-1990, to-Torque-Quip, Inc.,

for 1 Burks Pump Company centrifugal pump and motor assembly and two spare motors to be used for (and spares for).EDG standby _ jacket cooling water pumps. The parent component was not named or described and its safety function was expressed as that of the entire EDG system which was too far removed _from the parts description to provide meaningful 69

information for determining the part safety function.

The part safety function was marked " active," but described as pressure boundary only.

lhe critical characteristics listed were configuration, material, and functional flow.

Seismic qualification was not listed as a critical characteristic but was addressed via a PEERE.

This was also an example of an unsatisfactory source inspection. The original source inspector did not recognize the configuration and markings of the initially supplied items as deviating from the requirements.

it was apparently not discovered that the mounting flange was incorrect and the motor was totally enclosed-fan cooled (TEFC) when it was supoosed to be totally enclosed-non ventilated (TENV), until installation was attempted.

(12) FPC issued P0 F844659K, dated July 30, 1991, to Graybar Electric Company, Tampa, Florida, for 20 Bussman 600 volt, 30-60 ampere, fuse reducers for standard size fuses. The P0 specified that the supplier shall provide a CoC attesting that the item or service provided on this PO was processed in accordance with the following QA/QC program:

Bussman's QA policy manual BU1500 CTD, dated August 9,1988. The critical characteristics were listed on the FACCR as part number and description. The FACCR's basis of purchase was stated as "like for like, Drawing No. 201-063, V8-08 thru V8-15."

FPC received a CoC from Bussman dated August 9, 1991, which certified that the items listed were manufactured and tested in accordance with Bussman's specifications.

The CoC identified FPC P0 F844659K ts the " reference customer P0 number."

The team noted that it appeared Graybar had procured the fuse reducers directly from Bussman; however, there was no objective evidence to substantiate that fact. CR3 had not performed a survey at Graybar to determine that they had adequate warehousing controls in place. The team questioned the Bussman CoC since it referenced FPC's P0 number instead of a Graybar P0 number. The team requested FPC to provide additional objective evidence to show traceability from Bussman to Graybar. However, FPC was only able to provide a packing list from Graybar to FPC, and was not able to show any other evidence of tri.ceability from Bussman to Graybar. The team concluded that CR3 had not established adequate traceability from the manufacturer to Graybar and did not show that the components Graybar procured directly from Bussman were the same components shipped to FPC.

(13) FPC issued P0 F8447190, dated August 12, 1991, to Chemco Electric Supply for Allen Bradley 700-N400Al convertible pole relays for ESF logic circuits.

The part safety function (marked as active) was described as maintaining integrity of safety-related control power supplies - relay contacts used for load status indication, but did not state whether the relays must energize or deenergize to perform this function. The f atlure modes listed short circuits, but not open circuits. Other failure modes (including failure to change / maintain required state) were stated not to affect operation of associated safety-related loads. The critical characteristics listed on the FACCR included coil resistance, dimensions and configuration, and dielectric strength of coil to ground, but did not consider mechanical load f actors, pull-in and dropout I 70

voltages, insulation resistance on contacts, contact 1 resistance, or timing / synchronization. The relays were to be qualified on a like-for-like basis with-the relays being replaced.-

(14) FPC issued PO F740240K, dated April 5,.1990, to Consolidated Electric Supply Co. for Joslyn-Clark (JCC) type T8137-16 type 4t!6-130, type SUK8-7-76, and 506-2-76 convertible pole relays for the safety-related equipment status indication panel and for ESF logic-initiation upon loss of offsite power. Safety function of part was stated as the description of conventional-relay operation, but never stated whether the relay must change state, and if so, whether they must energize or deenergize to perform their safety functions. Critical characteristics were stated-only as part number and configuration and pull-in and dropout testing.

Not listed were: seismic-qualification, insulation resistance of. coil-and contacts, contact resistance and timing. The verification of critical characteristics was to be by source inspection. The annotated SIP and source inspection report did not give load factors nor provide coil turns data as required.

(15) FPC issued P0 F844454K, dated June 6, 1991, to Consolidated Electric Supply for 12 electrical replacement coils for JCC PM type relays. The critical characteristics listed on the FACCR included: part number,

. ~

catalog description, configuration, voltage rating, and mechanical load range. Section D of the FACCR stated, " relay coil is like-for-like to that used in originally qualified equipment." The stated safety.

function of the parent component was to provide engineered safeguards switching functions. FPC Engineering Letter ll87A, dated September 11, 1990, stated, in part, "...each relay shall be subjected to-[certain tests for electrical relays]... The manufacturer shall assure and provide certification attesting that there has been no design,. material, or manufacturing process changes made since October 1971...."

The team noted that although P0 F844454K procured JCC's electrical relay replacement coils, Letter ll87A incorrectly imposed inspection requirements for relays. The team's review of.the Ebasco SIR also identified that, although.it was required to verify that the vendor's CoC was supported by records, the SIR did not document that any relevant-records were reviewed other than the vendor CoC and the coil resistance and turns test report. The team also noted that PQA's subsequent' review of the SIR failed to identify that the source inspectors failed to verify the vendor's CoC basis.

(16) FPC issued P0 F844090V, dated February 26, 1991, to Interscan Corporation, Chatsworth, California, for two-vacuum < switches for Interscan's Toxic Gas Monitor System used in the control room complex-ventilation system. Section D of the FACCR stated, " certification required per letter 1040 (like-for-like replacement)." Engineering Letter 1040, dated October 12, 1988, required Interscan to furnish a CoC stating that all parts supplied were equivalent or superior to, and a

interchangeable in form, fit, function and structural integrity with, the parts procured under FPC PO F906740lV. ~ The SIP, dated February 20, 1991, stated, in part, ".. 2. verify that material meets the 71

.-~ -~.-

requirements of the procurement documents by review of vendor's

records... 3.- verify configuration is per vendor drawing I-8213, dated April 1988,-[and] 4. verify by review of vendor documentation that certification contents are accurate and are supported by-records (letter 1040)...

To witness operability of switch prior to shipment. -The switch must be installed in a like-type monitor, or a mock-up that has the same functional characteristics as the-monitor.

The monitor or mock-up must perform in accordance with vendor technical documents with-the FPC ordered parts installed (VSI set-point is 50 inches H 0)."

2 During the team's review of the Ebasco SIR, dated March 12, 1991,.the team identified that not all of the SIP verification requirements were adequately performed or documented.

For example, even though SIP item no. 7 required that the parts be installed and tested in a functional mock-up, the SIR does not indicate that it was performed. The report from the source inspector also indicates that a June 28, 1988, revision of drawing I-8213 was used for verification activities, instead of the drawing revision that was stated in the SIP.

This March 12, 1991, SIR was reviewed by a PQA representative on April 3, 1991.

The team concludes that certain acceptance criteria identified in the SIP were not adequately addressed, performed and/or documented by the source inspection report.

The lack of adequate derivation of safety functions and critical characteristics, inadequate translation of those characteristics into l

verification methods and acceptance criteria, and the inadequate verification l

of those critical characteristics (and/or inadequate review to ensure proper verification) resulted in the numerous examples of inadequate dedication found by the team.

The inadequate dedications of tne CG!s discussed above, some of which were installed, constituted a failure by FPC to perform and document an adequate review for suitability of application, and in some cases, adequate design verification (seismic or EQ), for items intended for safety service, contrary to the requirements of Criterion III of 10 CFR Part 50, Appendix B.

-The inadequate dedications also constituted a failure to _ verify that the items -

received met the specifications for their safety-related applications contrary to the requirements of Criterion VII of 10 CFR Part 50, Appendix B.

Representative examples of inadequate dedications listed in Section 3 are cited as Deficiency 91-201-01.

4 PROCUREMENT AND DEDICATION TRAINING The training for personnel involved in the procurement, handling, storage and--

dedication of components for safety-related use was reviewed.

Since the personnel functionally report to different sections of the organization, the p

licensee has provided the following procedures to prescribe training:

Nuclear Engineering Procedure (NEP) 121, " Indoctrination and Training to Nuclear Engineering Operations and Procedures,",

l 72

Training Department Procedure (TDP) 311, " Nuclear Operations Training Procedure,"

~

' Nuclear Procurement and Storage Manual, Section 1.5.3, " Training," and Warehouse inspection Group (WlG)-2 Personnel Qualification Record.

4 NEP 121 pertains to the Nuclear Procurement Engineering Services (PE) personnel.

This includes senior electrical, instrumentation and control, and-mechanical engineers. The employee's supervisor _ performs an evaluation to-determine the required training using the guidance given in the procedure. -

Required training must be completed by the employee prior to performing work independently.

TDP 311 describes the training of the specialized certifications training program for personnel who perform audits, inspections, nondestructive examinations, and calibration activities affecting quality. The procedurc is detailed and thorough.

It is intended to satisfy commitments to regulatory.

guides and applicable ANSI standards. The procedure covers both initial training for each employee and continuing training or re-training, In addition special training and on-the-job training is provided.

Each employee is evaluated through examinations and grading. The inspector found the procedure to be satisfactory.

NP&SM Section 1.5.3 describes training requirements for the Nuclear Procurement and Storage Manual. All personnel involved in procurement activities are required to understand and follow the-instructions outlined within the manual. The Nuclear Procurement and Storage Committee chairman contacts each manager of each department involved with procurement activities to ensure that personnel are trained on the manual and each manual revision.

Technical training is also identified by the Committee when needed. Records

- of attendence are maintained by the Training and-Records Management Department.

The final procedure, WIG-2, provides guidance and instructions for warehouse personnel.

Basically,-this procedure allows the Supervisor of Nuclear Stores to evaluate the training required for warehouse personnel, such as storekeepers and assistant storekeepers. The Supervisor-is responsible for maintaining the employee. qualified and may schedule re-training when determined necessary. A qualification record is maintained for each emp^1oyee.

Several records for storekeepers were reviewed by the inspection team and.

found to be zcomplete.

It-would appear that-the. training program for warehouse-i personnel-is satisfactory.

The training program was evaluated by examining training recordt. for several.

people who perform different parts of the procurement activity. The records were found to be complete and up-to-date. On February 24-28, 1992, special training entitled, " Nuclear Utility Procurement," was presented at CR3..The ~

course was presented by an EPRI NDE representative for 30 FPC people. The course content. material was reviewed by the inspection team. The course appeared to provide training in.the pertinent areas'of procurement and dedication. Licensee management stated that it is their intention to add this 73

special training to the regular training program for new employees and on-going training when appropriate. This action will be considered a strength in the training program when it is fully implemented, 5

EXIT MEETING On March 27, 1992, the inspection team conducted an exit meeting with members of the FPC staff and management at the CR3 site. During the exit meeting the team summarized the inspection findings and observations.

The following individuals were present, flgILdLftwxr CtrngrA11qn P. Beard, Senior Vice President, Nuclear Operations G. Boldt, Vice President, Nuclear Production B. Hickle, Director. Quality Programs P. Tanguay, Director, Nuclear Operations Engineering and Projects W. Conklin, Director, Nuclear Operations Materials and Controls R. Widell, Director, Nuclear Operations Site Support G. Oberndorfer, Manager, Procurement and Material Quality Assurance (QA)

E. Welch, Manager, Nuclear Procurement Engineering Services (NPES)

E. Froa'.s, Manager, Nucicar Compliance K. Wilson, Manager, Nuclear Licensing W. Watts, Manager, Purchasing and Contracting D. Kurtz, Manager, Site Nuclear QA K. Gardner, Manager, Material Control J. Colby, Acting Manager, NPES G. Becker, Manager, Site Nuclear Engineering Services (SNES)

A. Gelston, Acting Manager, SNES Santilli, Supervisor, Materials Quality Control t.

R. Yost, Supervisor, Quality Audits D. Bates, Supervisor, Quality Systems J. Buckner, Nuclear Regulatory Specialist T. Catchpole, Senior Nuclear QA Specialist

[M ltAC EtaHlal.ory Commissign B. Grimes, Director, Division of Reactor Inspection and Safeguards E. Merschoff, Deputy Director, Division of Reactor Safety, Region 11 U. Potapovs, Section Chief, VIB R. McIntyre, Team Leader, VIB 5, Alexander, EQ and Test Engineer, VIB W. Gleaves, Reactor Engineer, VIB J. Petrosino, QA Specialist, VIB M. Thomas, Reactor inspector, Region 11 F. Jape Section Chief, Region 11 P. Holmes-Ray, Senior Resident inspector, CR3

, 74

Other Personnel.

-.B !Bradley,:-Senior:-Project Manager, NUMARC P. Robinson.-Attorney _Winston and Strawn i

umu 75

_-_m-_.._mmm_2_ioum._-_

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

l f ucy&.

  1. c' UNITED STATES
  • k,, M '

i NUCLEAR REGULATORY COMMISSION i

I

-8 W ASHINGToN. D.C. 2245 c '.....,e AUG 2 41932 Docket No. 99901251 Mr. Mitchell S.

Bregman Vice President and General Manager GNB Industrial Battery Company Woodlake Corporate park 829 Parkview Boulevard Lombard, Illinois 60148-3249

Dear Mr. Bregman:

SUBJECT:

NRC INSPECTION REPORT 99901251/92-01 This letter addresses the inspection of your facility at Fort Smith, Arkansas, conducted by Messrs. R. C. Wilson and R. H. Moist of this office on August 4-6, 1992, and the discussion of their findings with members of your staff on August 6, 1992.

The purpose of the inspection was to investigate recent 10 CFR Part 21 reports that published one-minute load ratinge for safety-related batteries may be nonconservative.

The inspectors reviewed your bases for establishing original and revised battery ratings, and selectively reviewed the implementation of your quality assurance program for supplying Class 1E batteries for safety-related applications.

Areas examined during the NRC inspection and our findings are discussed in the enclosed report.

This inspection consisted of an examination of procedures and records, interviews with per-sonnel, and observations by the inspectors.

The inspection determined that published one-minute ratings for GNB type NCX batteries do not take into account a transient voltage dip effect known as the " Coup de Fouet," which can reduce the one-minute rating by 10 percent or more.

We believe that you have acted responsibly to investigate and report the problem.

During the inspection you committed to provide additional infor-mation to the industry, based on ongoing testing, by August 31, 1992.

I ask that you send copies of that information to the NRC Document control Desk and to me.

The inspection also determined that your procedures no longer required GNB to perform 10 CFR Part 21 evaluations and reports for batteries supplied as safety-grade prior to July, 1990.

You have committed to promptly revise your procedures to address that concern.

76

Mr. Mitchell S.

Bregman i In accordance with 10 CFR Part 2.790 of the.NRC's " Rules of i

Practice," a copy of this letter and its enclosures vill be placed in the NRC's Public Document Room.

Sincerely,.'

]

,a c~

}

/ Leif rrholm, Chief Vendor Inspection Branch Division of Reactor Inspection and Safeguards office of Nuclear Reactor Regulation

Enclosure:

Inspection Report 99901251/92-01 I

77

ORG A!112 ATIOll:

G!la INDUSTRIAL BATTERY COMPA!JY FORT SMITil, ARKAllSAS REPORT 110.

99901251/92-01 CORRESPONDEllCE Mr. Mitchell S.

Bregman ADDRESS:

Vice President and General Manager GilB Industrial Battery Company Woodlake Corporate Park 829 Parkview Boulevard Lombard, Illinois 60148-3249 ORGA!112ATIC

'l David M.

Lewis, Quality Assurance Manager CO!1 TACT:

501/646-8341 11UCLEAd IllDUETRY Major supplier of lead-acid station batteries ACT]VITY:

for commercial nuclear power plants IllSPECTIOli August 4-6, 1992 CO!! DUCTED:

20.92--

SIGt1ED:

Richard C.

Wilson, Senior Engineer

~ Date Reactive Inspection Section No. 2 Vendor Inspection Branch (VIB)

OTi!ER I!1SPECTOR:

Randolph 11. Moist, VIB APPROVED:

M1en.,

e f# L "Gregoff f. CWalina, Chief

' Date 7

Reactive Inspection Section No. 2 Vendor Inspection Branch IllSPECTION BASES:

10 CFR Part 21 and 10 CFR Part 50, appendix B I!1SPECTIO!1 SCOPE:

To review manufacturing and testing activ-ities affecting short-term battery ratings PLA11T SITE Numerous APPLICABILITY:

l 1

78

l 1

INSPECTION

SUMMARY

1.1 ppen Ittaa GND Industrial Battory Co. (GilB) committod to notify all nuclear safety-related customers of the Coup de Fouet-effect, and to issue and post a 10 CPR Part 21 reporting procedure for battories previously shipped as safety-related.

Nuclear Logistics Inc.

(NLI) committed to supplement its July 8, 1992, 10 CPR Part 21 report to cover additional battory types.

These actions are to be completed by August 31, 1992.

The NRC will monitor completion of those activities.

(See Section 3.8 of this report.)

2 STATUS OF PREVIOUS IllSPECTIOli FINDI!1GS There woro no previous NRC inspections of this facility.

3 li4SPECTIoll FI!1DIliGS AND OTHER COMME!1TS 3.1 Entrance and Exit Meetinos In the entranco meeting on August 4, 1992, the NRC inspectors discussed the scope of the inspection, outlined areas of concern, and established interfaces with G!lB's management and staff. -In the exit meeting on August 6, 1992, the inspectors discussed the findings and concerns with Gl1B's management and staff.

3.2 Innoection Scopo GNB, formerly named Gould, is a loading manufacturer of largo batteries.

The corporate headquarters is in Lombard, Illinois.

The first of the present design Gould/GNB buttories.for commer-cial nuclear power plants woro manufactured in Trontoli, Now.

Jersey,-and Kankakoo, Illinois, in the early 1970s..The Fort Smith, Arkansas, location has manufactured batteries for nuclear plants and for other applications such as telephone systems since 1974.

The Fort Smith plant has over 100,000 square feet and employs more than 100 persons.

Roughly five percent of the production is lead-acid batteries for nuclear plants.

GNB= maintained a quality assurance program conforming to Appen-dix B to 10 CFR-Part 50 frcm 1973 until June 20, 1990, when production-was converted.to commercial grado.

On August 8,

1991, GNB contracted to supply Class 1E batteries through HLI.

Two notifications under 10 CFR Part.21 have been submitted to the NRC concerning the-ability of GNB batteries--to meet published-one-minute load ratings:

by Commonwealth Edison Co. (CECO) for the Dresden Unit 3 nuclear power plant on May 22,-1992, and by NLI (for the Waterford nuclear power station) on July 8, 1992.

2 79

The NRC inspectors concentrated on reviewing the basis for GNB's published load ratings, and the tests being performed to estab-lish new one-minute ratings.

Manufacturing activities, previous and present quality controls, and reporting activities related to 10 CFR Part 21 were also reviewed as related to load ratings.

3.3 U1g_C_oup_dc Fouet ELinct The Coup de Fouet (crack of the whip) effect in lead-acid storage batteries has boon known for years; when the NRC laspectors asked for a description, GNB provided excerpts from a 1977 textbook.

Briefly, the effect involves a transient decrease in aid concen-tration at the cell plates, which results in a temporary voltage dip.

Although theoretically known, the Coup de Fouet offect apparently has not been observed in large stationary batteries until the one minute load test performed at CECO's Dresden 3 plant in October, 1991.

Load rating tests are conducted in accordance with IEEE Standard 450, "IEEE Recommended Practice for Maintenance, Testing, and Replacement of Large Lead Storage Batteries 'or Generating Stations and Sub-stations" (several revisions)

One-minute load rating currents are typically about 1000 amps or more.

The NRC inspectors observed data at GHB indicating that the transient voltage dip minimizes about 10 seconds after the cell is loaded, and disappears within less than 20 seconds.

The effect has not been observed in practice because (1) IEEE 450 only requires one voltage measurement during the rating period, and it specifies that the rated cutoff voltage must only be met at the_end of the rating period; (2) most licensee purchase orders have only specified long duration acceptance tests, where the test current is too smal) to produce a significant effect; and (3) measurements have usually been made with voltmeters that had to be reconnected from one cell to another and could not provide frequent measurements.

The 1991 CECO data and the new GNB/HLI data were obtained with data loggers capable of nearly continuous multiple point logging.

3.4 Ilards f or P@lished PerLQrmance Ratincts GNB conducted their first tests to determine nuclear battery ratings in 1971-2 on type NCX-2550 batteries, now called type NCX-35.

(GNB changed the part number designations in 1986 from the nominal amp-hour capacity for an eight hour limit of 1.75 volts per cell, to the number of plates per cell, to conform to industry practice.)

GNB conducted numerous discharge tests at various rates to generate a family of curves.

GNB personnel stated that the earliest load test data were taken at 0, 5,

30, and 60 seconds.

They stated that the o second measurement was open circuit, and that a slight voltage increase was noted from 5 to 30 seconds; however, the cell average always exceeded the 1.75 volt rating.

3 80

4 i

GND selected the NCX-2550 battery for initial testing because it had the least acid per plato, and would tend to produce the lowest long-duration capacity.

Ratings established for the NCX-2550 for various times, were extrapolated to other battery types by ratioing the number of positive plates por coll.

The 1971-2 published ratings were changed in 1986-8, when further testing shoved that the extrapolated long-term capacition woro genera]1y somewhat low.

The original ono-minute ratings have never been changed.

3.5 Licens2e Purchase Orders Licensco procurements have typically required verification of eight hour ratings for production cells, resulting in relatively low-current acceptance tests with few early data points.

For example, CECO Purchase Order (PO) No. 305664 dated March 26, 1986, _ procured batteries for the Dresden and Quad Cities nuclear power stations.

Sargent & Lundy Specification No. T-3349 Addon-dum 1 dated August 7, 1985, required an eight-hour acceptance test with the individual cell voltages at the end of eight hours to exceed 1.81 volts.

Hourly data were specified.

The NRC inspectors reviewed the test proceduro and data shoots for batteries shipped to the Quad Cities station.

GND test pro-cedure QAI-2004-S, Revision 10, dated July 31, 1986, required the first voltage readings 15 and 60 minutes after the start of. dis-charge. The first two data points on the August 6, 1986 GNB data shoots are at 15 and 60 minutos, and the current was 187 amps for the NCX-1500 cells.

The Coup de Fouet effect was not evident in the procurement acceptance tests because of the low current and' the infrequent data recording.

The published:one minuto current rating for the NCX-1500 cells at 2.81 volts was 1264 amps.

The NRC inspectors reviewed records for two ot'er Pos for whichi no data were taken during the fjrst minute._ Houston Power and Light Co. PO No. 14926-DF-978 dated April 28,-1983, specified an eight hour acceptance test.

Baltimore Gas and Electric Co. PO No. 32483-CX dated February 26, 1985, for the Calvert Cliffs nuclear power plant specified a capacity rating test and a two-hori load profile test.

'n ruviewing records for Toledo Edison Co. PO No. 056-Q096822A-JA tea te 6 May 6, 1986, the inspectors found.that one-minute and one-

~

ho a acceptance tests were required forfsix cells of each of four batteries, in-addition-to an eight-hour IEEE 450 test.

Test logs for batteries shipped to the Davis Besso nuclear power station showed digital _ voltmeter data taken at 0.2, 0.4, 0.6, 0.8,1and 1.0 minutes on-June 24, 1986, and o' O.3 0.8, and 1.0 minutes.on July 16, 1986.

Most of these dat..now a slight = voltage increase (10 millivolts or-less) between the first and second data points, 4

81

possibly suggesting the coup de Fouet effect.

These tests were conducted on type NCX-1500 cells at about 1500 amps.

3.6 Testina to Establish New Ratinas for NCX Cells Because of the CECO findings at Dresden, GNB performed a dis-charge test on five type NCX-21 cel)s at their Kankakee, Illinoin plant on March 13, 1992.

The test was performed in accordance with GND Specification TM-16-18-FST, Instructions for Prepara-tion, Test, and Documentation of Capacity and Short-Circuit Characteristics of a Stationary, Flooded String Battery NCX-21 cells," Revision - new, dated February 24, 1992.

GNB discharge-tested the cells per IEEE 450 at 1264 umps; voltage was measured 1

with an oscilloscope.

A CECO representative witnessed the test.

The low voltage point occurred at 10.40 seconds into the test; it was below the minimum acceptable value for one minute.

Based on those tests, NLI initiated a testing program at GNB's Fort Smith facilities to develop families of curves that will be used to define new one-minute ratings that take into account the coup de Fouet offect.

Approximately 89 tests had been performed for the following cell types:

NCX-17,

-21,

-27,

-33, and

-35.-

The inspectors reviewed documentation including the test pro-cedure, test equipment calibration status, test log sheets, and component identification.

The test procedure was NLI procedure NLI-PROC-02, " Inspection of GNB Activities," Revision 2, dated April 15, 1992, which referenced the IEEE 450 discharge test and the GNB Quality Assurance Instruction QAI-6020, " Stationary Battery Acceptance. Testing - High-Rate & Discharges Greater Than Two Hours - Lead-Acid Ret Cells," Revision 0, dated November,15, 1989.

The data acquisition equipment (Nicolet Model 500 data logger and Toshiba Model T3200 laptop computer) could monitor up to 20 channels in 30 millivolt increments.

The NRC inspectors witnessed NLI testing of three NCX-33 cells.

The test was conducted at a constant current of 1681.5 amperes (temperature-corrected from 1700 amps) until voltages of 1.78 volts per cell and 5.34 volts for the_ string were reached. -The current was maintained constant with a calibrated 0-3000 amp-shunt _with an accuracy of 1 percent.

The Coup de Fouet effect minimized at 9 seconds when the string voltage dipped to 5.3728 volts.

The. string voltage did not drop below 5.34 volts until-i 12.69 minutes, and the test was terminated at 15 minutes.

These data will be combined with others to develop new'_ ratings.-

At the time of the inspection, rating reductions of about 9 I

- percent to a cutoff voltage of 1.75, and 13 percent to 1.81 volts l

-appeared appropriate.

GNB personnel felt that thesi changes would have little impact on plants that=had sized their station batteries in accordance with IEEE Standard 485, " Recommended-I o

5 Z

82

.E,

.--.ym

,,--_.,.,-.,,,,,,ni,r.

m

..-..-,_,,,,g_,ne_.r_._-,

.-m.

__.r.._w,,,,,

.,y,.,,,.,y,, _,,

,,g,g-y-.

Practico for Sizing Large Lead Storage Batteries for' Generating Stations and Substations" (soveral revisions).

3.7 QualityJ aggranco Proarams When GNB converted its quality assuranco (QA) program from safety-grado to commercial grado in June 1990, the major changos made were in tho-organizational structuro and the internal audit program.

NLI performed two detailed audits to determino-what activities had to be added to GHB's implementation of their commercial grado QA program to provido safety-grade battories.

Although the emphasis of the NRC inspection was placed on the Coup de Fount offect, the inspectors selectively reviewed the implementation of GHB'r idt cresent QA programs, as well as m a i

the prosent GND-NLI ar12%4.'

The NRC inspectors solet: 991) c vicsod data packages for typo NCX-1500 battories supplie;.

eEC9+s Quad Cities Unit'l and 4

Baltimore Gas and Electric's calvert Cliffs nuclear power plants in 1985-6.

The-inspectors reviewed the requiroments that GNB i

imposed in vendor Pos (typically Appendix B-to 10 CFR Part.50, MIL-I-45200 for suppliers, and.HIL-STD-45602 for Calibration Services); receiving inspection records; supplier cortificates of analysis; GHB's analyses of critical raw products-(calcium, lead, and sulfuric acid); shop travel cards; and inspection records for casting, pasting, assembly, acid fill, formation (testing) and final inspection.

The NRC inspectors also selectively reviewed the GND Quality Program Manual, Revision 9, dated August 30,:1985 (the last safety-grado revision), and selected QA. Instructions.

B The JUtc inspectors toured several areas of GHB's plant including L

receiving inspection, manufacturing processes, assembly and. test-ing activities, and. final inspection to determine what controls l

woro in_placo-to support the current _ production _of batteries.

Assuming that similar manufacturing _and testing activities took-place in the past, the NRC inspectors saw no_ deficiencies in i

GHB's' implementation of its previous Appendix B QA program.

The NRC inspectors briefly. reviewed savoral GNB.neismic and environmental qualification reports, and several NLI documented reviews of GNB activities.

The-100; inspectors also discussed with GNB-and FLI representatives the current program for supply-ing Class 1E batteries. The approach is to superimpose NLI activities upon GNB activities so that the combination satisfies 10 CFR Part 50,-Appendix B, critoria.

NLI works so closvly with GNB-that the offect is loss like a third-party dedication than a-cooperativo offot. producing Appendix B level batteries.

For example, deviations and-ongineering change orders affecting nuclear batteries requiro NLI-approval,_and the final acceptance testing is performed in GNB's plant.

Conversely, NLI had not documented the normal _ dedication 6

.83.

x

..n,

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


e

activition of listing cafoty functions, deriving critical charactorintico from them, and identifying the verification method for each critical charactoristic.

Based on the reviews doccribed above, the NRC inopoctors found no indication that the batterion are not nafety-grado.

llowever, the NRC may conduct a futuro inspection at NL1's Forth Worth, Texas, facility to further review NLI's dedication and Part 21 rnporting activities.

3.8 10.CPR Part 21 Reparin Based on the March 1992 F.ankakoo tanting performed in ropponco to the CECO concern, on March 30, 1992, GHB notified all four nucicar utilitico using typo NCX-21 battorios in safety-ralated

-applications that the ono-minuto rating should be reduced from 1264 to 1050 amps for an and voltage of 1.81.

Daned on the ongoing testing at GND Fort Smith, NLI submitted a 10 CFR Part 21 notification that the one-minuto rating for type NCX-17 battorion supplied to the Waterford nuclear power plant should be reduced from 1306 to 1165 amps for an end voltago of 1.75.

Most nuclear-cafety-related GNB battorien are type NCX-17.

During thin inapoction, GND and NLI paraonnel committed to perform the following actions by August 31, 19921 (1) GND will advino all customers using GNU batterion in nuclear safety-related applications of the coup de Fouet of fect, and provido now ono-minuto ratings data an available.

(2) NLI will supplomont their 10 CPR Part 21 report dated July 8, 1992, to cover additional battery types.

The NRC inspectora datormined that GNB's previous Appendix B QA program, and NL1'n current QA program, both includo proceduron for conforming to the requirements of 10 CPR Part 21, including instructions and requirements for identification, control, and documentation of nonconformancon, llowever, CNB's present commer-cial grado QA program does not provide for continued reporting-responsibility for battorios that were supplied by GND as safouy-grado.

GND personnel agrood to issue and post an appropriato proceduro by August 31, 1992, and to address that proceduro in the next revision of the Quality Program Manual.

The NRC inspectors concluded that with thoso actions GND'and NLI are adequately addronning the notification requiremonta of 10 CFR Part 21 with respect to the Coup do Fouet offect, including both Part 21 notifications submittod to dato.

7 84 t

___._1___

t 4

PERSOllllEL CollTACTED

[

f GNB Lombard1 M. S.:-Drogman, VP and Gen. Mgr., Stationary Products J. J. Jorg1, Vice Pres., Engincoring.and R&D i

+

D. M. Fischer, Director of Quality Assurance GND FJ;trt Smith:

R. McReynolds, Plant Manager

+

D. M. Lewis, Quality Assuranco Manager D. Bateman, Quality Ascurance Coordinator likl:

+

A.

C.

Bol), Director, Nuclear Sparc Parts F. J. Mcdonald, Manager, Materials Equipment ~Engrg Attended the entrance meeting on August 4, 1992

+

Attended the exit mooting on August 6, 1992 F

1 4

8' 85-a

%rv--

y'Wr-er rm',Wway y=--%,www.yg<

wir-y w

yge-W.T y

7 og <W ag 9yw9p'my-emwyT"'-1

-T'--FT"t-V$

-- vr g P!4 m 9 P-g -

T--4-gae p,e w q

in*<t-fweg-19--

bgy-prt-'+er 9

tre y weedag 9

[pasoso og Cfi

'n UNITED STATES

! k'gg jf

- NUCLEAR REGULATORY COMMISSION

';1, 2J W ASHINoToN, o C. 20Z6

% [.. *#

AUG 141992 Docket No. 99900404 Mr. C. L. Caso, General Manager Nuclear and Advance Technology Division-Energy Systems Business Unit Westinghouse Electric Corporation Post Office Box 355 Pittsburgh, Pennsylvania 15230

SUBJECT:

NOTICE Of NONCONf0RMANCE (NRC INSPECTION REPORT 99900404/92-01)

Dear Mr. Caso:

This letter addresses the July 7-9, 1992, inspection of the Nuclear and Advanced Technology Division of the Westinghouse Electric Corporation's Energy Systems Business Unit in Monroeville, Pennsylvania. The ins)ection focused on motor operated valve data supplied to nuclear utilities by tie Mechanical

)

Equiament Design group. The inspection was conducted by Mr. Jeffrey B.

Jaco) son of this office as well as other members of the NRC staff. The inspection findings were discussed at the conclusion of the inspection with the Westinghouse representatives identified in the enclosed report.

Areas examined during the NRC inspection and our findings are discussed in the enclosed report.

This inspection consisted of an examination of procedures and records, interviews with personnel, and observations by the inspectors.

The inspection identified that your Quality Assurance (QA) program failed to meet certain NRC requirements. Specifically, Westinghouse failed to justify-the upward rounding of motor actuator stall thrust data supplied to New Hampshire Yankee for the Seabrook Nuclear Plant. The specific findings and references to the pertinent requirements for the above nonconformance are identified in the enclosed Notice of Nonconformance.

We are also concerned that similar stall thrust data supplied to numerous other utilities has not been appropriately validated for plant specific use.

This data was based on testing performed for Westinghouse by Limitorque which does not adequately replicate actual plant conditions. Consequently, this uata appears not adequate as a basis for establishing operability of a motor operated valve.

In addition, we are also concerned that Westinghouse draft qealification reports were being used as a basis for operability of motor operated valves by at least one utility. These reports attempt to extend the thrust ratings of the Limitorque actuators to ratings that exceed-the manufacturer's (Limitorque) specifications. During this inspection the team identified numerous weaknesses that would restrict the applicability of the reports.

86

Mr. C.

L. Caso In addition to the reply requested by the attached Notice of Nonconformance please indicate whether all users to whom Westinghouse supplied stall thrust data have been notified of the failure of the data t's represent actual plant considitions.

The response requested by the enclosed Notice is not subject to the clearance procedures of the Office of Management and Budget as required by the Paperwork Reduction Act of 1980, Public Law No.96-511.

In accordance with 10 CFR Part 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosures will be placed in the NRC's Public Document Room.

Sincerely, l

,,A/E G Mj!.

k l

7.W4Vs

/l /

eif f[orrholm, Chief L

j Vendor inspection Branch a

Division of Reactor Nspection and Safeguards Office of Nuclear Raactor Regulation

Enclosures:

1. Notice of Nonconformance
2. Inspection Report 99900404/92-01 1

87 4

p NOTICE Of NONCONf0RMANCE Wl511NGil0VSE [LLClRIC CORPORA 110N Docket No. 99900404/92-01 Pittsburgh, Pennsylvania During a U.S. Nuclear Regulatory Commission (NRC) inspection conducted at the Westinghouse Electric Corporation in Pittsburgh, Pennsylvania, on July 7-9, 1992, the NRC inspection team identified that certain of your activities were not conducted in accordance with NRC requirements that were contractually imposed upon Westinghouse by purchase orders f rom NRC licensees. The NRC has classified these itens as a nonconformance to the requirements of litle 10 of the (pde_qf fesixnLEcgyh11tu, Part 50 (10 CFR Part 50), Appendix B.

Criterion VI, " Document Control," of 10 CFR Part 50, Appendix B, requires that measures be established to control the issuance of documents, such as instructions, procedures, and drawings which prescribe all activities affecting quality. Criterion XI, "lest Control," of 10 CFR Part 50, Appendix B, requires that test results be documented and evaluated to assure that test requirements have been satisfied, Contrary to the above, Westinghouse supplied stall thrust data to New Hampshire Yankee in Westinghouse letter NAH-3219 which was inadequately evaluated and may also have been supplied to other users.

The data was taken from testing performed for Westinghouse by Limitorque and was rounded up in a nonconservative manner without justification.

(92-01-01)

Please provide a written statement or explanation to the U.S. Nuclear Regulatory Commission, A11N: Document Control Desk, Washington, D.C.

20555 with a copy to the Chief, Vendor Inspection Branch, Division of Reactor Inspection and Safeguards, Office of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting this Notice of Nonconformance, lhis reply should be clearly marked as a " Reply to a Notice of Nonconformance" and should include for each nonconformance: (1) a description of steps that have been or will be taken to correct these items; (2) a description of steps that have been or will be taken to prevent recurrence; and (3) the dates your corrective actions and preventive measures were or will be completed.

Dated at Rockville, Maryland this /y day of At j s (/1992 1

88 INSPECTION REPORT U.S. NUCLEAR REGULATORY COMMISSION Off!CE OF NUCLEAR REACTOR REGULATION DIVISION OF REACTOR INSPECTION AND SAFEGUARDS Report No.:

99900404/92-01 Docket No.:

99900404

(

Company:

Westinghouse Electric Corporation Post Office Box 355 Pittsburgh, Pennsylvania 15230 Organizational

Contact:

Wayne Walker 412-374-6471 Industry Activity:

Nuclear Steam Supply System Components and Services inspection Conducted:

July 7-9, 1992 Inspection Team:

Jeffrey B. Jacobson, Team Leader Thomas Scarbrough, NRR P. K. Eapen, Region 1 Michael Runyan, Region IV Prepared by:

4-/ b

. N/c/91 JeffreyA.,4ap6bson, Acting Section Chief Date-Reactive in yection Section No. 2 Vendor Inspection Branch Approved by:

i [4L L61fJ.INor/holm, Chief Date Vendor inspection Branch inspection Bases:

10 CFR Part 21 and 10 CFR Part 50, p

Appendix B l

Inspection Scope:

To review Westinghouse's program for upgrading the thrust ratings of Limitorque motor actuators and to review the basis for motor actuator stall thrust _ data supplied by Westinghouse.

Plants Affected:

Numerous 89

k f;y

f CONTENTS I

EASE 1

INSPECTION

SUMMARY

I 1.1 Nonconformances l'

l.1.1 Nonconformance 92-01-01 I

1.2 Unresolved items.......................

1 2 STATUS OF PREVIOUS INSPECTION FINDINGS................

1

'3 INSPECTION FINDINGS AND OTHER COMMENTS.................

1 3.1 Stall Thrust Data Review......... -..........

l>

3.2 Applicability to'Seabrook.................. -

3 3.3 Actuator Extended Thrust Rating Program............

3 3.3.1 SMB-00 Actuator Testing 6

3.3.2 SMB-2 Actuator Testing.................

7 3.3.3 SB-00 Actuator Testing..................

8 3.3.4 Conclusions 8

3.4 Applicability to Beaver Valley...............-.

9 4 EXIT MEETING.............................

9 90

/

1 INSPECTION-

SUMMARY

1.1 Nonconformances 1.1.1 Nonconformance 92-01-01 Contrary to Criterion VI and Criterion XI of 10 CFR Part 50, Appendix 0, Westinghouse supplied stall thrust data to New Hampshire Yankee for the Seabrook Nuclear Plant which was inadequately evaluated and may have supplied imilar data to other utilities. The data was taken from testing performed for Westinghouse by Limitorque and was rounded up in a nonconservative manner by Westinghouse without justification.

(see Section 3.2 of this report) 1.2 Unresolved items None 2 STATUS OF PREVIOUS INSPECTION FINDINGS Previous findings were not reviewed during this inspection.

3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 Stall Thrust Data Review Westinghouse has supplied numerous motor operated valves (MOVs) to various utilities throughout the last 20 years. Westinghouse procured the motor-operated actuators for its valves from Limitorque, Inc.

Prior to shipment from Limitorque, each actuator was mounted on a thrust test stand and subjected to a performance test. The test method was described in Limitorque drawing number 21-403-0003-2, Revision B, " Westinghouse (WEMD) Test Procedure Per Westinghouse Spec 565682, Revision H."

Each actuator was tested against a load cell at the maximum torque switch setting (per the limiter plate), the minimum setting, a middle setting, and with the torque switch bypassed (stall condition). Motor voltage, current, stem thrust, the torque switch drop out point, and the "SB" deflection (where applicable) were measured for each test.

This information was used by Westinghouse to establish a relationship of the torque switch setting versus delivered stem thrust and to determine-the 80 percent degraded voltage stall thrust capability of the motor actuator. The stall tests were generally performed at 100 percent voltage after which the resulting data were adjusted by calculation to estimate the 80 percent voltage capability.

If the calculated 80 percent stall thrust-capability was not considered acceptable,.the actuator was retested at 80 percent voltage.

Westinghouse typically supplied tiiis stall thrust data to utilities along with the hardware and has typically reissued such data to utilities when requested.

The NRC has learned that some utilities are relying on this stall thrust data as a basis-for establishing operability for those valves where standard

_ industry equations-show the motor operator incapable of stroking the valve.

1 91

~

1he team reviewed the above referenced test procedure (and the Westinghouse specification referenced in that procedure) and identified several concerns regarding the applicability of data used to establish the stall thrust capability of the motor actuators, lhe greatest of these concerns was that the stall tests were performed under a static condition (i.e., no differential 4

pressure or flow against the valve disc), but the results were being used in Westinghouse-supplied calculations _to demonstrate actuator capability under full-flow design basis dif ferential pressure (d/p) conditions. Westinghouse-had used this data in part to demonstrate that at 80 percent voltage the operator would be capable of delivering the design thrust required to position its asscciated valve prior to reaching a stall condition. However, a growing body of industry testing has shown that the thrust delivered under dynamic conditions may be less than that delivered under static conditions.

This load sensitive behavior or " rate-of-loading" of fcct is not fully understood in a physical sense but has been measured to be as high as 30 to 40 percent under certain high d/p conditions.

Therefore, the recorded stall thrust values usc=d in Westinghouse valvo calculations may not reflect the actual stall thrust available under design basis conditions.

Another consideration, which limited the apparent applicability of the stall test results, was that the lubrication conditions of the test stems used in the timitorque test stand were not documented and were not known.

Limitorque informed Westinghouse that the test stems may or may not have been lubricated and that the lubricant type, where used, was not known. Knowledge of the test stem lubrication conditions would be necessary to determine whether the stall thrust measured in the test stand could be achieved when the actuator was installed in the plant, If more desirable lubrication conditions existed-during the test, the decrease in available stall thrust under actual in-plant conditions (with aged and temperature-affected lubricant) could be significant. Torque was not measured during the stall tests and therefore an estimation of the test stem coefficient of friction cannot be made.

A third consideration affecting the appilcability of the stall test data was the accuracy involved in measuring the stem thrust.

The test results supplied to Westinghouse and subsequently used in the MOV calculations were raw data, uncorrected _ for measurement _ accuracy.

Limitorque informed Westinghouse that the intrinsic accuracy of the load cell used during the testing was 11.0 percent. Additional inaccuracies included the plotting precision of the chart recorder used to record the measurements and a reading error based on the resolution of the grid.

Based on an examination of several representative stripcharts, the team concluded that the reading error would be no less than 1500 pounds, The overall error in the measurement of stem thrust'could have exceeded 10 percent for some of the smaller actuators.

- Separate from thrust measurement accuracy, it appeared that in some-instances, the 80 percent voltage stall _ thrust value published and' disseminated by Westinghouse to its customers may hav? ocen arbitrarily adjusted or rounded up. An example of this practice was identified concerning an MOV at the

- Seabrook plant and is discussed in Section 3.2 below.

2 P

- n n-m e w,,,.- v l - w,, c s wee-~~v

-w

< w A -r v-mw mw... O,,-c e - n-o ne,-vm e m w m

-r+w w an -+w

- - n +r e - mv ~

m ver

Another factor which may affect the validity of the stall test data for MOVs exposed to a harsh environment is temperature effects. The actuators were tested in a laboratory environment but may be required to o>erate under high temperature conditions in the plant. Testing is currently >eing conducted by Limitorque to quantify the loss of efficiency experienced by AC motors under high temperature conditions. When this information becomes available, licensees will need to make adjustments as necessary to the stall thrust and other applicable data in their MOV program.

i in summary, the team identified five factors which may affect the validity of the stall thrust data used in Westinghouse MOV calculations. These were:

load sensitive behavior or rate-of-loading effect stem friction / lubrication thrust measurement accuracy rounding error AC motor temperature effect e

The cumulative effect of these factors may in some instances result in a r

decrease of available stall thrust of 50 percent or greater from the values published in the Westinghouse calculations. Consequently, the Westinghouse-supplied stall thrust data do not appear adequate as a means of demonstrating MOV operability.

3.2 boolicability to Seabrook The team reviewed the basis for the stall thrust data supplied in Westinghouse Letter NAH-3219 to New Hampshire Yankee.

Specifically, the basis for the rating of the actuators (serial numbers 269832 and 269844) for Seabrook Unit 1 motor-operated valves SIV-77 and SIV-102, respectively, was reviewed.

Westinghouse provided to the team the original Limitorque motor operator acceptance forms which documented the test data for these valves.

For SIV-77 on 2/18/78, the stall thrust at-470 volts was 23,200 pounds-and the calculated stall thrust at 80% voltage was 14,223 pounds. Contrary to the above, the 80%

stall thrust specified in NAH-3219 (14,500 pounds) was higher than the stated value in the motor operator acceptance form (14,223 pounds). Westinghouse could not justify the rounding up of the stall data.-

~

Westinghouse's 'ailure to justify the rounding up of stall data in the non-conservative direction is identified as a non-conformance in section 1.1 of-the report (92-01-01).

3.3 Actuator Extended lbrust Ratina Proaram Westinghouse has esta'alished a program (referred to as the " Westinghouse Re-Rating Program) intended to demonstrate that Limitorque actuators for MOVs can withstand thrust ~ greater than the ratings published by Limitorque. The Westinghouse Re-Rating Program applies only to thrust allowable limits 'and does not attempt to change the original torque limits for the Limitorque actuators.

In its Re-Rating Program, Westinghouse has arranged for Limitorque to test certain actuators at the Limitorque facility in Lynchburg, Virginia.

3 93

Westinghouse stated that nuclear power plant utilities Indiana Michigan Power Company (DC Cook), Alabama Power (Farley), and Houston Lighting & Power Company (South Texas) are currently participating in the program. Although the program is not complete, the NRC staff learned during nuclear power plant inspections that at least one facility (South Texas) is applying the preliminary results of the program.

During the inspection at the Westinghouse offices, the staff discussed the Re-Rating Program with Westinghouse personnel and conducted a conference call with Westinghouse and Limitorque personnel. The staff reviewed the " Report on Qualification Testing Program of Limitorque SMB-00 Valve Actuator for increased Thrust Rating for Use in a Westinghouse Nuclear Steam Supply System" (Draft Report B0335; february 3, 1992); " Report on Qualification Testing Program of Limitorque SMB-2 Valve Actuator for increased Thrust Rating for Use in a Westinghouse Nuclear Steam Supply System" (Draft Report 80336; April 8, 1992); Limitorque Corporation Report B0091 for Westinghouse Electric (February 12,1980), and " Thrust Cycling Testing of Limitorque SB-00-15 (3600 rpm) Actuator."

Through its Re-Rating Program, Westinghouse intends to increase the thrust allowable limits for the Limitorque actuators of size SMB-000, 00, 0, I and 2.

Westinghouse also is evaluating the potential for increasing the allowable thrust limits for the Limitorque SB actuators. The staff reviewed only the Westinghouse Re-Rating Program for Limitorque SMB actuators, because the program for SB actuators was not sufficiently established.

Westinghouse used the In*titute of Electrical and Electronics Engineers (IEEE)

Standard 382-1985, "lEEE Standard for Qualification of Actuators for Power Operated Valve Assemblics with Safety-Related functions for Nuclear Power Plants," in establishing SMB actuator groups. Westinghouse stated that an industry standard for the grouping of actuators for thrust limits does not exist and, therefore, followed the guidance of IEEE 382 which involves the environmental qualification of power-operated valve assemblies. According to Reports 80335 and B0336, Westinghouse compared maximum bore diameter, thrust rating, weight, and vertical center of gravity in establishing the 3MB actuator groups. From this comparison of actuator parameters, Westinghouse placed the SMB-000, 00, and 0 actuators in one group and the SMB-1 and 2 actuators in another group.

In discussions with Westinghouse and Limitorque personnel, the staff noted that the parameters selected to establish the SMB actuator groups did not address the differences of the actuators. For example, the SMB-0 actuator has a declutch mechanism significantly different from the SMB-000 and 00 actuators.

Further, Report 80335-indicates that.a Limitorque design review found that the safety factors of the SMB-0 are less than the factors for the SMB-00 for the housing, mounting bolts, housing

~

cover, upper drive bearing, drive sleeve for opening and closing, and the drive sleeve locknut-threads for closing. Westinghouse and Limitorque acknowledged the weaknesses in the justification for grouping the SMB actuators. Westinghouse stated that the grouping would be reassessed and documented justification for grouping would be provided in the Re-Rating Reports.

4 94

Westinghouse selected for testing an SMB-00 actuator as representative of the SMB-000, 00, and 0 actuator group and an SMB-2 actuator as representative of the SMB-1 and 2 actuator group.

In proposing new actuator thrust limits in-80335 and B0336, Westinghouse does not provide justification to account for the variation in actuator performance in light of its intent to test only one actuator for each group. As discussed below, a fire during the testing of the SMB-2 actuator caused Westinghouse to select another SMB-2 actuator for testing. Westinght cknowledged the weakness in its documented Justification to at

. for the sample size of one. Westinghouse stated that its proposed thrust

.lowable limits would be evaluated to justify the amount I

of margin necessary to account for the limited sample size.

The test plan for each tested SMB actuator in the Westinghouse Re-Rating Program consisted of an initial torque test, mechanical aging tests, seismic test, a final thrust overload test, and a final torque test. The. test stand for the thrust tests required the actuator to drive the stem against a load cell in both the open and close directions. The test plan required the tested actuators to be operated in the oper, and close directions under vario* s load conditions for a large number of cycles.

Based on satisfying the test plos, Westinghouse provided its conclusions in-B0335 and B0336 on the total number of cycles that the SMB actuators could withstand at various load conditions.

To apply the Westinghouse Program, the user would need to know the history of the actuator in terms of cycles and thrust values experienced by the actuator since its manufvture.

Westinghouse considered the higher-than-planned thrust exerted on the SMB actuators during the testing to account for the accuracy of the test equi) ment. Westinghouse should address the test equipment inaccuracy as part of tie consideration of the n,argin necessary to justify the testing of one sample from each actuator group. A user would need to include the inaccuracy of their MOV diagnostic equipment when applying thrust allowable limits from the Westinghouse Re-Rating Program.

During intermediate stages and after the testing, Westinghouse conducted inspections of the following components in the SMB actuators: the actuator mounting bolts; housing cover bolts; housing cover (or bevel gear-cartridge in the case of the SMB-00 side mounted handwheel assembly); drive-sleeve bearings; drive sleeve; stem nut locknut; and housing. /.lthough the stem nut will also experience thrust loads, Westinghouse did not include the stem nut as part' of its Re-Rating Program because of the: variations in-this component used in MOVs at nuclear-power plants. During the test program for the SMB-2 actuator, significant stem nut wear occurred which reduced the thrust delivered by the actuator. The team noted that users would need to consider the periodic inspection of this critical com)onent to identify accelerated wear that might reduce the thrust delivered )y the actuator.

- Although-the actuator mounting bolts and housing bolts were listed-as thrust-carrying components, Westinghouse did not include these bolts completely as part of its Re-Rating Program. A user would need to evaluate the adequacy of the actuator mounting bolts supplied by the valve vendor to ensure their 5

95'

adequacy (including tightness) with the increased thrust allowable limits. A user would also need to ensure that the housing bolts provided by Limitorque are maintained tightened to the torque value prescribed by Limitorque and periodically verified to this prescribed torque.

3.3.1 SMB-00 Actuator Testing in B0335 Westinghouse stated that the SMB-00 actuator was selected for testing as representative of the SMB-000, 00, and 0 actuators because of the high proportion of SMB-00 actuators used in nuclear power plants.

The thrust test data in 00335 showed 1865 opening and closing cycles at approximately 180% of the original thrust rating (14000 pounds of force),125 cycles at approximately 183% of the original thrust rating,10 cycles between 258 and 318% of the original thrust rating, 5 cycles near 186% of the original thrust rating, and I cycle at 258% of the original thrust rating. During and following the testing, Westinghouse inspected the actuator and found no noticeable damage to the thrust-carrying components included in the study.

Westinghouse stated that inspections revealed normal wear of gears but no unusual damage.

As a result, Westinghouse concluded that the Limitorque actuators of sizes SMB-000 (new design), 00 and 0 can withstand 1870 cycles at 135% of the original thrust rating,125 cycles at 160% of the original thrust rating, and 5 cycles at 250% of the original thrust rating.

In 80335, Westinghouse noted that the SMB-000 actuator had been redesigned in 1980 to strengthen the housing cover and housing cover flange, and to increase tha housing cover bolt size.

Based on the results of testing the SMB-00 actuator, the B0335 Report indicated that the thrust limits for the old design SMB-000 actuato.tould be 2000 open and close cycles at 135% of the original thrust limit (8000 pounds of force) and 1 motor stall cycle at 200% of the original thrust limit. Westinghouse did not have a documented justification for this change in the thrust allowable limits for the old SMB-000 design, further, the thrust allowable limit of 1 motor stall at 200% original thrust rating is less than the current limit of I cycle at 250% thrust rating published by Limitorque.

In response to staff questions Westinghouse and Limitorque stated that the test results were being evaluated for applicability to the old SMB-000 design and that nuclear power plant licensees would be notified if the conclusion of that evaluation indicates a need to reduce-the thrust rating for the SMB-000 old design.

The staff also noted the need for t

Westinghouse and Limitorque to address the reportability requirements of 10 CTR Part 21 following the completion of the evaluation of the thrust limits for the old SMB-000 actuator design.

During the seismic tests, the SMB-00 actuat.,r was unable to operate because of a problem with the declutch. lever. Westinghouse replaced the declutch lever on the actuator with a lighter lever designed to balance:the declutch mechanism. At the time of the staff inspection, limitorque was evaluating this problem for-appropriate corrective action by licensecs applying:the.-

Westinghouse Re-Rating Program. The team concluded that Westinghouse will need to consider guidance for users to ensure that declutch mechanisms do not prohibit _ operation of SMB-00 and other sized actuators covered by its Re-Rating Program.

Further, limitorque will need to address any 10 CFR part 21 reporting requirements in light of the declutch problem during seismic. testing.

6 96

3.3.2 SMB-2 Actuator Testing i

In B0336, Westinghouse states that the SMB-2 actuator was selected for testing as representative of the group consisting of the SMB-1 and 2 actuators because a Limitorque design review indicated that the safety factors for the majority of the thrust components in the SMB-1 and 2 actuators were lower for the SMB-2 i

actuator.

Two SMB-2 actuators were involved in the test program because of component failures. The first SMB-2 actuator was tested for 1058 open and close cycles at between 150 and 165% of the original thrust rating (70000 pounds of force).

The 80336 Report notes a problem with the collection of peak thrust data, but Westinghouse personnel stated that the lower (and, therefore, conservative) peak thrust value wt selected when a question arose about a particular data point. Af ter cycle 1058, a fire occurred in the SMB-2 actuator as a result of inadequate cooling during the frequent cycling of the actuator.

Because of fire damage to the first SMB-2 actuator, Westinghouse re-initiated the test program with a second SMB-2 actuator.

The thrust testing of the second SMB-2 actuator consisted of 1865 open and close cycles between approximately 143 and 175% of the original thrust rating,133 cycles between t

approximately 155 and 166% of the thrust rating, and 9 cycles between approximately 230 and 267% of the thrust rating.

During the inspection of this SMB-2 actuator following these cycles. Westinghouse discovered a crack in the housing of the actuator, fracture of the upper drive sleeve bearing cup, fracture of the worm gear, and other additional damage.

Westinghouse constructed a combined SMB-? actuator by replacing the housing of the second SMB-2 actuator with the housing from the first SMB-2 actuator.

Westinghouse tested this combined SMB-2 actuator with 6 open and close cycles at approximately 137% of the original thrust rating, and 5 cycles between 135 and 165% of the original thrust rating. Westinghouse then conducted seismic testing on the combined SMB-2 actuator. During the seismic testing, fasteners i

and set screws were checked for tightness and re-tightened as necessary. At one point, the actuator failed to operate during the seismic tests because a sharp edge in the limit switch compartment severed a motor lead. -Following the seismic testing, the housing of the combined SMB-2 actuator cracked during the first closing stroke at.277% of the original thrust rating. After inspecting the damage, Westinghouse performed a final opening stroke at 271%

of the original thrust rating.

Based on the piecemeal testing of the SMB-2 actuators, Westinghouse concluded

~

that the SMB-1 and 2 actuators can withstand 1875 cycles at 125% of the-original thrust rating, 125 cycles at 145% of the thrust rating, and 1 cycle at 250% of the thrust rating.

In B0336, Westinghouse states that neither of the 2 housings for the SMB-2 actuators went through the entire program, but that the original housing from the second SMB-2 actuator was_ subjected to an equivalent number of cycles at the increased thrust allowable limits without :

undergoing seismic testing. The team concluded that Westinghouse needs to reassess the adequacy of the test plan for its Re-Rating Program in light of _

failure of the housing of both SMB-2 actuators.

Based on that reassessment the team conicuded that, West.nghouse should either re-perform the testing 7

97

. a..

with a more realistic t plan or provide documented justification for the thrust allowable limits proposed from the program.

3.3.3 SB-00 Actuator Testing Westinghouse stated that a test program for the limitorque S8-00 actuator was under way.

In the interim, two licensees had requested assistance from Westinghouse in justifying the operability of SB-00 actuators that had experienced thrust greater than the original thrust rating (14000 pounds of force).

For this offort, Westinghouse referred to Limitorque Report B0091 from 1980 that described the testing of an SB-00 actuator. The strip chart data in the report indicated about 2000 opening and closing cycles with thrust between 15500 and 17500 pounds of force and 7 cycles between 30000 and 33000 pounds without failure. The report did not discuss margin to account for the single actuator sample, or the inaccuracies associated with the test equipment and the strip chart, or with reading data from the charts. Therefore, the staff did not consider the report sufficient in the long term to justify an increase in the allowable thrust limit for 58-00 actuators.

3.3.4 Conclusions The staff concluded that utilities should not currently be using the Westinghouse Re-Rating Program. On a case-by-case basis, some of the data in the reports could be useful for evaluating temporary justifications for operability of actuators.

The principal concerns with the Westinghouse program were:

(1)

The 80335 and B0336 test reports do not adequately establish similarity in grouping actuators for which the reports are purportedly applicable, for exanple, there is an inadequate basis for including the SMB-0 actuators with the SMB-00 and 000 actuators rather than the SMB-1 and 2 actuators.

(2)

The B0335 and B0336 test reports do not adequately address the margin necessary to account for statistical variations among actuators within each group of actuators in light of the sample of one actuator from each group for testing.

(3)

The 80335 and B0336 test reports do not justify the ratings for the SMB-000 actuator with the old design housing covers.

(4)

The 80335 and 80336 test reports need to clearly state that the Westinghouse Re-Rating Program does not include the actuator mounting bolts or the stem nuts.

(5)

The 80335 and B0336 test reports need to clearly state that the Westinghouse Re-Rating Program may only be applied to actuators for which the total number of operating cycles and thrust conditions are known.

(6)

The 80335 and B0336 test reports do not specifically address any margin provided to account for the inaccuracy of test equipment used in the Re-8 98

J Rating Program or used by licensees applying thrust allowable limits from the program.

(7)

The B0335 and B0336 test reports do not indicate that licensees will need to maintain proper torque of the housing bolts.

(8)

The B0335 test report for the SMB-00 actuator needs to resolve the issue involving the failed declutch lever.

(9)

The 80336 test report indicates multiple failures of the SMB-2 actuators. Westinghouse needs to reassess the test plan, and either re-perform the tests with a more realistic test plan or provide documented justification for the thrust allowable limits in light of the failures.

(10) The B0091 test report does nnt provide adequate justification for a long

[

term increase of the thrust limit to 16000 pounds for the SB actuator.

3.4 Apolicability to Beaver Valley The team reviewed the bases for the 58-00 operator thrust rating provided in Westinghouse letter WIN 284-6401 to Duquesne Light, dated 4/23/92.

Specifically, Westinghouse stated, "for the SB-00 size operators, limitorque conducted a proprietary qualification program for Westinghouse to show compliance with the operating requirements in the specification of 2000 cycles at a thrust rating of 16,000 pounds in lieu of the 14,000 pound catalog rating." Westinghouse provided Limitorque Corporation Report No. 130091, dated 2/18/80, that documented the test conducted for this purpose, The test consisted of a SB-00-15 (3600 rpm) operator constructed per B/M 680053-A which was tested to the requirements of the Limitorque operator test, dated 12/4/80, and R&D project 680053 dated 11/24/80. A review of the test data sheet indicated that the closure thrust varied randomly between 17,500 pounds and 14,500 pounds. Similarly, the running load varied between 4,000 pounds and 6,000 pounds.

The test report indicated that this SB-00-15 actuator completed 2,111 cycles.

However, it was not clear from the test report whether this actuator was subjected to greater than 16,000 pounds of closure thrust for 2000 cycles of operation. Based on this review, the team concluded that there was adequate test data to support the rating of this S8-00-15 actuator at 14,500 pounds of closure thrust for 2000 cycles. However, the basis for the stated claim of 16,000 pounds was not clear. Westinghouse is currently testing an SB-00 motor actuator as part of its Re-Rating Program to resolve the above and other identified concerns.

4 EXIT MEETING At the conclusion of the inspection the : team held an exit meeting where the team's findings were discussed. The following people were in attendance:

9 99

HAme affi1iation Ifile Jeffrey Jacobson NRC leam Leader, NRR Thomas Scarbrough NRC Sr. Mechanical Engineer, NRR P. K. Eapen NRC Section Chief, RI Michael Runyan NRC Inspector, Region IV Mark Harper Westinghouse Project Engineer Terry Matty Westinghouse Auxiliary Equipment Engineer Pamela Cortuzzo Westinghouse Licensing fledy Abromovitz Westinghouse Manager NATD Product Assuarance John Galembus Westinghouse Licensing Nick Liparulo Westinghouse NSRA Regis Petrosky Westinghouse PA Hanager Hank Sepp Westinghouse Strategic Licensing Issues Lamar Brown Westinghouse Operating Plant Licensing Pete Morris Westinghouse Safety Review Committee Bob Beer Westinghouse Engineering Technology Gary Schlemmer Westinghouse EMD Engineering Larry Walker Westinghouse NATD/AEE Ed Rusnica Westinghouse NATD/MBD Neil Morrison Duquesne light Co.

Rich Faix North Atlantic Energy Services Corp.

1. L. liarpster North Atlantic Energy Services Corp.

Chuck Rowland llouston Light and Power 10 100

g enea,

u

/

J' UNITED STATES n

3 i

NUCLEAR REGULATORY COMMISSION f

WA$HINoTON. D C XfA6

\\-

/

SEP 02 MU Docket No. 99900902 Hr. W. W. Holbrook Vice President of Eastern Operations Wyle laboratories P. O. Box 077777 7800 Governors Drive Huntsville, Alabama 35807 Ecar Mr. Holbrook:

SUBJECT:

NOTICE Of VIOLATION AND N011CE Of NONCONf0RMANCE (NRC INSPECTION REPORT NO. 99900902/92-01)

This letter addresses the inspection of your facility at Huntsville, Alabama led by Mr. R. H. Moist of this office on May 18 through 22, 1992, and the discussions of our findings with you and your staff at the conclusion of the inspection. The purpose of the inspection was to assess selected portions of Wyle Scientific Services & Systems Group Laboratories' (Wyle) Third Party Qualification (TPQ) comercial grade (CG) item dedication program and imple-mentation of selected areas of Wyle's quality assurance (QA) program.

Areas examined during the NRC inspection and our findings are discussed in the enclosed report. This inspection consisted of an examination of procedures and representative records, interviews with personnel, and observations by the inspectors.

The inspection team concluded that Wyle has generally developed an adequate program for performing its TPQ CG dedication activities. However, during this inspection it was found that the implementation of your QA program failed to meet certain NRC requirements. Many of the team's findings appear to be indicative of a lack of attention to detail in the establishment and implemen-tation of Wyle's dedication program.

The most significant issue identified was Wyle's failure to establish similarity to previously tested Automatic Valve Company (AVCO) solenoid valves that were shipped to the Browns Ferry Nuclear Plant.

In addition to the nonconformances identified, Wyle failed to include in its procedures the new requirements that were delineated in the July 31. 1991, revision to Title 10, Code of Federal Regulations, Part 21 (10 CFR Part 21).

The specific findings and references to the pertinent requirements are identified in the enclosed Notice of Violation and Notice of Nonconformance.

You are required to respond to this-letter and should follow the instructions j

specified in the enclosed Notice of Violation when preparing your response.

In your response, you should document the specific actions taken and any additional actions you plan to prevent recurrence.

In addition, you are 101

l l

(

Mr. W. W. Holbrook i requested to provide a written response to the enclosed Notice of Nonconfor-mance. We will consider extending the response time if you can show good cause for us to do so.

After reviewing your response to this Notice of Violation, including your proposed corrective actions and the results of future inspections, the NRC will determine whether further NRC enforcement action is necessary to ensure compliance with NRC regulatory requirements.

The respordes requested by this letter and the enclosed Notices are not subject to the clearance procedures of the office of Management and Bud 9et as required by the Paperwork Reduction Act of 1980, Public Law No.96-511.

In accordance with 19 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosures will be placed in the NRC Public Document Room.

Should you have any questions concerning this inspection, we will be pleased to discuss them with you.

Sincerely,

)

Leif J.

or olm, Chief Vendor Inspection Branch Division of Reactor Inspection and Safeguards Office of Nuclear Reactor Regulation

Enclosures:

1.

Notic? of Violation 2.

Notice of Nonconformance 3.

Inspection Report 99900902/92-01 h

I 1

102

l l

EllCLOSURE 1 NOTICE OF V10LAT10!i Wyle Laboratories Docket No. 99900902 Huntsville, Alabama 35807 During an NRC inspection conducted on May 18 through May 22, 1992, a violation of NRC requirements was identified.

In accordance with the " General Statement of Policy anJ Procedure for NRC Enforcement Actions," 10 CFR Part 2, Appendix C (1992), the violation is listed below:

Section 21.21, " Notification of f ailure to comply or existence of a defect and its evaluation," of Title 10, Code of Federal Regulations, Part 21 (10 CFR Part 21), states, in part, that each individual, corporation, partnership, or other entity subject to the regulations in this part must adopt appropriate procedures to evaluate deviations and failures to comply, in all cases, within 60 days of discovery.

Contrary to Section 21.21 of 10 CFR Part 21, Wyle failed to incorporate this time limit and other new requirements that are specified in the current revision of 10 CFR Part 21, dated July 31, 1991, in Section 19, " Reporting of Defects and Noncompliance per 10 CFR Part 21," of Wyle's quality assurance program manual, dated June 1988. (92-01-01)

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

Pursuant to the provisions of 10 CFR 2.201, Wyle is hereby required to submit a written statement or explanation to th s U.S. Nuclear Regulatory Commission, ATIN: Document Control Desk, Washington, D.C. 20555 with a copy to the Chief, Vendor Inspection Branch, Division of Reactor Inspection and Safeguards, Office of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting this Notice of Violation. This reply should be clearly marked as a " Reply to a Notice of Violation" and should include for each violatio'n:

(1) the reason for the violation, (2) the corrective steps that have been taken and the results achieved, (3) the corrective steps that will be taken to avoid further violations, and (4) the date when full compliance will be achieved.

Where good cause is shown, consideration will be given to extending the response time.

Dated at Rockville, Maryland this l " day of J T, 1992 q

t

_}_

103

l ENCLOSURE 2 NOTICE Of NONCONf0RMANCE Wyle Laboratories Docket No.

99900902 Huntsville, Alabama 35807 Based on the results of an NRC inspection conducted on May 18 through May 22, 1992, it appears that certain of your activities were not conducted in accordance with NRC requitements.

A.

Criterion 1, " Organization," of Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and fuel Reprocessing Plants," of 10 CFR Part 50, states, in part, that the authority and duties of persons and organizations performing activities affecting the safety-related functions of structures, systems, and components shall be clearly established and delineated in writing.

Contrary to 'he above, Wyle failed to adequately delineate in writing the duties and authority of its QC level I aM level 11 inspectors in its Quality Directive QD 11-2, "Engineerino pection and Testing Personnel Qualification Program," procedure i d August 1, 1988.

(92-01-02)

B.

Criterinn V, " Procedures, instructions, and Drawings," of Appendix B to 10 CFR Part 50 requires, in part, that activities affecting quality be prescribed by documented instructions or procedures of a type appropri-ate to the circumstances.

Instructions, procedures, or drawings shall include appropriate quantitative or qualitative acceptance criteria for determining that important activities have been satisfactorily accom-plished.

Section 5. " Instructions, Procedures and Drawings," of Wyle's QAPM, states, in part, that activities affecting quality will be implemented in accordance with documented instructions, procedures, or drawings.

Appropriate quantitative and qualitative criteria will be included to ensure that specified activities have been performed satisfactorily.

Contrary to these requirements, the following deficiencies were observed during a review of selected Wyle Job files:

(92-01-03) 1.

Wyle procedure NEQ 403, "TPQ Receiving and Homogeneity inspections,"

did not typically require adequate acceptance criteria to ensure that commercial grade items did not deviate from the procurement documents and design bases. Wyle's Receiving Inspection Reports and Component Homogeneity Inspection Checklists do not assure that like-for-like replacement, as defined in NRC Generic letter 91-05, was performed.

For example, (a) Wyle's Component Homogeneity Inspection Checklist.for Wyle Qualification Plan (QP) 30026-00 (Wyle Job File 30026/40883) governing the visual inspection of printed circuit

_].

104

i boards did not ensure like-for-like similarity to previously quali-fled units, in that the visual inspection of components was limited to resistors, (b) Wyle receiving inspection report was-inadequate i

for Wyle Job File 30030/40809 in that it did not require the docu-mentation of date codes or verification that the Molded Case Circuit Breakers (MCCBs) were from the same batch, and (c) during a recent TPQ for Job Number 30081/41430 for Browns Ferry Wyle failed to 5

verify that the coils supplied were identical to the ones that were qualified in 1985 and documented in NEQ 17514-1 and were not.of the type in which previous failures had occurred.

2.

Wyle's Test Procedure 6110-26 used for Wyle Jnb File 30030/40809 was inadequate in that it had failed to preclude the use of anomalous-data in the calculation of the mean pole resistance values for use in determining the acceptability of the_ individual pole resistance values for General Electric circuit breakers.

C.

Criterion V, " Procedures, Instructions, and Drawings," of Appendix B to 10 CFR Part 50, requires, in part, that activities affecting quality be prescribed by documented instructions or procedures of a type appropri-ate to the circumstances and shall be accomplished in accordance with these instructions, procedures, or drawings.

Section 5, " Instructions, Procedures and Drawings," of Wyle's QAPM,

[

states, in part, that activities affecting quality will be implemented in accordance with documented instructions, procedures, or drawings.

Appropriate quantitative and qualitative criteria will be included to i

ensure that specified activities have been performed satisfactorily.

l Contrary to this requirement, the following deficiencies were observed during a review of selected Wyle Job Files: (92-01-04) 1.

Wyle QP TPQ-QP-91-3 for Wyle Job File 41661 specified that a Certif-icate of Conformance-(CoC) is to be issued only af ter approval of the test report.

However, the Wyle CoC for printed circuit boards supplied to Boston Edison was found to be issued prior to test report approval.

2.

Although Wyle TPQ procedures require a Konconforming Material Report-1 (NMR) to be initiated when variations are identified during receipt inspection, no such document was apparently prepared when a noncon-forming printed circuit board for Wyle Job File 41661 was returned '

t.o its supplier.

3.

For Wyle Job File 42066, Wyle failed to follow its Table One sample size requirements of NEQ-402 for testing a minimum sample of 2 out of a lot of 4-8 items for 4 Agastat relays: shipped to Southern _

California Edisnn's (SCE) San Onofre Nuclear Generating-Station (SONGS),

Please provide a written statement or-explanation to the United States Nuclear Regulatory Commission, ATTN: Document Control Desk, Washington, D.C. 20555 105

with a copy to the Chief, Vendor Inspection Branch, Division of Reactor inspection and Safeguards, Office of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting thi:: Notice of Nonconformance.

This reply should be clearly marked as a " Reply to a Notice of Nonconformance" and should include for each nonconformance:

(1) a description of steps that have been or will be taken to correct these items; (2) a description of steps that have been or will be taken to prevent recurrence; and (3) the dates your corrective actions and preventive measures were or will be completed.

Dated at ockvil

, Meryland this d ay of 1992.

. _ _ 106

ENCLOSURE 3 ORGANIZATION:

Wyle Scientific Services & Systems Gro:4 ",aboratories Huntsville, Alabama REPORT NO :

99900902/92-01 CORRESPONDENCE Hr. W, W. Holbrook ADDRESS:

Vice President of Ehstern Operations Wyle Laboratories P.O. Box 077777 7800 Governors Drive Huntsville, Alabama 35807 ORGANIZATIONAL Robert Thomas, Quality Assurance Manager CONTACl:

(205) 837-4411 (Ext. 251)

NUCLEAR INDUSTRY Third Party commercial grade dedication of components ACTIVITY:

for safety-related applications and equipment qualification testing.

INSPECTION May 18 through 22, 1992 CONDUCTED:

LEAD INSPECTOR:

x<46$#1)/#$cd 9 JDly 92-R4ndolph'H. Moist, Team Leader Date' '

Reactive Inspection Section No. 2 Vendor Inspection Branch (VIB)

OTHER INSPECTORS:

Kamalakar R. Naidu, VIB Joseph J. Petr<Gsino, VIB Kennetit 'llivad Broo aven National Laboratories APPROVED:

Cw 7/3 2-

,y e fET ~BI/Jacobson,~ Acting CMef Dhte Reactive itspection Section No. 2 Vendor Inspection Branch-INSPECTION BASES:

10 CFR Part 21 and 10 CFR Part 50, Appendix B INSPECTION SCOPE:

Assess the adequacy of Wyle Laboratories' Third Party-commercial grade dedication program for components-used in safety-related applications and to follow up on one-issue concerning AVC0 solenoids.

PLANT SITE APPLICABILITY:

Numercus f

/

107

l.0 INSPECTION SUMHARY l

1.1 Violations 1.1.1 Violation 99900902/92-01-01 Contrary to Section 21.21 of 10 CFR Part 21, Wyle Scientific & Systems Group Laboratories (Wyle) failed to incorporate the new time limit and other new requirements that are specified in the current revision of 10 CFR Part 21, dated July 31, 1991, in the Wyle Quality Assurance Program Manual (QAPM),

dated June 1988.

(See section 3.7 of the inspection report) 1.2 Nontonformance 1.2.1 Nonconformance 99500902/92-01-02 Contrary to Criterion I of Appendix B to 10 CFR Part 50, Wyle f ailed to adequately delineate in writing the duties and authority of its QC level I and level 11 inspectors in its Quality Directive QD 11-2, " Engineering, Inspection and Testing Personnel Qualification Program," of August 1,1988.

(See section 3.3 of the inspection report) 1.2.2 Nonconformance 99900902/92-01-03 Contrary to Criterion V of Appendix B to 10 CFR Part 50, Wyle's procedures did not ensure that activities affecting quality were adequately prescribed in the QA program documents in the following instances.

(1) Contrary to Section 5, " Instructions, Procedures and Drawings," of Wyle's QAPM, Wyle procedure NEQ 403, "TPQ Receiving and Homogeneity inspections," did not typically require adequate acceptance criteria to ensure that commercial grade items (CGis) did not deviate from the procurement documents and design basis. Wyle's Receiving Inspection Reports and Component Homogeneity Inspection Checklists did not assure that like-for-like replacement, as defined in NRC Generic Letter 91-05, was performed.

For example, (a) Wyle's Component Homogeneity Inspection Checklist for Wyle Qualification Plan (QP) 30026-00 (Wyle Job File 30026/40883) governing the visual inspection of printed circuit boards did not ensure like-for-like similarity to previously qualified units, in that the visual inspection of components was limited to resistors, (b) Wyle receiving inspection report was inadequate for Wyle Job File 30030/40809 in that it did not require the documentation of date codes cr verification that' the Molded Case Circuit Breakera (MCCBs) were from the same batch supplied to River Bend nuclear power station, and (c) during a recent TPQ for Job Number 30081/41430 for Browns Ferry Wyle failed to verify that the coils supplied were identical to the ones that were qualified in 1985 and documented in NEQ 17514-1 and were not of the type in which previous failures had occurred.

(See sections 3.5 (1),

3.5 (4), and 3.6 of the inspection report) 2 I

108

-.=

(2)- Contrary to Section-5. " Instructions. Procedures.and Drawings," of Wyle's QAPM, Wyle's Test Procedure (TP) 6110-26 for Wyle Job File 30030/40809 was inadequate in that it had failed tu preclude the use'of anomalous data ^ in the calculation of the mean pole resistance values-for use in determining the acceptability of the individual pole resistance values. (See section 3.5 (4)'of the inspection report) 1.2.3 Nonconformance 99900902/92-01-04 Contrary to Criterion V of Appendix 8 to 10 CFR Part 50, Wyle personnel did not follow documented procedures in the following instances:

(1) Contrary to Section 5, " Instructions, Procedures and Drawings," of-Wyle's QAPM, for Wyle Job File 41661, Wyle 'f ailed to follow the require-ments of QP TPQ-QP-91-3 by issuing a Wyle CoC: for the printed circuit board supplied to Boston Edison prior to test report approval. (See section 3.5 (2) of inspection report)

(2) Contrary to Section 5, " Instructions, Procedures and Drawings," ~of Wyle's QAPM, for Wyle Job File 41661, Wyle failed to follow the require-ments of NEQ 409 by not initiating a nonconforming material report for a nonconiorming printed circuit board returned to its supplier. (See section 3.5 (2) of inspection report)

(3) Contrary to Section 5, " Instructions, Procedures and Drawings," of Wyle's QAPM, for Wyle Job File 42066, Wyle failed to follow its Table One sample size requirements of NEQ-402 for testing a minimum sample of 2 out of a lot of 4-8 items for 4 Agastat relays shipped to Southern California Edison's (SCE) San Onofre Nuclear Generating Station (SONGS),

(See section 3.5 (3) of the inspection report) 2.0 STATUS OF PREVIOUS INSPECTION FINDINGS No previous inspection findings were reviewed during this inspection.

3.0 INSPECTION FINDINGS AND 0THER COMMENTS 3.1 Entrance and Exit Meetinas During the entrance meeting on May 18, 1992, the NRC inspection team (team) discussed the scope of inspection, outlined areas of concern and established interfaces with Wyle management and staff. At the conclusion of the inspec-tion on May 22, 1992, the NRC team summarized their findings and contcrns, and' Wyle management and staff acknowledged this information.

3.2 Procram'and Procedure Review As part of this inspection of Wyle's program for qualification and. Third Party Qualification (TPQ) dedication of Commercial Grade items (CGis), the team 3

109

.=-- ---.

reviewed the following Wyle Nuclear Environmental Qualification (NEQ) engi-neering procedures used for TPQ.

NEQ 402 - Sampling Requirements for TPQ Program NEQ 403 - TPQ Receiving and Homogeneity inspections NEQ 404 - TPQ Engineering Analysis NEQ 406 - Instructions, Procedures, and Test Reports NEQ 409 - The Wyle TPQ Process

+

The team determined that Wyle's TPQ program and procedures for performing dedication of CGis were generally adequate except for some isolated instances.

However, Wyle's procedures for performing receiving and homogeneity inspec-tions did not typically require adequate acceptance criteria as detailed in section 3.4 of this report.

3.3 Receivina and Homoaeneity Insoection Areas Section 6.0, " Procedures," of Wyle's NEQ Engineering Procedure 403, "TPQ Receiving and Homogeneity Inspections," Revision A, dated September.23, 1991, states, in part, that "[a]n appropriately certified Quality Control (QC) inspector will be notified by TPQ personnel to begin the Receiving and Homogeneity Inspection after the Qualification Plan (QP) has been approved, or the documentation required to begin the inspections has been approved. The inspector shall... inspect the items... [and] sign and date the [ Homogeneity)

-l checklist. The completed checklist... shall be forwarded by the inspector to the TPQ Project Engineer. The TPQ project engineer shall_ evaluate the results of the verifications for acceptability... [and) shall document his approval of the checklist (s) and test control record by signing in the Approved by block at the end of the sheet... The project engineer shall designate the require-ment for Level 1 or Level 2 inspector by marking the appropriate block under Inspector Level Requirements on the checklist..."

The NRC team conducted an inspection of the QC receipt and homogeneity inspection area, and observed QC receipt inspection activities. The NRC team also conducted discussions with QC inspectors and Wyle Energy Services Division (ESD) project engineers. During this review the NRC team identified several inconsistencies regarding the manner in which Wyle established or executed portions of its QA program in its receiving inspection and homoge-neity inspection areas.

The NRC team also asked the ESD project engineers what criteria were.used by them to decide whether a Level 1 or Level 2 QC inspector would perform the receipt and homogeneity inspections, as required in Section 6 of NEQ 403. The project engineers indicated that the main deciding factor was the complexity of the CGI that was to be inspected. The project engineer indicated that a level 1 inspector would perform the inspections for less complex items and a level 2 inspector would inspect the more complex components. However, the NRC team observed one of ESD's level 1 inspectors, inspecting Agastat electro-

. pneumatic time delay relays. When the NRC team asked for the criteria, it discovered that no criteria was established by Wyle in either its QA '"Direc-tive P_rocedures," program or'its ESD " Engineering Procedures" program to 4

i

[

110

~.-

delineate specific guidance for the ESD project engineers to make such a-determination-about the required conpetence level of a QC inspector that would be used.

it was determined by the NRC team that.Wyle failed to adequately delineate in writing the duties and authority of its QC level I and level !!

inspectors in its Quality Directive QD 11-2, " Engineering,' Inspection and Testing personnel Qualification Program," of August 1,1988.

(See Nenconform-ance 92-01-02 in section 1.2.1 of the inspection report) 3.4 Bfeceivina inspection and Comognent Homoaeneity latpfction Checklish The NRC team reviewed Receiving Inspection Reports and Component Homogeneity inspection Checklists in each Wyle dedication package and also reviewed selected Wyle procedures in conjunction with the packages. For example, the NRC team reviewed NEQ 403, "TPQ Receiving and Homogeneity inspections." The review was performed to determine whether Wyle's Receiving Inspection Reports and Component Homogeneity inspection Checklists complied with the NRC's regulations and met the intent of industry dedication guidance, such as Generic Letter (GL) 89-02, " Actions to Improve the Detection of Counterfeit and Fraudulently Marketed Products," GL 91-05, " Licensee Commercial grade Procurement and Dedication Programs," and Electric Power Research Institute's (EPRl's) industry guidelines relating to dedication of CGis.

During the review of Wyle's CGI dedication packages, it was noted thct several of the Receiving Inspection Reports and Component Hcmogeneity Inspection Checklists were not comprehensive regarding the critical characteristics that were expressed, and verified.

In addition, some Receiving inspection Reports and Component Homogeneity Inspection Checklists were very general in their format and did not appear adequate to ensure "like-for-like" parts were being supplied.

NRC's GL 91-05 states that "a like-for-like replacement is defined as the replacement of an item with an item that is identical."

The NRC team's review of completed Receiving Inspection Reports identified

(

that the Receiving Inspection Reports appeared to be strictly for verification of such items as quantity received, enclosed receiving documents, damage, markings, desiccant, and whether a manufacturer's CoC was enclosed.

A review of several completed Component Homogeneity Inspection Checklists identified that the Component Homogeneity Inspection Checklists contained QC inspector verification requirements such as,. catalog /part number verification, damage, homogeneity, assignment of Wyle serial numbers, and minor variations.

The NRC team concluded that Wyle's Receiving-Inspection Report and Component-Homogeneity Inspection Checklist do not consistently ensure an adequate like-for-like replacement, as defined in NRC Generic Letter 91-05.

(Examples are listed in Nonconformance 92-01-03 in section-l.2.2 (1) of the inspection report) 3.5 Component-Specific Dedication Reviews To assess the adequacy of implementation of Wyle CGI dedication procedures, the NRC team reviewed selected Wyle Job File packages and interviewed the cognizant Wyle personnel.

The NRC team found the job files to be comprehen-5 111

sive and complete. Each file typically contained the customer P0, technical specifications (or reference thereto) imposed by the customer, Wyle QP, Wyle Test Plan (TP), any CoCs issued by Wyle and Wyle sub-tier vendors, shipping memos, material receipt inspection records and related correspondence.

However, several deficiencies related to Wyle's implementation of Appendix B to 10 CFR Part 50 were identified as a result of this review.

Specific exampics include:

(1)

Wyle Job File 30026/40883, "Rosemount Circuit Board for Use in Chlorine Detection Systems (Rosemount Part No. XH3115-OlB)," was prepared in response to Wolf Creek Nuclear Operating Corporation PO 537332, dated June 6, 1992, for the Wolf Creek Generating Station (WCGS).

Technical Requirement 2.03 of Wolf Creek P0 537332, required the circuit boards ordered under this P0 to

"...be reviewed to ensure that there have been no changes in materials, fit, form, or functional properties from those originally supplied to WCGS."

Subsequently, Wyle QP No. 30026-00, Rev A, dated November 8, 1990, and Wyle TPQ Test Report (TR) No. 30026-99 Dated November 14, 1990, "for Rosemount Analytical Uniloc/ Delta Division Chlorine Detector Parts for Use In Nuclear Power Generating Stations," was prepared to document the method and procedures to be employed for the environmental qualification of the circuit board which was procured commercial grade by Wyle from the Original Equipment Manufacturer (0EH) Rosemount.

Since the unit was to be shown by Wyle to be identical to previously qualified circuit boards, Wyle determined that qualification by a similarity analysis was suitable. As stated in Section 1, " Purpose " of the Wyle QP: "The tests outlined in this QP will demonstrate the subject equipment is identical (emphasis added) to the test specimens in Wyle TR 40130-2".

Due to a number of f actors, including the complex nature of printed circuit boards used in electronic instrumentation, and the frequent substitution of sub-components by manufacturers of CGIs, the inspectors questioned the method employed by Wyle t9 satisfy its stated objective of demonstrating that the board was identical to a previously qualified unit.

Section 3 of the Wyle QP, which details the Qualification Program, was found to require a visual inspection of all components.

The stated purpose of the visual inspection was to identify any noticeable differ-ences between the circuit board being qualified and the previously qualified unit. To aid in accomplishing this task the Wyle qualifica-tion program included inspection forms which graphically depict the circuit board and its subcomponents. This form was included in the Wyle Qualification Plan as Attachment 2, Homogeneity Inspection Record, Form A2-7.

Form A2-7 was to provide the basis for Wyle's qualification of a commercial component by similarity, However, the NRC team review of this form found it to lack sufficient detail to achieve its intended 6

112

objective of demonstrating that the units currently being qualified are identical to units previously qualified under Wyle TR 40130-2. While the form required a fairly detailed comparison of critical characteris-tics of certain components, such as resistor " color-codes," it did not require a comparison of any solid state devices (transistors, diode.;,

integrated circuits). _ Therefor _e, Wyle failed to document _whether variations have occurred in the construction-or manufacturing process of these devices from those originally qualified.

The NRC team was concerned, therefore, that Wyle'_s failure to adequately verify that the critical characteristics of all subcomponents mounted on the circuit board were compared to those originally qualified, may have resulted in undetected variations between the units, thereby placing;the validity of the similarity analysis in question.

Based on the above, the NRC team concluded that Wyle procedures govern-ing the visual inspection of components to assure like-for-like similar--

ity to previously qualified units were inappropriate for the circum-stances. As a result, the objectives of Wyle QP 30026-00 were not fully demonstrated. This deficiency constitutes a nonconformance with_ respect to Criterion V, " Instructions, Procedures, and Drawings," of Appendix B to 10 CFR Part 50 which requires that activities affecting quality be prescribed by documented instructions, procedures, and drawings which include appropriate quantitative or qualitative acceptance criteria for determining that important activities have been satisfactorily accom-plished.

(See Nonconformance 92-01-03 section 1.2.2 (1) of the inspec-tion report)

(2)

Wyle Job File 41661, " Nuclear Environmental Test Program on a Basler Electric Part No. 90-32101-102 Voltage Regulator Printed Circuit Board,"

was prepared for Boston Edison Company's (BEC0's) Pilgrim Plant.

BECO issued PO RRR001021 on January 24, 1991, to Wyle to qualify one (1) of two (2) Basler voltage regulator circuit boards being supplied for use in a mild environment at its Pilgrim plant. The circuit boards to be qualified were procured commercial grade by Boston Edison from Public Service Electric and Gas (PSE&G) and. drop shipped to Wyle for qualifica-tion.

~

Wyle QP TPQ-QP-91-3, QP for Basler Electric Part Number 90-32101-'102 Voltage Regulator Printed Circuit Board For Use in Nuclear Generating Stations, Rev. A, dated January 26, 1991, was prepared to delineate the qualification method and procedures necessary to qualify the commercial grade circuit boards for use in a Class IE system and satisfy the BEco procurement-document requirements.

Among its specific technical requirements, BECo P0 RRR001021 required-Wyle to verify, and state in its CoC,:that the qualified circuit board is " identical to information contained on Drawing _M6-58-1, Revision E3, as supplied."

In its CoC, dated January 29, 1991, Wyle_was found to state, in part, that: "the Basler Electric Printed Circuit Board furnished to BECo under Wyle Job Number 41661 is identical to informa-tion on BECo Drawing No. M6-58-1, Revision E3, and is Class lE qualified 7

113

= -. -

to the requirements of BECo Purchase Order No. RRR001021 in accordance with 10 CFR Part 21, 10 CFR Part 50, Appendix B, IEEE 323-1974, IEEE 344-1975...".

The inspection team identified the following anomalies during the review of information contained in Wyle Job File 41661:

Section 7 " Certificate of Conformance," of Wyle QP TPQ-QP-91-3, states, in part:

"A CoC shall be issued when the test report is approved." Contrary to this requirement the NRC team's review of-the Wyle CoC found it to be dated January 29, 1991, while the Wyle TR No. TPQ-TR-91-3 was found to be dated February 8, 1991. The NRC team determined this to be a nonconformance to Criterion V, "In-structions, Procedures, and Drawings," of Appendix B to 10 CFR Part 50 which requires, in part, that activities affecting quality be performed in accordance with, documented procedures.

(See Noncon-formance 92-01-04 section 1.2.3 (1) of the inspection report)

When variations are identified during receiving inspection Wyle procedure NEQ 409, "Wyle IPQ Process," requires, in part, that a Nonconforming Material Report (NMR) be issued. By letter dated January 25, 1991, from Wyle to PSE&G, Wyle stated that it was returning 1 of 2 boards furnished to BEco because it was not identi-cal. However, contrary to Wyle procedure NEQ 409, Wyle had not issued an NMR. The NRC team found this to be another example of a nonconformance to Criterion V of Appendix B to 10 CFR Part 50 which requires, in part, that activities affecting quality, such as receipt inspection, be performed in accordance with documented procedures.

(See Nonconformance 92-01-04 section 1.2.3 (2) of the inspection report)

(3)

Wyle Job File 42066, was for 27 commercial grade Agastat model 7032 PDC electropneuotic, double-head, time delay relays, that Southern Califor-nia Edison Company (SCE) ordered for Class IE safety-related applica-tions at its San Onofre nuclear generating station (SONGS). The relays were ordered from Wyle on SCE PO 6T071027, of July 31, 1991. The P0 imposed Appendix B to 10 CFR Part 50, 10 CFR Part 21, required that Wyle impose traceability requirements on the manufacturer to serialize the relays, seismic testing requirements, and other requirements such as requiring that SONGS receive four Agastat 7032 PDC relays no later than August 1, 1991, in conjunction with the NRC team's review of Wyle Job File 42066, several of Wyle's QA program procedures were also selec-tively reviewed, such as NEQ 402, " Sampling Requirements for TPQ Program," Revision A, dated June 1, 1988.

Subsequently Wyle-issued P0 4-3273-P to Amerace, dated July 25, 1991. This P0 ordered 35 commercial grade Agastat 7032 PDC relays from the manufacturer, Amerace Corporation.

Wyle required that a " certificate of same-lot /same date code," and a bill of materials be provided by Amerace.

The NRC team's review found-a few deficiencies in Wyle's package regarding compliance with its QA program procedures and QAPM.- For example, although SCE required Wyle to " impose appropriate controls on 8

i 114 i-

i Amerace to assure traceability (by placement of serial numbers on each relay) to assure unique identification of each relay supplied to SCE/ tested oy Wyle," Wyle did not impose that requirement on Amerace, nor did Wyle identify and document serial numbers or date/ lot codes as characteristics to t, serified and recorded on the initial shipment of five relays. Wyle's IPQ Test Report TR-91-28, did not contain any objective evidence.o show whether the five Agastat 7032 PDC relays, that were supplied to kyle with Amerace's internal " ship ticket No." 289043-01 contained Amerace serial numbers or date/ lot codes. The NRC team also found that Amerace did not supply the entire order at one time.

Instead, Amerace supplied five relays that were manufactured and controlled as one lot, and 30 relays that were manufactured and con-trolled as another lot at a later date.

Consequently, in order to meet SCE's requirement of receiving four relays no later than August 1, 1991, Wyle performed its dedication and qualification testing on one test specimen out of the five that were first received by Wyle. Af ter successful testing of the one sample, and functional testing of all five, Wyle shipped the four relays to SONGS with a Wyle CoC, dated August 2, 1991, that they met all imposed requirements. However, Wyle's NEQ 402 procedure required that a minimum of two samples be tested from a lot size between 4-8 items.

In summary, Wyle did not impose the required SCE serial number traceability require-ment on Amerace, failed to verify and record date/ lot codes and serial numbers for the initial lot of five relays, failed to test the requ' red minimum number of test specimens for the initial lot of five relays, and certified that the four relays supplied to SONGS met SCE's P0 require-ments.

(See Nonconformance 92-01-04 section 1.2.3 (3) of the inspection report)

(4)

Wyle Job File 30030/40809 - Wyle prepared Test Report No. 30030-99,

" Nuclear Environmental Qualification Test Program On General Electric (GE) Catalog TFJ224 Molded Case F225 Line Circuit Breakers," for the Gulf States Utilities Company (GSU), the licensee for the River Bend Nuclear Power Station.

GSU issued P0 89-K-73108 of October 13, 1989, to Wyle to supply four GE TFJ224150 type circuit breakers for Class IE service.

In the scope of the P0, GSU stated: Wyle shall provide copies of the P0's procuring the molded case circuit breakers (MCCBs) which are being supplied (to GSU) as qualified replacement parts.

In lieu of this requirement, Wyle may use a method of marking or identifying the breakers and provide a statement on the CoC which identifies the breakers supplied and traces them through documents to the OEM. GSU stated that 10 CFR Part 50, Appendix B and 10 CFR Part 21 were applicable to the P0. Consequently, Wyle issued P0 4-3139-P of October 5,1989, to Matthew Electrical Company (MEC), a local distributor of GE products, to supply 15 TFJ 224150 and 23 TFJ 224100 GE MCCBs.

Wyle qualified four MCCBs for GSU based on the successful completion of thermal aging and seismic tests on a sample selected from the lot of 15 9

115

IFJ 224150 MCCBs. These tests included post-thermal aging functions, post-functional and post-inspection tests on seismic specimens. Wyle performed electrical tests on four TfJ 224150 MCCBs, determined the test results acceptable and shipped them to GSU.

lhe NRC team reviewed the test report and relevant documentation, and determined the following:

lhe provisions in Wyle's Receiving Inipection Report to verify homogeneity of MCCBs were inadequate.

For instance, the ESD QC inspector who completes the Receiving inspection Report was not required to document the date codes on the MCCBs, or to verify that all the MCCBs were from the same batch.

(See Nonconformance 92-01-03 section 1.2.2 (1) of the inspection report)

The results of the electrical tests were acceptable, except in the case of pole resistance tests.

In Notice of Anomaly (NOA) 2 of the NLQ report, Wyle identified that the pole resistance values of two MCCBs measured prior to seismic tests were high.

Wyle measured them again after the completion of other tests and found they decreased to the normal values.

The NRC team determined that Wyle incorrectly included the high resistance values to calculate the mean pole resistance values for both N0A 2 and for post-seismic test results.

This error resulted in the value for the post-seismic pole resis-tance manufacturer's mean exceeding the lowest pole resistance reading by more than 50-percent, lhe NRC team informed Wyle representatives that contrary to 10 CFR Part 50, Appendix 0, Criterion V, Wyle's Test Procedure 6110-26 was inadequate in that it had failed to preclude the use of anomalous data in the calculation of the mean for use in determining the acceptability of the individual pole resistance values.

(See Honconformance 92-01-03 section 1.2.2 (2) of the inspection report) 3.6 InhaluLLvALutimL9LEmtipmaLQ1tallLlatinttBmr_Li lhe NRC team reviewed the Wyle Nuclear Environmental Qualification Test Report (NEQ) 17514-1, dated March 14, 1985, on a pneumatic control assembly (PCA) manufactured by the Automatic Valve Company (AVCo), Novi, Michigan, for the Tennessee Valley Authority. The NEQ is also valid for the Detroit Edison Company (fermi 2), the Pennsylvania Power and Light Company (Limerick) and the Commonwealth Edison Company (Dresden and Quad Cities).

The PCA is a manifold that is assembled on an actuator to operate a main steam isolation valve (MSIV) or a main steam relief valve (MSRV).

The PCA directs air to the actuator to operate HSIVs or MSRVs.

The PCA is an assembly of multiple solenoid operated valves (SOVs) which control air to and from air valves, which in turn directs motive air under a piston operator to open the MSIV/MSRV or on top of the piston to drive the MSIV/MSRV closed.

TVA had issued P0 TV-56071A to Wyle to qualify the PCA manuf actured by AVCo to satisfy the intent of the institute of Electrical and Electronic 10 116

~..

= -..

1 Engineers (IEEE) Standard 323-1974, "lEEE Standard for. Classifying Class IE Equipment for Nuclear Power Generating Stations," and'IEEE Standard-382-1980, "lEEE Standard for Qualification of Safety-Related Valve Actuators."

The NRC team reviewed two issues documented in this NEQ report', one related to the use of Hougton H-620 type lubricant and the other related to the failure of solenoid operated valves (SOVs) manufactured by Airmatic-Allied, incorporated, Wilmington, Ohio installed in the PCAs.

Regarding the Houghton H-620 lubricant, the NEQ stated that Super-0-lube type lubricant was not qualified for use in the PCA because Super-0-lube solidified during tests. Wyle attributed the valve qualification-failures in the PCA to " excessive amounts" of Super-0-lube applied to the valves (Wyle test report, page xviii, item 15). As a result of the failure of the valves with Super-0-lube, Wyle stated that it used Houghton H-620 type lubricant and successfully qualified the PCA.

The above result appears to be contrary to the experience of the General Electric Company Nuclear Energy (GE NE), San Jose, California, which in Service Information Letter (SIL) No. 329, dated June 1980, informed its customers that the use of Houghton SAFE # 620 in MSIV operators had shown a tendency for early seal deterioration and cylinder galling. On December 10, 1986, after installing AVCo PCAs in MSIVs the Pennsylvania Power and Light Company (PP&L) experienced failures of the AVCo PCAs for all 16 MSIVs at its Susquehanna, Unit 2 plant. The Houghton H-620 l

lubricant corroded the unanodized aluminum manifolds in the PCA. -On December 19, 1986, with copies to Ceco and PP&L, AVCo notified the NRC Region Ill Office of the corrosion problems experienced at Susquehanna, Unit 2.

On February 10, 1987, AVCo notified Detroit Edison (Fermi 2) and TVA (Browns Ferry) of the problems which had been experienced with Houghton H-620 lubricant at Susquehanna.

The NRC team reviewed a Wyle letter of April 30, 1985, to TVA in which Wyle seemed to reassure TVA that Hougton H-620 was _a compatible lubri-cant.

In this letter, Wyle stated-that it~ discussed the compatibility of H-620 with a Hougton representative who-advised that:

"H-620 was compatible with some aluminum but not all.

If there is incompatibility, the fluid will attack the aluminum rather quickly, within approximately 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> at normal temperatures."

The responsible Wyle engineer informed the team that he was unaware of the corrosive nature of Hougton H-620 and the failures of MSIVs'experi-enced at Susquehanna. The NRC team informed him of NRC Information

-Notice 90-11 (IN 90-11) which discussed the non-compatibility of Houghton H-620. He acknowledged that he read IN 90-11 but had not noticed the significance. Wyle stated they would review this informa-tion under their 10 CFR Part 21 program.

11 117 l

I The second item concerns the use of SOVs manufactured by Airmatic-Allied Incorporated ( AAI), Wilmington, Ohio, supplied to AVCo for installation in its PCAs.

AAI purchases the electrical coils used in the assembly of its SOVs from two different vendors, the five Star Company and the Quality Coils Company, both are located in Connecticut. NEQ 17514-1 documented that the coils of four S0Vs failed during the qualification and that Wyle was unable to determine the identity of the coil manufac-turer from AAl.

In a letter of November 8,1984, AAI stated that it could not determine which of its two suppliers, the five Star Company or the Quality Colls Company supplied the electrical coils which were assembled in the SOVs that failed.

During a recent TPQ for Job Number 30081/41430 for Browns ferry Wyle failed to verify that the coils supplied were identical to the ones that were qualified in 1985 and documented in NEQ 17514-1 and were not of the type in which previous failures had occurred. (See Nonconformance 999C3902/92-01-03 section 1.2.2 (1) of the inspection report) 3.7 10_ CfR Part 21 P01L10! Land Procc@tc.

lhe NRC team observed that Wyle f ailed to ensure that its 10 CFR Part 21 procedure encompassed all of the requirements of the July 31, 1991, revision of 10 CFR Part 21. Wyle stated that it was not aware that a new revision to 10 CFR Part 21 had been issued.

Therefore, the NRC team provided Wyle with a copy of the July 31, 1991, revision of 10 CFR Part 21. Wyle took immediate corrective action to revise its 10 CFR Part 21 procedure and its posting document that Wyle used to describe the regulations and its procedure. NRC issued NRC Information Notice 91-76 on November 26, 1991, to inform all holders of operating licenses or construction permits and vendors for nuclear power plants that the NRC had ammended its regulations regarding the reporting of safety defects under Parts 21 and 50.55(e) of Title 10 of the Code of federal Regulations. (See Violation 92-01-01 section 1.1.1 of the inspection report) 12 l

118

'l 4.0 PERSONNEL-CONTACTED

-WYle laboratories

+*

W. Holbrook, Vice President of Eastern Operations

+

S. Hyten, Energy Services Division General Manager

+*

J. Gleason, Director-of Engineering Services Division

+*

_C. Thibault, Deputy Director of Engineering Services Division

+*

R. Thomas, Quality Assurance Manager

+*

T. Patterson, Third Party-Qualification Manager

+*

T. Brewington, Equipment-Qualification Manager

+* F.-Johnson, Nuclear Equipment Qualification Manager

+*

K. Monaco, Senior Operations Analyst

+*.E. Schum, Senior Engineer s

J. Thomason, Engineer P. Lubeski, Engineer D. Tonn, Engineer G. Wiggins,-Third Party Qualification Technician B. DeFour, Third Party Qualification M. Howard, Senior Engineer R. DeFour, Engineer A. Stebbins, Secretary.

B. Quinn, Senior Staff Engineer T. Hamilton, Quality Assurance Engineer A. Horsman, Senior Staff Engineer Nuclear Reaulatory Commission J. Jacobson, Acting Section Chief, VIB, NRR, NRC

  • Attended the. exit meeting

-+ Attended the entrance meeting 13 119

Selected Bulletino and Information Noticos Concerning Adoquacy of Vendor Audits and Quality of Vendor Products IDEdlED TlThE 1.

Information Notice 91-52

- Nonconaarvativo Errora -irl Supplomont 1 overtemperature Delta-Temperaturo (OTdT) Sotpoint Caused By Impropor Gain Setting 2.

Information Notice 92-40 Failure of Exido Batteries:

3.

Information Notico 92-51 Misapplication and Inadoquato Teoting of Molded-Capo Circuit-Breakota 4.

Information Notico 92-56 Counterfeit Valvoa in the Commercial Grado supply Syntom 6.

Information Notico 92-63 Cracked Insillators in ASL Dry Typo Transformora Manufactured By Weat35,nouse Electric Corporation 6.

Information Notico 92-66 Accoon Donlod to NRC Inapoctora at Fivo Star.

Products, Inc..and Construction Products Roacarch, Fairfield, Connecticut 7.

Information Notico 92-67 Doficiency in Donign Modifications to-Addreas

-Failurco of flillor Actuatora=

Upon a Gradual'Lona of Air Proasurc

.8.

Information Notico 92-68 Potentially Substandard-Slip-On, Wolding Nock, and-Bl.ind.

Plangea 9.

Information Notico 92-70 Westinghouse Motor-Operated:

Valve Performance Data Supplied to Nuclear _ Power Plant Liconacon 120 I

e a

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

=

)-l.

(

r k

CORRESPONDENCE RELATED TO-VENDOR ISSUES 9

d

}

e e

1

( *.

?

e 121 3

1 w

e-ww ien 6

-wog de-e me-r d=-e-mem erw*

' W es w-g.

eop. gem,.w.ay--ey.,+

y v

w.

g--

m m

yatf z--e e*

r--

-w w-+-?-

ti

l

\\

/pm assuq'c,

'n UNITED STATES 3

i NUCLEAR REGULATORY COMMISSION

('h K"b/e/

WASHINGTON, D C 20566 g s,-

AUG 12 m2 Docket flo. 99901227 Mr.

R.

S.

Orr liuclear Administrator Mail Stop 7A-47 Boeing computer Services P.

O.

Box 24346.

Seattic, Washington 98124-0346

Dear Mr. Orr:

SUBJECT:

RESPONSE TO 10 CFR PART 21 INQUIRY By letter dated June 12, 1992, you requested the U.

S. Nuclear Regulatory Commission's (NRC) position regarding your unsuccessful attempt to contact certain former Boeing customers to inform them of error notices received from your former suppliers of nuclear related codes.

We have described our position in Enclosure 1 to this letter.

The current revision to 10 CFR Part 21 dated July 31, 1991 is also provided for your information as Enclosure 2.

The latest address and phone number information on file at the NRC for the requested companies is provided as Enclosure 3.

Should you have any further questions, please contact Mr. Ronald Frahm, Jr. of my staff at (301) 504-3216.

Sincerely, b.

/

/

I J

L if J.

orr olh, Chief Vendor nspe'dtion Branch Division of Reactor Inspection and Safeguards office of Nuclear Reactor Regulation

Enclosures:

1.

Response to Questions 2.

July 1991 Revision to 10 CFR Part 21 3.

Customer Address Information 122

ENCLOSURE 1 RESPONSE TO BOEING COMPUTER SERVICES LETTER QUESTION (paraphrased):

Does the NRC require: (1) a copy of every supplier error notice on the enclosed list which was not received by a former customer and (2) a copy of.cVery future error notice received by Boeing which is not sent to one or more former customers?

NRC RESPONSE:

Part 21 to Title 10 of the Code of Federal Regulations (10 CFR Part 21) requires that suppliers of basic components, or services associated with basic components, evaluate deviations or notify the customers if the supplier does not have the capability to perform the evaluation.

The intent of 10 CFR Part 21 is to assure that users of basic components are made aware of any potential defects which could create a substantial safety hazard.

Your letter states that Boeing has attempted to inform all of their purchasers when error notices were reported so that the customers could evaluate the safety significance of the deviation.

In the cases where Boeing was unable to contact former customers, Boeing should notify the NRC promptly so that the NRC cr.n determine if the deviation is of significance to j

warrant generic communications to the nuclear industry.

The notifications should be made initially via facsimile or telephone communication to the NRC Operations Center followed by written transmittal to the NRC Document Control Desk as-detailed in 10 CFR Part 21.

123

l l.

I i

UNITED STATES NUCLEAR REGULATORY COMMISSION l

RULES and REGULATIONS D1 CLOSURE 2 Tiftt 10. CH APfift 1. CODI Or f t DIR AL fitout ATIONS - INIROV pg g,

_.y.-

L m. _m-

__.a 21.1-21.31a)

~

PART REPORTING OF DEFECTS AND NONCOMPLIANCE osasau Peonuomo safety huards or (b) that the facility, and failures to comply under this part 8M activity, or bute componerd supplied and the responsibthey ofindividual 31 I Puri" to suf h fullity or activity containe de-director $ end responsible ofkers of j

lette, erhtch could treate a substaritteJ euch be4 nous to report defects under 38 e InterpreteHone E 8&I8%F hM*Fd.10 Imth'dlet'IF h0tLIY section 22 of the Energy iteorynlubon at a communneuena a the Cornmiselon of such f atture to Act of gg74 si e Poetine enutremenu

", tiomply or such delect, urdtse he hu

[c} to, p,,,one tg,nnd to operei,,

31 1 3 s ernpuens utual knostedet that the Commla.lon as e intmenei.oa son.sima sept,em..w bu been adequately infortned of such mlear per pleM undu pan M of tMe M 5 e e'u W delect or f ailure to comply.

thopter. evolustion ol potenual defecto W ep prim up of defe Hoetettation f

under il 50 7L M 73 or i 73ft of thle g3g

x. u me-t

. -,e, or as iereen.neeen o nso.e i

8[*['EPU emnte ur e strM and no evekswa

' I"

  • eve etion, notification, and reportit g Pa c esu m h u m e othnwise in pene 3h 34. 35,39,00,(10, geHon to report defute under this 31.38 Precitteinent documenk 01,71 or art 72 of this chaplet, to each part end the responsibihty of lndividual g"

g indMdue, partnerehlp, c.orporeHon, or darntore end responsible offerers of other entity beenud pursuant to the eur.h luenene to report delet te under 21 el inse"'

)s 2t il Me nwnana and impalmend emnds regulebone in this chepter to poones.

pchon 2m of the Enngy hwgemsehon r neln @m the Med Mew AC'"II874 t'meoersusse suurte metenel, byproduct rnalertal, (d) Nothing in then regulet one 36 el reliun to nourt.

opecial nucleer metenal. and/or spent should be deemed to preclude either en fuel and high level radioecuve wute. or

& individual. e manufetturer, or a suppher t

to construct, manuf erture, ponen, own.

y of a commercial grade item (en

\\

Auttmet'r Sec tet e4 Stat 64 es operate or trenefer within the UnHed i 213(eW) not oubbct to the en ended m 13e e)Ide% en es amended States eny production or uuhtetton reguleHone in this part from reporting to lel U 5 C W1.12AJ) ou 21. u evnended lecthty or independent opent fuel the Commiselon, a known or eueperted lui se lien 1242, se esiended llee 142 U SC storage instellebon (ISl%Il or monllored defect or fellore to cornply and, so e

e sten imed undee e c in tot retneveble storage tnotellation (MRe.),

authonsed by law, the identity of

$ Pub 1. ehen se sin In*,22ei tu U R C' end to each dirntor and rnponenble en)one so reportma will be withheld

, tm n tm ey ofhter of such a hunen, The from disclosure NRC regional offices g

Foe the purynees of set In se Stat en e, regulettone in this port apply eloo to end hudquarters will no i collut O amended let U $ C lit)) 18 71 sk 21litel e ach individual. corporation, telephone celle from indivl uale who and il 31 em ieewd undes m, telb 6e Sten partnership or other entity dotng wish to speak to NBC representoune usa o einenaed let u s C 12mlbli er 4 business within the United Stein, and conenning nuctor eefety related 18 at n 21 et end 31 at ere seused endee m.

enh director and rnponsible officet of problerne the locshon and telephone tels 6e Siet, em ee emended lC U EC sut h organitehon. that constructs a riumbers of the five regions (onewered

[D

$ production or ubhsaunn facihty hcensed during regular wurb6ng hours) are heled

$ for manufetture.conetructlon, or in appendin D to part 30 of this chepter.

GentuaL Penstalone D 'nehon pursuant to art 50 of thle The telephone number of the NitC epter, en 1551 for t e olorage of opent Opermisone Cenin (enewered 74 hours8.564815e-4 days <br />0.0206 hours <br />1.223545e-4 weeks <br />2.8157e-5 months <br /> e r

IIU P""

fuel brenud pureuent to part 72 of this doy-including hohday4)is p01) 911 The regulattoru in thl.a part eatsb thopter, a MitS for the otorega of opent 0530 llah procedures and recutremente for fue) og high ley,t radioscqiy, weete

[

kmplementation of section 206 of the purovent to part 72 of this chapter, or a g flJ Dehninene Trursy He^rgantation Act of 1914-That ention requires any IndividueL1 eologic repoentory for tho disposel of thrector or resporalble officer of a tsh level radiusctive weste under Eett taHl) "Dule cornponent? S hen ap

- f 6tm cranstructing, omning, operating

  • E

"'"P lilled to nuclear pos er reartors rneans I or supplytrig the componente of any tomPmnte for a facthty or echve a pl W st m t e e> m n e ent

% f aelhty or activity which ta licensed or hunnd other then for emport uru er or part thereof neceuary to asuure (H E otherewtee regulated pursuesnt to th, perte 30. 44 M 00. 01,70,71. or pert y2 nf

,e the integrtty of the reactor coolant 1 Momic Energy Act of 1954. u arner@

ed, or the Energy litorgantaation Act (b) for pnoone konsed to construct a presAure bourutar), till the capability e

to shut doun the reacter and metntain m

of 1914, who obtains information res.

faciht under a construction iermtl

! 11 in a safe shutdoen condition, or (till sonably indicating' (a) That the f actil, leeve under l 50 23 of this c ispter, 14 the espubtilty to prevent of mitigate

'y. utivity or t,ule component sup-nelohon of potential defute and the corutouences of arcadente which died to such fultity or activity Iatte to fathues to mmply and reportma of cout(s result in potentle.1 ofIsite eate comply with the Atomic frergy Act of d*fects and failures to comply under sures comparable to th4at referred to 1954, u amended, or any appH(able l SO SHe) of thte chapter settelin enh in i 100 li of thta chapter rule, regulation, order, or lleense of person e evelustion. nohficallon. end (21' tam component." s hen opphed

~

the Contmtaelon relattng to substantled repntung chhgebon to report defecte to other f atihties and t hen apphed to I

21 1 July 31,1991

2@

PART 21 o REPORTING OF DEFECTS AND NONCOMPLIANCE l

other activities Licensed punuant t.o Parta 30, 40, $0. 60. 61,10.11, or 13 of

12) The installation, use, or oper-Im)
  • Substantial sefety hasard" snesna this chapter, rneans a component, ation of a bule cornponent containing a loss of safety function to the extent I structure, system, or part thereof that a defect u def tned in paragrsph (dMll that there is a major reduction in the I ts directly procured by the lleenace of of this section; or degree of protection provided to public

[ a f acility or activity subject to the reg.

(3) A deviation in a portion of a fa.

beelth and safety for any facihty or

= ulations in tttts part and tri which a c111ty subject to the construction

- actalty liceped. other than for emport,

defect (see i 21.3(d)) or f ailure to permit or manuf acturtng licensing re' $ pm4M to Parts M M M M u comply with any applicable regulation quirements of Part 50 of this chapter g 74 71* ce 72 of this cb 1"*

P la this chapter, order, or license tsaued provided the deviation could, on the e by the Commhslon could create a sub.

buis of an evaluation, create a euk

- stantial safety hazard (see i 21.3(k)).

Stantial safety hazard and the portion E of the f acility contatning the deviation has been offered to the purthater for (n}..Bupplying" or " supplies" menna (3)In all cases basic component Includes safety related design, analysis, j "CC'Dl'UC Of contrsetually responsible for a baalc component used or to be used in a fa.

y inspection testmg fabncation.

a M replacement parts or consultmg

[

(4) A condition or circumsta. nee in, cility or activity which is subject to volving a bule component that could the regulations in th14 part.

f services that are associated with the

~

3 component hardware whether these contribute to the exceedtng of a safety semees are perfortned by the limit, as defined in the technical specl.

L component suppber or othert fications of a lleense for operation tsaued pursuant to Part 60 of this I 21.4 Interprecations.

(4) A commercial grade item la rwa a

  • 'f[$eviation" means a departure - Except sa spectitcally authorized by (e

Cohlon in Ming, m inter.

part of a be sic component until aNe from the technical requirementa in

pretation of the rneantng of the regu-dedication (see ( 21.3(o-1))

cluded in a procurement docuinent a lations in this part by any officer or (e-1)"Commerc!al grade ttem* meane (see ( 21.3(IH.

Eloyte f th Co lon oth en item that is (1) not subject to design (f) " Director" rnear a an tndlMdual, I h*an u

or specific.ation requirements that are appointed or elected according to law, General Counsel will be recogntzed to unique to facihtles or activities licensed who la authorized to manage and be btnding upon the Comrnhston.

direct the affatts of a corporation.

4 pursuantio Part: 30. 40. 50. co.a i.7(L 7t. oe

$ 72 of this chapter and (2) used La partnership or other entity. In the can of an indiddual proprietonhlp, applications other than facilities or r-e director means the individual,

{ actidties licensed pursuant to Parta 34 b i 21.8 CommuNescons.

40. M 00.aL 70. 71,At 72 of this chapter Except where otherwise specified in and (3) to b, ordered from the

> (g) Discovery means the completion of this part, all wntten commumcations manufacturer /suppher on the basis cf the documentation first identifytng the

, and reports conceming the regulations epecifications set forth in the existence of a devietion or failure to g in this part must be addressed to the enanufacteer's published product comply potentially associated with a x Document Control Desk. U.S Nuclear description (for eaample a catalog).

substantial safety hazard within the e Regulatory Commission. Washmston.

DC 20555 in the case of a heensee. a ev aluation procedares dacussed in y

L l 21.21. (a).

copy must also be sent to the Q

(b) "Cornmtssion" means the Nucle.

appropnate Regional Admmistrator at at Regulatory Commir.sion or its duly the address specified m appendix D to eL authortzer representatives.

(h) Evoluotion means the process of part 20 of this chapter.

?_

determmma whether a particular deviation could create a substantial du" m an he a y ad

[s d

l 21.8 Posting requirementa g,

p c

(a) Each individual. partnership, cor-manufacture. f abncation. placement.

substantial safety hasard.

ere ction. inst alla tion. modification, poration or other entity subject to the N

on mean8 pWc regulations in this part, shall post cut.

inspection, or testing of a facihty or

. c mmunicati n to the NRC Operations rent copies of the follosang dxuments E actmty which is subject to the

$ regulations in this part and consultmg f Center or written trenomittal of in a conspleuous posttion on any prem-information to the NRC Document tses, utthin the United States shere services related to the f acihty or activity e Controt Desk.

the activttles subject to this part are thst ate safety related.

E bl Operating or operation rneans the conducted (1) the regulations in this operation of a facihty or the ronduct of part,(2) Seetton 206 of the Energy Re-a hcensed actmty which is subtect to

- organization Act of 1974, and (3) pro-2 the regulations in this part and cedures adopted pursuant to the regu-

ll te-1) " Dedication" of a commercial consulons services related to operstmns

% lations in this part.

grade item occurs af ter receipt when n

(b) If posting of the regulations in that are safety related that item ts designated for use as a

' thts part or the procedures adopted

';. basic component 7 pursuant to the regulations in this

{

(kl " Procurement document" means part is not practicable. the beensee or a contrset that defines the require-firm subject to the regulations in this ments shich f acilities or baale cornpo-part may. in addition to posting acc-F, (d)" Defect" means:

nenta must meet h order to be conald.

tion 206 post a nottee shich describes (1) A devtation (see 121.3(en in a ered acceptable by the purchuer.

the regulations / procedures, includmg

basic component delivered to a pur.

til " Responsible officer" means the the name of the individual to shor-chaser for use in a factitty or an a,etivt.

president, vice president or other indi-reports may be made, and states the a

! ty subject to the regulations in this vidua! in the organir.ation of a corpo-they may be examined.

part 11. on the basts of an evaluation ration, partnership, or other entity (c) The effective date of this section

  • (see i 21.3(g u. the deviation could aho la vested with executive authority has been deferred untti January 6 create a substantial saf ety hazard or over activities subject to this part.

1978.

125 July 31,1991 21-2

21.21(d)

PART 21 e REPORTING OF DEFECTS AND NONCOMPLIANCE l Ill E s em ptions.

(3) Ensure that a director or has been received should be mede by The Commtaston may, upon applica, responsible officer subject to the cathng the NRC Operations Center, This tion of any tnterested person or upon regulations of this part to informed as peregraph does not apply to intenm f-its own in1ttative, grant such exemp.

soon es prochcabli and. m all cases.

reports descnbed in i 2121[s)(21-l tions f rom the requirements of the withm the 5 workmg days after (ii) Wntten notificahon to the NRC et regulations in this part as it deter-completion of the evolustion desenbed the address opccified in l T15 within 30 mines are authortted by las' and mill in i 2121(a)(1) or i 2121(ell 2)if the des followmg receipt of mformation by not er.danser life or property or the construction or operation of a facil,ty or the director or responsible corporate corrunon deferae ownd security and ue actmty.or a besic component supphed officer under perngraph (e)(3) of this others tse in the public interest.

for such fatility or activity--

section. on the identification of a defect (i) Fails to comply with the Atomic or a f ailure to comply.

?

g Energy Act of19M es amended, or any b pilers of com.mertial grade items ar applicable rule, regulation. order. or PO d

e x e mpt from the prostslons of this license of the Commasion relating to e not be limited to, the following infor-part to the e Alent that they supply Gubstantial tafety heterd. or m
commertial gtade itema.

(ii) Contains a defect.

tD Ns.me and address of the indjvidu-com 1 da ve d by a pl er of to besic componente. or semcas (U) Identification of the f aciuty, the

=

g,,3

% g, g,,,,g,ocag p g associated with besic components, and activity, or the htAlc component sup-the suppher determines that it deres not pUed for such f acility or such activity

,L=

(e) ne Nuclear Regu! story heve the capabthty to perform the within the United Utates which f alla to Commission has submitted the evalusu o to determine lf a defect comply or contains a defect.

trtfortnatico coUection to ulismente contained la this port to e Offics of utsts, then the suppher must inform the (tu) IdentLfication of the firm con-purthesers or effected bcensees within struettns the facility or supplying the Management and Dudget (OMB) for five workm3 days of this determination baste component thlch fe.Lis to comply appros al as requked by the Paperwork so that the purchasers or effected or contains a defect

Reduction Act of 19r10 (44 U S C. 5W1 et 1.censees may evaluate the devtetion or (le) Nature of the defect or f ailure to t seg) OMB has approved th*

failure to c.omply, punuent to l 2121(e) comply and the safety huArd which la M

informenon cutlection requirements (clit) A dir,ciar or responsible officer erested or could be create I 2121 Notmcanon or feiture to compty se

  • f','ihCI C

^3 f acilities or actMties subject to the g

esistence of a detect enc tte evalueuort St. 70,71, or 72 of this chapter and that le) Each individual, corporehon.

is within his or her orgerdzeuen's "g* "h "[

o partnership, or other entity sub ect to responsibility; or timme of the individual or organization i

the regulahons m tha part must adopt (ii) A basic component that is within responsible for the action; and the appropnete procedures to--

his or her organization's responsibility length of time that hu been or will be

11) Evaluate deviehons and failures to and is supplied for a facility of an taken to complete the action.

comply to idenufy defects and failures actaity within the United States that le

( vill) Any advice related to the

~

to cornply associated with substantial subject to the licensing requirements defect or failure to comply about the safety harerde ns soon es prechcable.

under part 30. 40, 50. 00, 61. 70. 71, or 72 f actitty, activity, or bute component and except es provided m peregraph of tha chapter.

that hu been, is being. or till be given (alf 2) of this section. In all cesee within (2) The notification to NRC of a failure to purchasers or lleensees.

60 days of discovery. in order to 6denufy to comply or of a defect under (5) The director or responalble offl.

a reportable defect or failure to comply peregraph (c)(1) of this section and the eer may authortre an indtvidual to that could create a substantial safety evaluation of a failure to comply or a proside the nottfication required by

, heterd, were it to remain uncorrected, defect under peregraphs (e)(1) and (e)(2) thts paragraph. provided that, this sha.1 not tellese the director or re.

Q and of this section, are not required if the 1

A A Ensure met if an evaluahon of an director or responsible officer has actuel eponsible officer of his or her resportst.

it identif ed deviation or failure to comply knowledge that the Commission has bihty under this paragraph.

O potenheny sesociated with a subatentiel been notified in writmg of the defect or (d)Indmduels subiect to tha p rt safety hasard cannot be completed the failure to comply, may be required by the Commmmn to withm 60 den from discovery of the (3) Nohfication required by peregteph supply addihonalinformenon relatni to deviehon or f ailure to comply, en (c)(1) of this sectmn must be made as a defect or failure to compl>

intenm report is prepared and submitted followe--

Commission schon to obt.nn,nt.bhon41 to the Commasion through a director or (6)Inihel notificahon by facsimile.

Informehon may be li.ord on rrports M responsible othcer or designated person which is the preferred method of defects from other trpothnu i nnte es darussrd m i 2121(c)($) The interim notificat on. to the NIAC Operat ons report should describe the deviehon or Center et M1-492-818? or by telephone failure to comply that is being evaluated at M1-9514NO within two day a and should also state when the following receipt of mfortnation by the evaluehon will be completed This director or responsible corporate afhcer mienm report must be submitted in under peregraph (s)(1)of this sechan on wnhrg wdhin 60 days of discovery of the identificahon of a defect er a fedurr the dewehon ur failure to comply to comply Venficahon that the facsimile 126 21 3 July 31 199_t

f-PART 21 o REPORTING OF DEFECTS AND NONCOMPUANCE 2W Psocceruc'er Doctrusarts EnroacturxT

[21J1 Procurement documents.

Each individual, corporstion, part-nership or other entity subject to the '

g3:43 peaure to nocty, regulations in this part sha.11 assure that each procurement dwuroent for a

^87 diC' 0' F"P0^8ADI' 'II'C" subject to the regulebons in this part factljty, or a basic component tasued o

by htrn, her or it on or esiter January 2 who knowingly and cortaciously fetls to 6.1818 speelfles, then applicable, that O provide the notice required by i 21.21

  • the provtalons of 10 CFR Part 21 [ shall be subject to e civil penehy equel E SPDIY-to the amount prosided by section D4 of e

Insrserions. Reconne

l!I.41 laspections.

the AtomicfnerTy Act of1954. ee amended.

Each individual, corporation, part-nership or other entity subject to the regulations in thLs part shall pertnit duly authorized representatives of the I% mM e9 FR 19623)

Commlaston, to inspect -its records, prerntsee activttles, and baalc compo.

nente as necessary to effectuate the purposes of this part.

I y s.tui uam.nene. w % of r e.oe.

(a) Esch indtvidual, corporetion, partnership, or other entity subject to the regulations in this part must prepare end maintain records necessary to accomplish the purposes of this part, specifically-(1) Retain evaluations of all deviations and f ailures to comply for a marumum of five years after the date of the evaluation:

(2) Supphers of basic components must retam any notifications sent to purchasers and effected licensees for e ounimum of five years after the date of

- the notarication.

f (3)Supphers of besic components must retain a record of the purchasers of e

besic components for 10 years after w

3 dehvery of the basic compenent or service associated with a basic component.

(b) Each individual, corporetion, partnership.or other entity subject to the regulations in this part must efford the Commission, et all reasonable times.

the opportunity to inspect records pertaining to basic components that relate to the discovery, evaluation, and reporting of deviations, failures to comply and defects, including any advice given to purchasers or licensees on the placement, erection. Installation.

operelion. maintena nce. modifice tion, or inspection of a basic component.

127 l

July 31,1991 21 4

ENCLOSURE 3 RESPONSE TO BOEING COMPUTER SERVICES LETTER BOEING CUSTOMER ADDRESS INFORMAT. ION The following 3 addresses differed from those you had listed:

ABB Impell Corporation 5000 Executive Parkway San Ramon, CA 94583 (415) 275-4770 Nuclear Power Services One Harmon Plaza Secaucus, NJ 07094 (201) 865-6550 URS/ John A. Blume & Associates 130 Jessie Street San Francisco, CA 94105 (415) 397-2525 The NRC has the same address as Boeing for Reactor Controls, Inc.

The NRC has no information available for Echo Energy Consultants, Inc. in Oakland, California or Nuclear Applications and Systems Analysis Company in Japan.

t 128

,f 4

4' '

r UNITED STATES -

e, 3

1 i

NUCLEAR REGULATORY COMMISSION

'i,, k

,f wassisovos. o c. nossa s,

s.- f JUl.131932 Docket No. 99901250 l

Mr. Michael S. Morris President Bridgeport Testing Laboratory, Inc.

23 Willow Street Bridgeport, Connecticut 06610

Dear Mr. Morris:

SUBJECT:

RESPONSE TO 10 CFR PART 21 INQUIRY By letter dated April 8, 1992, to Mr. Gregory.CWalina, you requested clarification on the NRC's position as described in a

'.ter dated April 26, 1991, to the James C. White Company.

The April 26 letter responded to several Part-21 related inquiries including the need to audit suppliers of calibration services for) measuring and test equipment used on safety-related items.

We have clarified our position, as well as addressed the limited application you described in your letter, in an enclosure to this letter.

Please note that our response is limited to the NRC responsibilities of Title 10 of the Code of Federal Regulations (10 CFR) and does not address the American Society of Mechanical Engineers ( ASME)- code requirements.

Official positions regarding_

compliance with standard industry codes should be obtained from the affected organizations.

Should you have any further questions, please contact Mr. Gregory Cwalina of my staff at (301) 504-2984.

Sincerely, j

W Leif orrholm, Chief Vendor Inspection Branch Division of Reactor Inspection and Safeguards Office of Nuclear Reactor Regulation

Enclosure:

Response to Questions 129

l

}

ENCLOSURE RESPONSE TO BRIDGEPORT TESTING LABORATORY LETTER QUESTION:

What is the requirement for on site audits under the following limited situation:

1)

Service being supplied is a calibration service 2) service is being supplied to equipment located in our promises, and the service is performed on our premises 3)

Supplier has his records concerning traceability to NIST in his possession at the time service is supplied 4)

Supplier has in his possession at the time service is being performed the written procedures to which he operates 5)

Our Quality Control Department monitors his actions during the calibration process to assure that the written procedures are followed 6)

Comprehensive documents supporting the calibration performed (ie Strain charts) are retained at our facility pending receipt _of the written calibration report NRC RESPONSE:

Several of the quality assurance requirements specified-by Appendix B to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR Part 50) relate to calibration services.

Criterion XVIII states, in part, that planned and periodic audits shall be carried out to verify compliance with all aspects of the quality assurance (QA) program.

If the scope of services is limited to calibration of equipacnt on your premises only, a complete audit at the offices of the calibration company may not be necessary to adequately evaluate compliance with their QA program.

A review of their QA manual, written procedures, personnel qualifications, and documentation confirming traceability to the National Institute of Standards and Technology (NIST) may constitute the basis for a sufficient audit.

Additional measures, such as telephone checks, may be necessary to establish the integrity of the traceability documentation.

A further review of the calibration company's internal audits should provide insight into the effectiveness of their overall QA program.

Those activities affecting quality should also be observed in process to assure the written instructions are being followed.

The audit results must be documented and available for review.

Sufficient records must also be maintained supporting each calibration service performed, including documentation that the services conform with the specific requirements as stated on the procurement documents.

130

y

-NRC FORM 333 U.B. NUCLEAR REGULATORY CONDASStON

1. SEPORT NUMBEN -

(*C)

(Assigned by NRC. Add Vol.,

l-W4CM 1102,-

Supo., Rev., and Addandum Num-l am, 3202 -

BIBLIOGRAPHIC OATA SHEET <

    • . H any l NUREG-0040 (See mstruet ons nn the reverse)

. Vol.16, No. 3

[..

. inLE eo suemtE

3. DATE REPORT PVDESD Licensec Contractor and Vendor inspection Status Report yyy November 1992 Quarterly Report July-September 1992
4. Fiw oR or. ANT NuMaEn
6. AulnUR S)
6. TYPE OF REPORT t

-Quarterly

7. PERICO COVERED (incJusive DatJe)

July-September 1992

8. Pt.RFORMING ORGANIZAf TON - NAMLi AND ADDRESS (tf NHC, provice Division, Othee or Repon, U S. Nuclear Regwatory Commmsen, and maeling address; if contractor, provtds name and malling address.)

Division of Reactor inspection and Licensee Performance Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, DC 20555

9. SPONSO&W#J ORGANIZATION - NAME AND ADDRESS U1 NRC, type 'Same as above"; it contractor, provKle NRC Des:en, Ott:ce or Raguri, U.S. Nuclear Regulatory Commisson, and maihng address.)

Same as above

10. 6UPPLEMENT ARY NOTES
11. ABSTRACT (200 words or less)

This periodical covers the results of inspections performed by the NRC's Vendor Inspection Branch that have been distributed to the inspected organizations during the period from July through September 1992.

12. KEY WORDS/DESCRFTORS (Ust words or prirases that will assist researchers in locating the report.)
13. AVAiLABdJTY STATEMENT Unlimited
14. SECURITY Ct ASSIFICATION vendor inspection mis e no Unclassified-(This Report)

Unclassified

16. NUM8th OF PAGES 16 PHICE NRC FORM 335 (2-es)

.= -

F s

)

4 Printed on recycled paper Federal Recycling Program

t I

UNITED STATES sacet rounmetass mTE i

NUCLEAR REGULATORY COMMISSION PosrAos awo ruts pao WASHINGTON, D.C. 20555-0001 PEPWT NO. G-67 OFFtCfAL BUS! NESS PENALTY FOR PR!VATE USE, $300 g g*!1'N

, ^

4 - 3VCE e 1v P 31 f.-T1,n c,

,,, e 93

[3., ) 9 ( C, ', y y g t L,i

- 'i

"; p r ?,

,q55" rots

t;

.ro_-

r, s

u g *; 7,

  • V V

?== 7 1,, p' 4

  • I '

. m g 4II, b

u 4