ML16342C372
| ML16342C372 | |
| Person / Time | |
|---|---|
| Issue date: | 01/14/1992 |
| From: | Matthews S Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML16342B847 | List: |
| References | |
| REF-QA-99900772 NUDOCS 9202030055 | |
| Download: ML16342C372 (60) | |
Text
Enclosure 2
INSPECTION REPORT U.S.
NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION DIVISION OF REACTOR INSPECTION AND SAFEGUARDS Report No.:
Docket No.:
99900772/91-01 99900772 Company:
Industry Activity:
Inspection Conducted:
Inspection Team:
NRC Consultant:
NEI Peebles - Electric Products, Inc.
17045 Euclid Avenue Cleveland, Ohio 44112 NEI Peebles - Electric Products, Inc.
(P-EP) supplies power generators and static exciter voltage regulators for use in emergency power systems.
August 5 through 9, 1991 Steven M. Matthews, Team Leader, NRR Walter P.
- Haass, Sr. Reactor Engineer, NRR Michael R. Snodderly, Reactor Engineer, NRR Kenneth Sullivan, Brookhaven National Laboratory Prepared by:
Steven M. Matthews, Team Leader Reactive'nspection Section 1
Vendor Inspection Branch Approved by:
Inspection Bases:
Uldis Potapovs, Chic Reactive Inspection Section 1
Vendor Inspection Branch 10 CFR Part 21, Appendix B to 10.CFR Part 50, and ANSI N45.2-1971 Date Inspection Scope:
Plants. Affected:
To assess P-EP's compliance with regulatory requirements and licensees'rocurement requirements through a performance-based evaluation of its engineering, procurement, processes, inspections, and tests All licensees with P-EP power generators
)
I g
CONTENTS 1
INSPECTION
SUMMARY
~ae 1
1.1 Nonconformances 1.1.1 Nonconformance 99900772/91-01-01 1.1.2 Nonconformance 99900772/91-01-02 1.1.3 Nonconformance 99900772/91-01-03
/
1.2 Unresolved Item (99900772/91-01-05) 1.1.4 Nonconformance 99900772 91-01-04 2
STATUS OF PREVIOUS INSPECTION FINDING (99900772/86-01-01) 3.
INSPECTION FINDINGS AND OTHER COMMENTS
~
1
~
2
~
2 3
3 3 '
Entrance and Exit Meetings 3
3.2 Background
3.3 Compliance With 10 CFR Part 50, Appendix B
3
, 5 3.4 3.5 APPENDIX Design Control 3.4.1 Design-Bases Documentation 3.4.2 Design Interface 3.4.3 Selection of Critical Items..
3.4.4 Selection and Review for Suitability Instructions, Procedures, and Drawings
~
~
~
~,
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
A
~
7 8
8 9
14
~
~
~
25 A-1
1 INSPECTION
SUMMARY
1.1 onconformances 1.1.1 Nonconformance 99900772/91-01-01 Contrary to.Criterion III, <<Design Control," of Appendix B to Title 10 of the Code of ederal Re ulat'o s, Part 50 (10 CFR Part 50),
and American National Standards Institute (ANSI)
N45.2.11, "Quality Assurance Requirements for the Design of Nuclear Power Plants" (1974),
endorsed in NRC Regulatory Guide (RG) 1.64, "Quality Assurance Requirements for the Design of Nuclear Power Plants" (Revision 2, June 1976), in Section 3, "Design Control," of "Quality Assurance Manual No. 100,<<
(QAM-100), revision dated November 1,
- 1984, NEI Peebles - Electric
- Products, Inc.,
(P-EP) failed to (1) establish adequate measures to control changes in design, materials, and manufacturing processes commensurate with those controls applied to the original design, (2) provide for performing design verifications of the changes in design, materials, and manufacturing processes, (3) demonstrate that the changes in the design were controlled commensurate with the design controls applied to the original
- design, and (4) demonstrate that'he original design basis had been correctly translated into revised specifications,
- drawings, procedures, and instructions (see Section 3.4.1 of this report).
1.1.2 Nonconformance 99900772/91-01-02 Contrary to Criterion III, "Design Control," of Appendix B to 10 CFR Part 50, and ANSI N45.2.11-1974, in Section 3, "Design Control," of QAM-100, P-EP failed to (1) establish adequate measures to control the activities between it and its sister organization, Peebles Electrical Machines (PEM) of Edinburgh, Scotland, that consisted of the review, approval,
- release, distribution, and revision of documents involving their respective design interface, (2) demonstrate that the results of PEM's design translation activities were equivalent to the design requirements specified by P-EP, (3) adequately document the critical requirements or acceptance criteria compared during the equivalency evaluation, and (4) adequately document the results of the equivalency evaluation or other bases to support P-EP's conclusion that PEM's procedures'and specifications were equivalent (see Section 3.4.2 of this report).
1.1.3 Nonconformance 99900772/91-01-03 Contrary to Criterion III, "Design Control," and Criterion VII, "Control of Purchased Material, Equipment, and Services," of Appendix B to 10 CFR Part 50, and ANSI N45.2.11-1974, in Section 3, "Design Control," and Section 7, "Control of Purchased Materials, Equipment, and Service," of QAM-100, P-EP failed to (1) establish adequate measures to provide for the selection and
review for suitability of the application for materials,
- parts, and equipment that were procured as commercial grade items and were essential to the generator's ability to perform its intended design and safety-related
- function, (2) ensure the suitability of the stator coil's resistance temperature detectors, slip rings, adhesives, and mounting sleeve insulator for the slip rings, and (3) ensure the suitability of the materials, parts, and equipment PEM procured (see Section 3.4.4 of this report).
1.1.4 Nonconformance 99900772/91-01-04 Contrary 'to Criterion V, "Instructions, Procedures, and Drawings," of Appendix B to 10 CFR Part 50, in Section 5,
<<Instructions, Procedures, and Drawings," of QAM-100, P-EP failed to establish adequate measures to ensure (1) that activities affecting quality were prescribed by documented instructions, procedures, or drawings; (2) that activities affecting quality were accomplished in accordance with these instructions, procedures, or drawings; and (3) that instructions, procedures, or drawings include appropriate quantitative or 'qualitative acceptance criteria for determining that important activities were satisfactorily accomplished.
P-EP also failed to demonstrate that the activities affecting quality (1) to fit the dovetail rotor pole assemblies to the rotor spider assembly, (2) to perform the brazing required to fabricate the rotor spider
- assembly, and (3) to perform. brazed joint spliced-connections in the field coil winding were documented or accomplished in accordance with instructions, procedures, or drawings that contained quantitative or qualitative acceptance criteria and were equivalent to those specified by P-EP (see Section 3.5 of this report).
1.2 Unresolved Item 9990077 05 P-EP's original quality assurance manual (QAM-100), in effect during the design, manufacture, and test of PGSE's generator, did
'ot include measures to adequately control all of the activities affecting the quality and safety-related function of components and parts.
Although P-EP's second quality assurance manual (QAM-101) superseded QAM-100, it contained several weaknesses that required strengthening before its implementation.
Because the team did not evaluate the implementation of QAM-101, this concern will be evaluated in more detail during a future inspection (see Section 3.3 of this report).
2 STATUS OF PREVIOUS INSPECTION FINDING CLOSED Nonconformance 9900772 86 Contrary to Criterion ZV of Appendix B to 10 CFR Part 50, and to the Parson Peebles Electric Products Inc.
(EPI) Quality Assurance Manual Procedure EQ 2.5.1, EPI did not indicate the applicable drawings, revisions, specifications, or quality requirements on a purchase order (PO) for a safety-related manual voltage regulator.
P-EP's attempts to correct this problem were inadequate as evidenced by Nonconformance 99900772/91-01-03 of this report.
Although a list of clauses for use on POs for nuclear Class 1E materials and components was developed, additional examples of POs for commercial grade materials and parts showed that P-EP continued to procure items that are critical to the generator's ability to perform its safety-related function as commercial grade without specifying the technical and quality requirements.
Therefore, the previous inspection finding (Nonconformance 99900772/86-01-01) will be closed and this issue will be tracked
,.under Nonconformance 99900772/91-01-03.
3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 ntrance and Exit Meetin s During the entrance meeting on August 5, 1991, the NRC's inspection team met with P-EP's staff and discussed the scope of the inspection, outlined areas of concern,. and established working interfaces.
The team observed activities, held discussions with P-EP's staff, and reviewed records and procedures.
The specific areas and documentation reviewed and the team's findings are described in Sections 3.3, through 3.5 of this report.
The persons who participated in and who were contacted during the inspection are listed in the appendix to this report.
During the exit meeting on August 9, 1991, the team summarized the-inspection findings, observations, and concerns with P-EP's management.
P-EP's facility was originally known as Electric Products Incorporated (EPI) and, under various
- names, supplied over 120 power generators to the nuclear industry.
XPI was purchased by
- Portec, Inc. in 1969, and was known as the Electric Products Division of Portec, Inc.
Portec sold the company in 1979 to Parson
- Peebles, a subsidiary of Northern Engineering Industries Limited (NEI) in England.
NEZ is a wholly owned subsidiary of the Industrial Power Group of Rolls-Royce.
The Cleveland facility was known at that time as Parson Peebles Electric
- Products, Znc.
(also EPZ) and was inspected by the NRC on
'0
February 11 through 13, 1986.
The results of that inspection
.'were documented in Inspection Report 99900772/86-01.
After Parson Peebles'urchase of the Cleveland facility, NEI reorganized its Parson Peebles operations under the name of NEI Peebles Limited and the Cleveland facility became NEI Peebles-Electric Products, Inc.
(P-EP).
P-EP's manufacturing facility in Cleveland was closed September 1984 and the power generator work was moved to NEI Peebles Limited's Pilton Works facility in Edinburgh,
- Scotland, known as Peebles Electrical Machines (PEM).
The NRC conducted an inspection of PEM October 6 through 8, 1986, and the results of that inspection were documented in Inspection Report 99901065/86-01.
The organizational structure of NEI Peebles Limited at the time of this inspection was such that the Vice President and General Manager of P-EP reported directly to the Managing Director of PEM.
According to P-EP, most of its U.S. business was spare and replacement parts; about 50 percent of those parts were for nuclear safety-related items.
Since
- 1984, PEM has manufactured the power generators and many of the spare and replacement parts that P-EP has supplied to the nuclear industry.
P-EP recently completed a safety-related power generator for PG&E's new sixth (No. 2-3) emergency diesel generator (EDG) set for its Diablo Canyon Nuclear Power Plant Unit 2 (DCNPP2).
P-EP also was performing the design and procurement activities for a safety-related power generator for Washington Public Power Supply System's Nuclear Project.2 (WNP2), which will be manufactured by PEM.
Although the team focused their inspection activities on the completed power generator for PG&E's DCNPP2, the concerns discussed in this report may have generic implications for WNP2's power generator and any other power generator, or spare and replacement parts, purchased by other licensees.
PG&E's five existing operating power generators (serial 16908022 through 16908026),
installed on DCNPP's EDGs (Nos. 1-1, 1-2, 1-3, 2-1, and 2-2), were procured in 1969 from the Electric Products Division of Portec, Inc.,
and manufactured in the Cleveland facility, described above.
In 1986, PG&E procured a
spare power generator (serial 38604851) from P-EP on the basis that it be identical (i.e., like-'for-like) to'CNPP's five existing operating power generators that were manufactured in 1969 by Portec, Inc.
The 1986 spare power generator was manufactured by PEM in its facility in Edinburgh, Scotland.
The power generator for PG&E's No. 2-3 EDG for DCNPP2 was procured by PO ZS-1539-AB-9, Revision 0, dated January 16, 1990.
PG&E requested P-EP to supply one 4.16-kV, 2600-kW, 3-phase, 60-Hz, 900-rpm, single-bearing, engine-driven, AC synchronous power generator.
The generator was to be supplied as a design Class lE basic component in accordance with PG&E's Engineers
Material Memo (EMM), DC2-3322-BRH-E, Revision 0, dated January 5,:
1990.
In the EMM, PG&E required that the generator be like-for-
'like with DCNPP's 1986 spare generator and DCNPP's five existing operating generators.
PG&E's PO to P-EP also invoked the reporting requirements of 10 CFR Part 21.
PG&E's EMM, Attachment A, "Specification for Supplier's Quality Assurance
- Program, Spec.
No. SP-D-Peebles" (SP-D-Peebles),
Revision 3, dated October 11, 1989, required in Section 4.0, "Quality Assurance Program (Cleveland Facility)," that P-EP's quality program for equipment and components comply with British Standard 5750: Part 1 (IS0-9001-1987),
Part 2, and Part 3; and that P-EP's quality. program for engineering services comply with Appendix'B to 10 CFR Part 50 and ANSI N45.2, "Quality Assurance Program Requirements for Nuclear Power Plants" (1971),
endorsed in NRC RG 1.28, "Quality Assurance Program Requirements (Design and Construction),"
June 1972.
SP-D-Peebles also imposed the requirements of numerous other ANSI standards.
Included among the ANSI standards imposed by SP-6-Peebles was ANSI N45.2.11 (1974),
endorsed by NRC RG 1.64, Revision 2, June 1976.
PG&E's generator was completed and tested by. PEM during January and February 1991.
On March 1,
- 1991, PEM shipped the completed generator to PG&E's contractor, GEC Alsthom of Toronto,
- Canada, for the final assembly and skid-mounting of the EDG set, and the combined testing of the diesel engine, the generator, and the EDG's auxiliary systems.
3.3 Com liance With 0
CFR Part 50 end'x B
P-EP's activities affecting the quality of the generator for DCNPP2 were controlled by its quality program described in QAM-100, originally issued on July 10,
- 1976, and its latest revision issued on November 1,
1984.
QAM-100 was superseded by QAM-101/
issued on February 1,
- 1991, and its Revision 1, issued on July 24, 1991.
P-EP's quality programs described in both QAM<<100 and QAM-101 committed to implementing the requirements of Appendix B to 10 CFR Part 50, ANSI N45.2-1971, and the Canadian Standards Association's "CAN3-Z299.1-1978 Quality Assurance Program Standards."
The QAM-100 program was in,effect during the time that PG&E's generator and its component-parts were designed and procured by P-EP and manufactured and tested by PEM.
QAM-100 and QAM-101 controlled P-EP's design, procurement, fabrication,
- assembly, inspection, tests, corrective actions, and commercial grade dedication activities to produce safety-related generators and spare and replacement component-parts.
The results of the team's evaluation is described in the following paragraphs.
QAM-100 was originally issued when P-EP's facility manufactured electrical components.
The latest revision was intended to address the changes in P-EP's operations that resulted from the closing of the Cleveland manufacturing facility in 1984.
- However, QAM-100 continued to contain obsolete requirements that were not.implemented by P-EP and that were not applicable to P-EP's current, operations.
The program also failed to establish adequate measures to 'control several activities affecting the quality of DCNPP2's generator.
For example, in the QAM-100
- program, P-EP did not describe adequate controls for changes in the design, materials, and manufacturing processes commensurate with those controls applied to the original design for reconciliation of changes in the design, materials, and manufacturing processes to'he original engineering design bases
~
for the revision of design documents involving the design interfaces between P-EP and PEM
~
for review, approval, and design-bases reconciliation of PEM-initiated changes that affected P-EP's design 'bases for the selection and review for suitability of application of materials,
- parts, equipment, processes, and services procured from subsuppliers without adequate quality programs
~
to ensure that the technical and quality requirements for
'aterials, equipment, and services had been, accomplished by source evaluation and selection, objective evidence of quality furnished by the subsupplier, inspection at the subsupplier's
- source, or the examination of products upon delivery Although the QAM-101 program was an improvement over the QAM-100 program, it did not adequately address all of the controls over activities affecting the quality of safety-related generators to the extent consistent with their importance to safety.
In the QAM-101 program, P-EP failed to establish measures to identify and control the design interface between P-EP and'PEM, and to provide procedures for review, approval,
- release, distribution, and revision of design documents involving the design interfaces between P-EP and PEM.
I Although QAM-101 addressed the qualification of commercial grade subsuppliers by establishing documented, acceptable subsupplier performance history data, P-EP failed to ensure that the performance data was directly applicable to the item's critical
characteristics and its intended safety-related application and to verify by audit that the subsuppliers'easures for the control of design, processes'nd material changes were adequately implemented.
Although the'commercial grade dedication program was described in QAM-101, P-EP failed to establish measures to determine (1) the effect of parts on the component's design function, (2) the part's properties or attributes that are essential for the item to perform its design function, and (3) the effect of the part's credible failure mechanism on the component s design function. In
- addition, P-EP failed to ensure that special inspections and tests (the only acceptance method prescribed in QAM-101) will adequately verify the critical characteristics for all parts supplied by commercial grade subsuppliers.
Therefore, P-EP's quality program (QAM-100) in effect for PGGE's generator did not establish measures to adequately control all of the activities that affect the quality of the components and parts that are directly applicable to the generator's ability to perform its intended safety function.
In addition, several weaknesses were identified in the QAM-101 quality program.
This concern is Unresolved Item 99900772/91-01-05 and will be evaluated in more detail during 'a future inspection.
3.4 esi n Control Criterion III, "Design Control," of Appendix B to 10 CFR Part 50, and ANSI N45.2.11 (1974),
as endorsed in RG 1.64, require, in part, that measures shall be established to assure that applicable regulatory requirements and the design bases are correctly translated into specifications,
- drawings, procedures, and instructions and that design changes shall be subject.to design control measures commensurate with those applied to the original design.
Measures also shall be established for the identification and control of design interfaces and for coordination among participating design organizations and shall include the establishment of procedures among participating design organizations for the review, approval,
- release, distribution and revision of documents involving design interfaces and for the selection and review for suitability of application of materials,
- parts, equipment, and processes that are essential to the safety-related functions of the component.
P-EP maintains the overall. engineering and design control responsibility, in addition to providing sales and service support, for the generators and other power generating equipment procured by the nuclear industry.
The team evaluated P-EP's design activities in the four areas described separately below.
3.4.1 Design-Bases Documentation Xn its PO, PG&E required that the generator be like-for-like to the 1986 spare generator and DCNPP's five existing operating power generators (original 1969 design basis).
The team reviewed P-EP's control of the generator's engineering design basis, which would be necessary to establish the like-for-like relationship of the new generator to the design bases of the generators previously supplied.
Specifically, the team reviewed the synergistic effect of the changes that were made to the original engineering design basis, since 1969, to determine what; if any, effect those changes had on PG&E's like-for-like procurement requirement.
P-EP's design-basis reconciliation to the original 1969 design basis consisted of a drawing change review dated June 24, 1991.
P-EP's review'ncompassed the drawings associated with PG&E's generator since 1984, including all revisions.
However, the design control measures of Section 3, "Design Control," of P-EP's QAM-100 did not provide for reconciling changes in the design, materials, and manufacturing processes to the original engineering design basis.
In addition, P-EP's design-basis reconciliation of design changes for PG&E's generator was documented and verified only to 1984 when the Cleveland manufacturing facility closed.
P-EP could not substantiate that the new generator was like-for-like to PG&E's five existing operating power generators.
Therefore, in its design-basis reconciliation, F-EP failed to demonstrate (1) that the established design control measures were commensurate with those applied to the original design and (2) that the original design basis had been correctly translated into revised specifications,
- drawings, procedures, and instructions.
This is Nonconformance 99900772/91-01-01.
3.4.2 Design Interface A significant design interface existed between P-EP and PEM.
Although P-EP maintained the overall responsibility for engineering and design control, PEM's engineering and design organization was completely independent of P-EP's organization and it performed independent design translation activities.
P-EP provided its design drawings, procedures, and material specifications to PEM, and PEM's engineering organization translated them into PEM specifications,
- drawings, procedures, and instructions, including converting dimensions and tolerances from English values to their metric equivalent.
The measures established in Section 3, "Design Control," of P-EP's QAM-100 did not provide for adequate procedures between P-EP and PEM for the review, approval,
- release, distribution, and revision of documents. involving their respective design
,I
interface.
This deficiency appeared to have resulted from the "sister company" relationship of P-EP and PEM, and the daily interface of their respective staffs.
Although PEM issued Departmental Procedure
- DP03A004,
<<Processing of Engineering Change," Revision 0, dated December 17, 1990,- it did not affect P-EP's control of the design inte'rface activities during most of the fabrication and assembly of PG&E's generator.
- Moreover, P-EP failed to establish reasonable assurance that PEM's procedure adequately controlled the design interface activities that were P-EP's responsibility.
Equivalency evaluations of PEM's procedures and material specifications used to fabricate and assemble PG&E's generator were completed by a P-EP's engineering staff in July 1991 and reviewed by P-EP's QA manager in August 1991.
(The, generator was completed by PEM in February 1991.)
P-EP performed these evaluations to ensure that PEM correctly translated 'the design bases into procedures and material specifications.
The equivalency evaluations were not auditable because (1)
PEM's equivalent procedures or material specifications were not always available for comparison to P-EP's procedures or material specifications and (2) the evaluations consisted of only a brief summary of the procedures or material specifications.
P-EP's equivalency evaluations failed to adequately document (1) the critical requirements or acceptance criteria compared during the evaluation and (2) the results of the evaluation or bases to support P-EP's conclusion that the documents were equivalent.
This is Nonconformance 99900772/91-01-02.
3.4.3 Selection of Critical Items PG&E's generator is a complex component, composed of several critical parts that directly affect the ability of the generator to perform its design and safety-related functions (i.e., the credible failure mechanism or long-term degradation of the'art could adversely affect the generator's ability to perform its safety-related function).
PG&E selected and identified the generator s critical items in its PO.
'G&E s
PO (described in Section 3.2 of this report) was modified and issued as Revision 1, February 2,
1990, to add Attachment F, "Critical Items Listing
& Dedication Testing," to its EMM.
Attachment F listed 14 critical items and their associated critical characteristics and required P-EP to verify the PG&E-identified critical characteristics for each of the 14 critical items by performing tests.
PG&E further required that the verification tests to be performed and their respective acceptance criteria be furnished to PG&E for approval before the materials and parts were installed or used.
Revision 2 to PG&E's PO, dated February 22,
- 1990, addressed specific data that P-EP was to provide to enable PG&E to perform the seismic analysis of the generator.
Revision 3 to PG&E's PO, dated February 6,
1991, included significant revisions to PG&E's Engineers Material Memo (EMM)
SP-D-Peebles, Attachment A and.the critical items list of Attachment F.
Attachment A, Revision 5, dated November 15,
- 1990, imposed numerous requirements on P-EP that were not previously imposed'y SP-D-Peebles Revision 3 included in PG&E's original PO.
The most significant additional requirements follow:
~
added section 4.2.6(.1),
the requirements for critical material, parts, or components that were procured as commercial grade items added section 4.2.8; the requirements for the identification and control of materials and items added section 4.2.9, the requirements for a test program to identify and document all testing required to demonstrate that items will perform satisfactorily in service
~
added section 4.2.10, the requirements for the control of measuring and test equipment The EMM's Attachment F,
changed the list of critical items from 14 (shown in Revision
- 1) to 27 (in Revision 3).
Several of the critical characteristics for those items that were to be verified by P-EP also changed.
In other changes imposed by the revision, certain sub-assemblies that were previously identified as critical items were divided into individual parts of the sub-assembly and listed separately (e.g.,
brushes and brush holders was identified as item 7 of Revision 1 and the critical characteristics were identified as.(1) size and shape, and
'(2) final generator test for resistance,
- material, and contact
- pressure, however, Revision 3 listed the brushes and the brush holder separately as item 20 and 19, respectively, and listed configuration as the only critical characteristic for both items).
Table 1, "Cri:tical Items Procured by P-EP,"
on page 15 and Table 2, "Critical Items Procured by PEM," on page 17 of this report provide a comparison of the critical items and their critical characteristics as expressed in Revisions 1 and 3 of the PO.
These changes were the result of discussions between the staffs of PG&E and NEI-Peebles at QA (quality assurance) audit meetings held in Cleveland during December 1989 and in Edinburgh during October 1990.
P-EP's generic failure modes and effects analysis (FMEA) was applicable to all rotating electrical machinery pioduced and was part of P-EP's technical documentation that demonstrated a
generator's compliance with the requirements of the Institute of
Electrical and Electronics Engineers (IEEE) Standard
- 323, "Qualifying Class 1E Equipment for Nuclear Power Generating Stations,"
and IEEE Standard
- 344, "Recommended Practice 'for Seismic Qualification of Class 1E Equipment for Nuclear Power Generating Stations."
The FNEA included the credible failure mode for each individual part.of the generator assembly and assigned a criticality level (see definitions below) to the part, on the basis of the effect of the part's credible failure mode on the ability of the generator to perform its safety-related function.
~evel l catastrophic failure (i.e., will not operate at all, extensive repair needed)
~evel 2 - severely degraded (i.e., operates far off-normal giving warning that a failure will soon occur, extensive repairs needed)
~evel 3 - degraded (i.e., operates off-normal hut with adeguate warning of an impending failure, 'repairs simple if done promptly)
~evel 4
minor degradation (i.e., operates near-normal hut gives a warning of eventual failure, situation very slowly deteiiorates; repairs are simple)
~evel 5
no effect (i.e., part does not affect operation, repairs are part of maintenance)
According to P-EP, PG&E's PO did not impose qualification of the generator'to the requirements of IEEE Standards 323 or 344, and PG&E did not procure P-EP's FMEA documentation for use in the selection of critical items or their critical characteristics.
P-EP also stated that the extent of its involvement in PG&E's selection of critical items or their critical characteristics was limited to only an agreement with PG&E to perform testing necessary to verify the critical characteristics of the critical items identified by PG&E in Revision 1 of its PO.
P-EP reported that it had not been involved in PG&E's selection of the critical items, or their critical characteristics, listed in Revision 3 of PG&E's PO.
Furthermore, PG&E's generator was completed when Revision 3 was issued; therefore P-EP did not consider Revision 3 during its design, procurement, and manufacturing activities.
Because of the minimal involvement of P-EP's engineering organization in PG&E's selection of critical items and their critical characteristics, listed in Revision 1, the team was concerned that PG&E's selected list of critical items may not have been sufficiently comprehensive to ensure that all items were included, specifically, those items with a credible failure mode or that, in a degraded condition, could adversely affect the generators ability to perform its design and safety-related 11
function.
The team reviewed P-EP's generic FMEA and discussed the technical bases for the critical items and their critical characteristics with P-EP's engineering staff to determine whether PG&E's Revision 1 list of 14 critical items, or its Revision 3 list of 27 critical items included ~l parts that are critical to the generator's ability to perform its design and safety-related function.
According to P-EP's FMEA, the generator's two major design parameters with regard to the effects of long-term degradation and cyclic fatigue were (1) its operating temperatures and (2) cyclic loading or high vibration forces.
On the basis of these design parameters, FMEA criticality levels 1 or 2 were assigned to critical items such as the stator windings, leads and their connections, rotor pole windings, roller bearings, rotor shaft, coil supports, and slip rings.
P-EP's generic FMEA documentation indicated that PG&E's lists of critical items did not adequately envelope all of the generator's critical parts having a design or safety-related function.
Two examples are discussed below.
(1) in ou t'leeve su ator The generator was designed with a brush and slip-ring assembly to carry DC excitation voltage to the field coils mounted on the rotor shaft.
The slip-ring assembly was concentrically mounted on the rotor shaft.
P-EP incorporated a slip-ring mounting sleeve insulator in its design to prevent establishing a current path to ground between the slip-ring assembly and the rotor shaft.
The mounting sleeve insulator consisted of a tube of insulating material, with approximately 0.25-inch wall thickness, installed between the shaft and the slip-ring assembly.
P-EP's generic FMEA documentation indicated that, if the mounting sleeve insulator between the slip-ring assembly and the rotor shaft failed, DC excitation voltage would be lost and result in catastrophic failure of the generator.
In
- addition, a short-to-ground failure in the insulator could occur from wear or erosion, establishing a current path to ground.
(2) em erasure and V'b at'on d'cat'n ev'ces P-EP provided six resistance temperature detectors (RTDs) to monitor the generator's stator coil operating temperatures and provide' conservative indication of the generator's overall temperature.
-The operating temperatures of the generator, including localized thermal stresses, affect the stability of the insulation and adhesive materials (e.g.,
thermal breakdown, aging, fatigue, and wear) which directly affect the fragility of unisotropic structures (e,g., rotor 12
.windings) duping the installed life of the generator.
A limit of 105 C rise over an ambient temperature of 40 C for the maximum generator operating temperature (145 C) was established in P-EP's design basis.
Although the RTDs were included in PG&E's Revision 1 list of ciitical items, they were not included in the Revision' list of critical items.
RTDs were not provided to monitor the temperature of the shaft's single roller bearing, even though the roller bearing and its operating temperature were identified in P-EP's Ft1EA as critical items.
Zn addition to high temperature, fatigue from cyclic loading or high vibration forces on the generator may also directly affect the performance and reliability of the single roller bearing.
The roller bearing may be subjected to cyclic loading or high vibration forces caused by an unbalanced rotor shaft, the diesel 'engine with its crankshaft directly connected to the generator's rotor shaft, and other sources from the skid-mounted EDG assembly.
During the installed life of the generator, subtle damage to the generator may occur from short-to-ground or asynchronous events (e.g., paralleling the generator out-of-phase) that cause significant forces on the stator coils and rotor pole windings.
P-EP indicated that PGEE's generator was designed
. to withstand short-to-ground events that produce magnetic forces on-the stator coils, which were mechanically supported by the stator frame's welded structure.
The end sections of the stator coils, however, were installed in a cantilevered arrangement with stiffeners to support the coils and prevent or minimize their distortion.
An asynchronous event may produce centrifugal forces on the rotor pole windings of such magnitude to cause separation of
,the windings and an unbalanced rotor shaft.
P-EP stated that the generator was not constructed to withstand an asynchronous event.
- However, PGRE did not identify vibration indicating devices as critical items.
For a complex assembly such as a generator, the selection of critical items and the determination of their critical characteristics would require the involvement of both the licensee's and supplier's engineering staffs.
P-EP considered this interface activity to be limited for those critical items
'identified in Revision 1 of-PGEE's PO, and it believed the interface activity was nonexistent for the critical items identified in Revision 3 of the PO.
Furthermore, P-EP had completed PGEE's generator when Revision 3 was issued; therefore,'evision 3 was not'considered during the design, procurement, and manufacturing activities of the generator.
13
3.4.4 Selection and Review for Suitability P-EP or PEM procured each of the critical items identified in Revisions 1 and 3 of PG&E's PO as commercial grade items.
The critical items procured by P-EP and PEM are identified in Tables 1 and 2 of this report'.
P-EP procured 7 of the 14 items listed in Revision 1 of PG&E's PO (or 10 of the 27 items listed in Revision 3),
and supplied them to PEM for installation in the generator assembly.
P-EP's procurement practice consisted of purchasing items from subsuppliers that were selected on the basis of their performance history, which was 'determined through the general knowledge and experience of P-EP's staff.
The
. performance history data that was documented and verified during the manufacture of PG&E's generator did not establish an adequate basis for the qualification of the subsuppliers of critical items; Most of P-EP's.subsuppliers were not audited to verify that their measures to control design, processes, and material changes were adequately implemented.
Therefore, the critical
-=items procured by P-EP for PG&E's generator were procured as commercial grade items.
The POs for these items did not impose any quality requirements or the reporting requirements of 10 CFR Part 21 on the subsupplier.
PEM procured the 7 remaining critical items listed in Revision 1
of PG&E's PO (or 17 of the 27 items listed in Revision
- 3) from its subsuppliers in Europe.
PEM was qualified as a subsupplier to P-EP through P-EP's audits of PEM dated September 30,
- 1985, and August 7 through 9, 1989.
P-EP stated that its audits qualified PEM to supply components and parts produced to a quality program equivalent to the requirements of Appendix B to 10 CFR Part 50 and in compliance with the reporting requirements of 10 CFR Part 21.
- However, P-EP's reports for the 1985 and 1989 audits of PEM did not document adequate objective evidence to substantiate whether PEM's quality program was adequate to perform commercial grade dedication of critical items.
P-EP also failed to demonstrate that PEM's dedication activities for critical items that it procured as commercial grade resulted in establishing reasonable assurance that the generator and its critical items will perform their respective design and safety-related functions.
PEM's commercial grade dedication activities will be reviewed by an NRC team in a separate inspection.
This report only documents the results of the team's evaluation (1) of P-EP's procurement and commercial grade dedication activities for the items it procured and supplied to PEM and (2) of P-EP's responsibility for establishing reasonable assurance that PEM's commercial grade dedication activities were adequate to ensure that the critical items will perform their design and safety-related function.
14
Table 1 - Critical Items Procured by P-EP C 'ica 1 tems:
G&E's 0
~ev is on':
arac er'sties Insulators (5-kV in terminal box)
Rev.
No.
1 Item-1 Rev.
No.
3 Item-22
~ Dielectric strength
~ Size and weight
~ Dielectric strength
~ Configuration Insulating bushings (lead wires through motor case)
Rev.
No.
1 Item-3 Rev.
No.
3 Item-24
~ Size and shape
~ Configuration Insulating material (sheets,
- tape,
& rings)
Rev.
No.
1 Item-5 Rev.
No.
3 Item-26
~ Thickness
~ Thickness Bearing seals (felt)
Rev.
No.
1 Item-6 J
~ Thickness and shape
~ Texture Rev.
No.
3
. Item-23
~ Configuration
~ Texture Brushes and Brush Holders Brushes Brush Holder Rev.
No.
1 Item-7 Rev.
No.
3 Item-20 Rev.
No.
3 Item-19
~ Size and shape
~ Final generator test:
resistance,
- material, and contact pressure
, ~ Configuration
~ Configuration
II
Table 1 continued Cr't'cal tems:
~tea C aracter's Stator resistance
'temperature'etectors (RTDs)
Rev.
No.
1 Item-8
~ Shape and size
~ Shop test:
continuity, resistance, and insulation Current transformer and test switch Current transformer Current transformer test switch Rev.
No.
1 Item-9 Rev.
No.
3 Item-21 Rev.
No.
3 Item-25
~ Size and weight
~ Dielectric strength
~ Continuity
~ Configuration
~ Mounting
~ Insulation
~ Resistance
~ Continuity
~ Configuration
. ~ Dielectric strength
~ Continuity Slip rings Rev.
No.
3 Item-17
~ Configuration
~ Material Adhesives Rev.
No.
3 Item-27
~ Material
Table 2 - Critical Items Procured by PEM C 't'cal Items:
Lead wire 8
"'ev.
No.
1 Item-1 Rev.
No.
3 Item-16
~ritical haracte ist'cs:
~ Dielectric strength' Number of strands Marking on cable Insulation thickness
~ Configuration Magnet wire Rev.
No.
1 Item-4 Rev.
No.
3 Item-3
~ Size and shape
~ Resistance
~ Insulation
~ Dielectric strength
~ Material
~ Insulation
~ Dielectric strength Copper bus (in terminal box)
Rev.
No.
1 Item-10
~ Size
~ Resistance
~ Silver plating Lead to coil termination Rev.
No.
1 Item-11 o Brazing
~ Weld materials Roller bearing Rev.
No.
1 Item-12
~ Size/type
~ Visual inspection
~ Catalog number
~ Tolerances Rev.
No.
3 Item-6
~
~ Part number
~ Configuration 17
~
s
Table 2 continued r't'cal Items:
Shaft/casting G&E's PO Pevisiio Rev.
No.
1 Item-13 Rev.
No.
3 Item-1 Crrtical a acte 'st'cs:
~
PEM test
~ Material
~ Configuration
~ Integrity Stator and Rotor core Stator coils Rev.
No.
1 Item-14 Rev.
No.
3 Item-15
~
PEM test (losses)
~ Configuration e Chemical composition
~ Coating insulation Stampings Rev.
No.
3 Item-2
~ Configuration
~ Material Bearing bracket Rev.
No.
3 Item-4
~ Configuration
~ Material Stud/threaded rod Rev.
No.
3 Item-5
~ Dimensions
~ Material
~ Welding Spider end rings Rev.
No.
3 Item-7
~ Configuration Pole end rings Rev.
No.
3 Item-8
~ Configuration,
~ Material Short circuit bars Rev.
No.
3 Item-9
~ Configuration
~ Material Pole head Rev.
No.
3 Item-10
~ Configuration 18
9 ~
I
~ ~
~
Table 2 continued C 'ica1
- tems:
~evis'on:
ttac me Q~t'.ca 1 Tapered keys, Rev.
No.
3 Item-11
~ Configuration
~ Material
~ Hardness Rotor wedge Rev.
No.
3 Item-12
~ Material Rivets Rev.
No.
3 Item-13
~ Configuration Insulating washers Rev.
No.
3 Item-14
~ Configuration
~ Material
~ Dielectric strength Stator frame Rev.
No.
3 Item-18
~ Configuration P-EP's commercial grade dedication program was governed by Procedure DED-100, implemented on August 2, 1991.
The program was not in effect during the procurement and commercial grade dedication of the critical items supplied to PEM for use in PG&E's generator.
P-EP considered its standard material receiving activities adequate to dedicate commercial grade items, on the basis of its understanding of commercial grade dedication requirements that existed before P-EP's development and implementation of DED-100.
The commercial grade dedication activities performed by P-EP for the items procured and supplied to PEM for PG&E's generator were, therefore, not controlled by documented instructions or procedures.
Although P-EP agreed to perform the testing necessary to verify the critical characteristics of the items identified in Revision 1 of PG&E's PO as critical, P-EP did not (1) identify the items critical to the generator's ability to perform its intended safety-related function qr (2) perform a technical evaluation of the items identified in Revision 1 of PG&E's PO to determine the 19
~
C ~
adequacy of PG&E's list of critical characteristics.
For the critical-characteristics selected by PG&E, P-EP failed to demonstrate their relevance (1) to the properties or attributes of the item necessary to withstand the effects of long-term degradation,.
(2) to the credible failure mode of the item, and (3) to the ability of the item to perform its safety-related function.
P-EP failed to substantiate that PG&E's list of critical it'ems included ~l parts that are required for the generator to perform its safety-related function and that PG&E's list of critical characteristics were adequate to that the item will perform its safety-related function.
Consequently,,an evaluation of P-EP's generic FMEA identified additional critical characteristics for certain items that weie not identified or.
verified by P-EP during its commercial grade dedication activities and were not identified by PG&E in its Revision 1 to the PO.
These characteristics are described, as applicable, in the summary of P-EP's commercial grade dedication and verification activities for the items it procured and supplied to
- PEM, as given below.
The four 5-kV insulators were installed in the terminal box.
PG&E identified their critical characteristics as dielectric
- strength, size, and weight in Revision 1 of its 'PO, and dielectric strength and configuration in Revision 3..
The acceptance criterion for the insulators dielectric strength was not obtained from their supplier.
P-EP verified the insulator's weight and dimensions, including length,
'outside-diameter (OD), and bolt hole center location.
P-EP found the results of its verifi,cation activities acceptable, even though it did not obtain the supplier's certification of dielectric strength or perform the test to verify the dielectric strength.
(2) nsulatin Bushin s (Revisions 1 and 3)
The insulating bushings were installed in the lead wire penetration through the generator housing.
PG&E identified the critical characteristics as size and shape in Revision 1
of its PO and configuration in Revision 3.
P-EP verified the bushing's dimensions, including the thread OD, length, bushing OD, inside-diameter (ID), and the overall length.
P-EP found the results of its verification activities acceptable.
- However, P-EP did not verify their dielectric strength or concentricity, which were not identified as described above.
20
I
(3) nsulatin Materials (Revisions 1 and 3)
The insulating materials were installed on the stator's so called "diamond" (shape) coil windings.
PG&E identified thickness as the critical characteristic.
The insulation materials procured and the amount sampled by P-EP consisted of
~
Mica paper tape,.60 rolls;
~
Mica paper tape, 60 rolls.
~
B-Stage mica paper tape, 162 rolls
~
B-Stage mica wrapper, 4 rolls.
.6 rolls sampled
.6 rolls sampled 17 rolls sampled
.4 rolls sampled P-EP verified the thickness of the insulating material on the rolls sampled.
P-EP found the results of -its verification activities acceptable.
- However, P-EP did not verify (1) the batch or lot homogeneity of the insulation material to ensure that each batch or lot was sampled and traceable to each batch or lot and (2) the material constituents of the insulating materials or their properties or attributes with regard to the generator's design-basis operating temperature requirements.
The bearing seal is a felt disc installed on the rotor shaft.
PG&E identified the critical characteristics as thickness,
- shape, and texture in Revision 1 of its PO and as configuration and texture in Revision 3.
P-EP verified the seal's dimensions, including ODg ID, thickness, and shape.
According to P-EP, the determination of acceptability of the felt bearing seal's texture was a matter of judgment.
P-EP found the results of its verification activities acceptable.
- However, P-EP did not consider that the different "weights" of oil-seal felt that have different porosity and lubricant holding properties.
(5) rushes and Brush Holders (Revision 1)
The brushes and brush holders were installed on a brush holder stud and positioned above the slip-rings on the rotor shaft.
PG&E identified the critical characteristics as
- size, shape, final generator test to verify resistance,
- material, and contact pressure.
~ushes (Revision 3)
The brushes and their wire leads and terminal connections were installed in the brush holders.
PG&E identified the critical characteristic as configuration.
P-EP verified the brushes'imensions, including height, length, and width.
P-EP found the results of its verification activities 21
e ~
~
t
~
acceptable.
P-EP did not, however, verify the brushes'1) material constituents, (2) wire lead size or type, (3) wire lead terminal connections, and (4) electrical resistance, which were not identified as desciibed above.
PG&E identified the critical characteristic as configuration.
P-EP verified the brush holders'imensions, including the opening size for the brush and its overall shape.
P-EP found the results of its verification activities acceptable.
P-EP did not, however, verify (1) the spring tension on the brushes or (2) the technical and quality requirements, or the critical characteristics of the Grade X Spaudite Bakelite cylinder (bushing) that fits over the brush holder stud and functions as the insulator for electrical separation between the brush holder and the generator frame, which were not identified as described above.
(6)
Stator Resistance em e ature Detectors RTDs (Revision 1)
The stator RTDs were installed in the stator coil assembly.
PG&E identified the critical characteristics as shape,
- size, shop test for continuity, resistance, and insulation.
Although continuity and resistance were listed separately as characteristics for several items without further explanation, the team noted that if a quantitative value for resistance was desired then continuity would also be demonstrated without performing a separate test.
Specifying continuity in addition to resistance would, therefore, normally be considered redundant.
PG&E's critical characteristics,
- however, did not specify (1) the temperature at which the shop test for resistance was to be conducted or (2) the linearity requirements over the test range.
(See the description of temperature and vibration indicating devices, on page 12 of this report.)
P-EP failed to demonstrate documented dedication and verification activities for the commercial grade stator RTDs.
(7) urrent ransformers and est Sw tch (Revision 1)
The current transformers and test switch'orm a sensing device.
PG&E identified the critical characteristics as size, weight, dielectric strength, and continuity.
Current ansformers (Revision 3)
PG&E identified the critical characteristics of current transformers as configuration, mounting, insulation, resistance, and continuity.
P-EP verified the current transformers dimensions, including the height, the length 22
~
~
k
measured at the feet, the length of the body, location of mounting holes, weight, insulation resistance, and continuity.
The acceptance criterion for the dielectric strength of the current transformers was not obtained from the manufacturer.
P-EP found the results of its verification activities acceptable.
- However, P-EP's, verification activities did not consider (1) the electrical loads supplied by the current transform rs or whether the current transformers supplied a current to the static exciter voltage regulator or instrumentation and protective circuits and (2) the ratio of the primary to secondary
- currents, which were not identified as described above.
est Switch for the Cur ent ransfo ers (Revision 3)-
PG&E identified the critic'al characteristics of the test switch for the current transformers as configuration, dielectric strength, and continuity.
P-EP verified the test switch s dimensions, including the cover size, location of mounting holes, weight, dielectric strength, and continuity.
P-EP found the results of its verification activities acceptable, even though it did not obtain the supplier's certification of dielectric strength or perform the test to verify the dielectric strength.
(9)
Sli -Rin Assembl (Revision 3)
The slip-ring assembly was installed on the rotor shaft.
PG&E'identified the critical characteristics as configuration and materials.
- However, P-EP failed to demonstrate documented dedication and verification activities for the commercial grade slip-ring assembly.
The epoxy adhesive (resin) was applied during the forming of the rotor pole windings.
PG&E ident'ified the critical characteristic as material.
P-EP Shop Order S-1128 required the use of epoxy resin instead of a polyester resin (polyester resin was used for PG&E's five existing EDGs) because an environmental qualification report showed that the performance characteristics of epoxy resin were acceptable and it was an acceptable substitute for the polyester resin.
P-EP, however, did not establish similarity of the commercial grade epoxy resin purchased to the epoxy resin described in the environmental qualification report.
P-EP also failed to demonstrate documented dedication and verification activities for the commercial grade epoxy resin.
r
~ I k
(11) li -Rin Mountin Sleeve su ator (not identified)
The slip-ring mounting sleeve insulator was installed between the shaft and the slip-ring assembly and provided not only the electrical separation of the slip-ring assembly and the rotor shaft, but also formed'the mounting structure for the slip-ring assembly.
PG&E did not identify the slip-ring mounting sleeve insulator as a critical item.
(See the description of slip-ring mounting sleeve insulator, on page
-12 of this report.)
P-EP's material routing incoming order review of April 12,
- 1990, showed that P-EP supplied the slip-ring mounting sleeve insulator "to PEM as a commercial grade stock item.
Additionally, P-EP Drawing No. A-29412, "Slip Ring Mounting Sleeve Insulator," Revision 3, dated December 20,
- 1967, showed an obsolete material specification for the sleeve insulator.
P-EP stated it would, update the drawing.
P-EP failed to demonstrate documented dedication and verification activities for the commercial grade slip-ring mounting sleeve insulator.
(12) Vibration I dicatin Dev'ces (not identified)
The vibration indicating device would be used to detect high vibration resulting from various sources, including an asynchronous event.
(See the description of temperature and vibration indicating devices on page 12 of this report.)
- However, P-EP failed to include vibration indicating devices in its evaluation of critical items.
P-EP supplied the generator to PG&E as a basic component that complied with the quality requirements of Appendix B to 10 CFR Part 50; therefore, P-EP was responsible for establishing reasonable assurance that the generator and its critical items will perform their respective design and safety-related
.functions.
Although PG&E selected and specified the critical items and their critical characteristics for its generator, P-EP agreed to perform the tests necessary to verify PG&E-specified critical characteristics.
P-EP did not demonstrate that the critical characteristics, specified by PG&E, were relevant to the
~
critical item s (1) design characteristics, (2) credible failure
- modes, (3) ability to perform its safety-related
- function, and (4) properties or attributes necessary to withstand the effects of long-term degradation and cyclic fatigue.
P-EP also failed to demonstrate that PEM s dedication activities, for critical items procured by PEM as commercial grade, resulted in establishing reasonable assurance that the generator and its critical items will perform their respective design and safety-related functions.
24
0
~
~
~
P-EP failed to demonstrate reasonable assurance that the technical bases for the critical items and their critical, characteristics chosen by PG&E and verified by P-EP during the commercial grade dedication and verification activities adequately (1) ensured that the critical items and the generator will perform their safety-related function and (2) ensured that the critical items have the properties or attributes necessary to withstand the effects of long-term degradation or cyclic fatigue.
This is Nonconformance 99900772/91-01-03.
3.5 st uc 'o s oc u es d
w Criterion V, "Instructions, Procedures, and Drawings," of Appendix B to 10 CFR Part 50, require; in part, that measures be established to ensure that activities affecting quality be prescribed by doc'umented instructions, procedures, or drawings; that activities affecting quality be accomplished in accordance.
with these instructions, procedures, and drawings; and that instructions, procedures, and drawings include quantitative or qualitative acceptance criteria for determining that important activities have been satisfactorily accomplished.
The most significant concerns identified during this review are discussed below.
P-EP's facsimile transmittal to PEM, dated February 11,
- 1987, provided the instructions for fitting the dovetail rotor pole assemblies to the rotor spider assembly, even though P-EP did not have a documented procedure that prescribed this activity.
PEM incorporated these instructions into Engineering Standard R-6097, "Assembly Procedure for Wound Rotors of Class 1E Generators Having Dovetail Poles."
P-EP did not approve PEM's procedure or perform an equivalency evaluation because it did not have a
documented procedure to compare to PEM's procedure.
P-EP Drawing C-66827, "Rotor Pole Assembly," Revision 2, specified the use of Brazing Specification EB-4.4 for the fabricat'ion of the rotor pole'stampings.
P-EP did not perform an evaluation to determine the equivalency of PEM's Brazing Procedure R-6092, "Preparation 'and Procedure for Brazing Copper/Copper Alloy Rotor Bars to Short Circuiting Rings for Use in Normal Industrial Environments," Issue 1, Revision 0, dated December 7,
1990.
In addition, P-EP did not perform an equivalency evaluation of Peebles Power Transformers Procedure
- 5275, "Process Specification Responsible Department Fabrication,"
dated March 26, 1987.
PEM. used this procedure as a broad-based procedure that allowed the user to choose between several welding and brazing processes and joint geometries.
PGRE's generator was designed with eight field coils mounted on the rotor.
Each field coil consisted of 404 turns of magnet wire that were wound on a laminated steel, rotor pole core with high permeability.
The application of a DC excitation voltage, 25
(E s
~
~
supplied from the brush and slip-ring assembly, will cause the field coils to generate a magnetic field, and in combination with the rotation of the rotor shaft, this field generates the output voltage.
An electrically shorted, or open field coil winding, may result in the failure of the generator to'erform its intended design and safety-related function.
The field coil windings are also subjected to centrifugal forces from the rotation of the shaft and the resulting mechanical stresses that may affect the integrity of the field coil windings.
PEM's Manufacturing Procedure R-6096, "Manufacturing Procedure for Strip-On-Flat Field Coils Wound Directly Onto Laminated Poles for use In Class lE Generators,"
Revision 0, dated January 4,
1991; stated, in part, that spliced joints in the magnet wire were permissible where a continuous length of magnet wire was not available during field coil (rotor pole) fabrication.
In the event, that the amount of magnet wire available on' single spool was not sufficient to complete the coil winding operation, or where the magnet wire was damaged or broken during the manufacturing process, PEM's Procedure R-6096 permitted making a brazed joint spliced-connection in the field coil winding. If the fabrication and brazing of such a joint was not adequately controlled by procedural guidance and proper quality techniques, the results may be (1) a mechanically weak spliced-connection or (2) a high electrical resistance at the brazed joint, which may not be readily detectable after completing the field coil winding.
P-EP's Production Specification R-6028, dated May 1968, provided the engineering guidance to perform a resistance brazed spliced-connection of magnet wire during the fabrication (i.e., winding) of the field coils.
In addition to providing procedural guidance, this specification contained precautionary comments that (1) silfos (a brazing rod material) is brittle and considered to be only half as good as an electrical conductor as that of copper and (2) the phosphorous in the brazing material will bubble, if over heated, resulting in a mechanically weak spliced-connection with high resistance.
A high-resistance spliced-connection may cause a thermal "hot spot," leading to an electrical short within the field coil winding.
According to P-EP's generic.FMEA document, such an occurrence may lead to the generator s failure.
The high resistance would be localized in a single winding.
Therefore, any change in total field coil resistance may be masked, and not readily detectable, when the field coil was completely assembled.
Where the brazing operation resulted in a mechanically weak spliced-connection, mechanical
- stress, induced by centrifugal forces and/or vibration, may also cause a short or open circuit in the affected field coil winding.
26
e
~
~
P-EP stated that,'o its knowledge, no spliced-connections were made during the fabrication of the field coils and produced a
certificate of conformance that indicated that a sufficient quantity of magnet wire per spool was ordered for each field coil
, assembly.
The certificate of conformance, however, did not establi'sh reasonable assurance that PEM.had not made spliced-connections as a result of damage to the magnet wire during the winding process, and P-EP did not demonstrate documented verification that PEM-did not perform spliced-connections.
P-EP did advise the team that spliced-connections may be necessary for the field coil windings for WNP2's generator.
PEM Manufacturing Procedure R-6096 did not produce the guidance or the precautionary statements contained in P-EP Production Specification R-6028.
P-EP did not perform an equivalency evaluation of PEM's procedure.
PEM's'manufacturing procedure did not include quantitative or qualitative acceptance criteria for spliced connections such as resistance measurements
.and tensile strength tests following the brazed joint splicing operation.-
These instances were the result of P-EP's QAM-100 failure to establish adequate measures to ensure (1) that all of the activities affecting quality were prescribed by documented instructions, procedures, or drawings and were accomplished in accordance with these instructions, procedures, and drawings, and (2) that the instruction's, procedures, and drawings include quantitative or qualitative acceptance criteria for determining that.important activities were satisfactorily accomplished.
P-EP also failed to demonstrate that the activities affecting quality (1) to fit the dovetail rotor pole assemblies to the rotor spider
- assembly, (2) to perform the brazing required to fabricate the rotor spider assembly, and (3) to perform brazed joint spliced-connections in the field coil winding were documented or accomplished in accordance with instructions, procedures, or drawings that contained quantitative or qualitative acceptance criteria and weie equivalent to those specified by P-EP.
This is Nonconformance 99900772/91-01-04.
27
l(
C C
L 4
APPENDIX A PERSONS CONTACTED The U.S. Nuclear Regulatory Commission staff participating in the evaluation of NEI Peebles - Electric Products, Inc. with regard to its design, procurement, commercial grade dedication, and manufacture of a power generator for Pacific Gas and Electric Company's Diablo Canyon Nuclear Power Plant, Unit 2 and the persons contacted during-the inspection are listed below.
E eebles -
lect 'c roducts c ~
~ *
~ *
- Clasen, Robert C.
Marino, Frank D.
Mossbrugger, Charles J.
Politi, Ron B.
- Rossman, Richard A.
Senior Design Engineer Manager, Quality Assurance
- Manager, Engineering Vice President and General Manager Manager,- Materials U.S. Nuclear e ulator Commi ss on
- Haass, Walter P.
~ *
- Matthews, Steven M.
~
- Snodderly, Michael R.
~
Sullivan, Kenneth Senior Reactor Engineer, Special Projects Section, Vendor Inspection Branch (VIB), Division of Reactor Inspection and Safeguard (DRIS), Office of Nuclear Reactor Regulation (NRR)
Team Leader, Reactive Inspection
'ection 1 (RIS1), VIB/DRIS/NRR Reactor Engineer, RIS1/VIB/DRIS/NRR NRC Consultant, Brookhaven National Laboratory
~ ~ Attended the Entrance Meeting
- ~ Attended the Exit Meeting
k
~
l