ML20149H980
| ML20149H980 | |
| Person / Time | |
|---|---|
| Issue date: | 07/22/1997 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20149H948 | List: |
| References | |
| REF-QA-99900358, REF-QA-99900359 99900359-97-01, 99900359-97-1, NUDOCS 9707250275 | |
| Download: ML20149H980 (18) | |
Text
..
U.S. NUCLEAR REGULATORY COMMISSION 0FFICE OF NUCLEAR REACTOR REGULATION ORGANIZATION:
Yuasa Exide, Inc. (YEI) 2366 Bernville Road, Post Office Box 14145, Reading, Pennsylvania 19612
. DOCKET / REPORT N0(s).:
99900358/97-01 (Reading, Pennsylvania) 99900359/97-01 (Richmond, Kentucky)
ORGANIZATION CONTACT:
Richard P. Bender, Vice President, Quality Assurance and Purchasing (610)208-1972 Lana West, Quality Assurance Manager (606)624-7346 l
NUCLEAR INDUSTRY:
Manufactures and supplies stationary batteries and ACTIVITY battery racks.
l j
INSPECTION Reading, Pennsylvania, March 4-7, 1997 CONDUCTED:
Richmond, Kentucky, April 28-May 2, 1997 TEAM MEMBERS:
K.R. Naidu, Team Leader, NRR J.J. Petrosino, NRR S.D. Alexander, NRR S.N. Saba, NRR Approved By:
G.C. Cwalina, Chief Vendor Inspection Section Special Inspection Branch Division of Inspection and Support Programs Office of Nuclear Reactor Regulation 9707250275 970722 PDR GA999 EMVYUASA 99900358 PDR y7
1 INSPECTION
SUMMARY
The purpose of this inspection was to evaluate the quality assurance (QA) program and its implementation in the design, qualification and manufacture of rectangular, vertical plate, lead-acid battery cells (Type "GN") by Yuasa-Exide, Inc. (YEI), used in Class IE/ vital station batteries at nuclear power plants.
Conducted at YEI's engineering facility in Reading, Pennsylvania, and its factory in Richmond, Kentucky, the inspection focused on: (1) the implementation of the manufacturer's process controls, (2) procedural adequacy 4
(including consistency with established requirements) and procedural compliance, (3) procurement and acceptance of purchased parts and materials used in battery manufacture including commercial-grade dedication of components and parts for resa(le as basic components, e.g., battery rack replacement parts), and (4 certificates of conformance) purchase orders (P0s) from NRC licensees and licensees.
(C0Cs) and associated documents provided to NRC Inspection bases were:
Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Processing Plants," to Part 50 of Title 10 of the Code of Federal Requlationi (10 CFR Part 50, Appendix B) 10 CFR Part 21, " Reporting Defects and Noncompliance" The inspectors identified two minor violations of 10 CFR Part 21 (621.21(a) and 521.6) (See Section 3.1); one Level IV violation of 10 CFR 21.21(b) (See Section 3.2); one nonconformance with respect to Criterion XV of 10 CFR Part 50, Appendix B (See Section 3.5); nonconformances with respect to Criterion III of 10 CFR Part 50, Appendix B (See Section 3.7); nonconformances with respect to Criterion VII of 10 CFR Part 50, Appendix B (Section 3.7); and one inspector followup item (Section 3.6).
2 STATUS OF PREVIOUS INSPECTION FINDINGS There have been no NRC inspections conducted since Exide became Yuasa-Exide, Inc. (YEI) and after the company reorganized under new management.
3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 10 CFR Part 21 Procedure and Postina Eq_op_e a.
q The inspectors reviewed YEI's quality assurance procedure (QAP) for reporting in accordance with 10 CFR Part 21 (Part 21), QAP 80.0, "10CFR21 - Procedure for Reporting Non-Conforming Material," dated November 2, 1994.
QAP 80.0 was developed by YEI to address Part 21 requirements at the two YEI facilities 2
which perform activities relating to " basic components."
The inspectors also observed and reviewed the document that was posted at both facilities to comply with 9 21.6 of Part 21, " Posting requirements."
b.
Observations and Findinos The inspectors determined that QAP 80.0 included certain provisions of 10 CFR 21.21(c) (NRC notification procedures), which are not required by the current revision to 10 CFR Part 21 to be included in procedures adopted pursuant to the regulation.
However, QAP 80.0 did not contain any of the required provisions of 521.21(a) (evaluation of deviations and failures to comply, interim reports, and notification of directors or responsible officers).
The inspectors also determined that the Part 21 posting at YEI's Reading facility was not in accordance with the requirements of 10 CFR 21.6.
The inspectors observed that YEI-Reading, apparently opting for a 621.6(b) posting, had posted only a notice that indicated that a copy of 10 CFR Part 21 was available for review in its administrative office.
However, the notice lacked the other information required by 621.6(b), i.e., a description of the regulation and the Part 21 procedures, the location where the procedures (as well as the regulation itself) may be viewed, and the name of the person to whom reports should be made.
In addition, YEI-Reading did not post Section 206 of the Energy Reorganization Act of 1974 as required by both s21.6(a) and 621.6(b).
At its Richmond, Kentucky, facility, YEI had complied with the posting requirements of f21.6(a).
The inspectors discussed with the YEI staff the provisions of 621.21(b), which require that deviations or failures to comply, discovered by a supplier of basic components, for which the supplier determines that it does not have the capability to perform the evaluation of 621.21(a)(1) to determine if a defect exists, must be reported to the purchasers or affected licensees within five working days of this determination so that the purchasers or affected licensees may evaluate the deviation or failure to comply.
The inspectors explained that although 521.21(b) is not specifically required to be included in the procedures adopted pursuant to 10 CFR part 21, it is perhaps the most important provision for a particular vendor / supplier's disposition of deviations or failures to comply, because most vendors or suppliers do not have the capability to perform a 521.21(a)(1) evaluation. As also defined in Section 21.3, the 921.21(a)(1) evaluation is the process of determining whether a particular deviation constitutes a defect, i.e. whether it could create a substantial hazard or lead to exceeding a license technical specification safety limit, or determining whether a failure to comply (with the Atomic Energy Act of 1954, as amended, or any rule, regulation, order or license of the NRC) is associated with a substantial safety hazard.
YEI agreed that this process would normally be beyond its capability because, although YEI can advise a licensee or purchaser of the effect of a particular deviation on the performance of the battery, it cannot determine the ultimate effect on plant operation, reliability or safety.
The inspectors also explained that nothing in the regu M on should be construed as prohibiting a report to the NRC by anyone J: L concerned that a deviation may be a defect or that a failure to comply may be associated with a 3
4 substantial safety hazard, even if they are not capable of performing a 4
521.21(a)(1) evaluation.
However, a vendor who is not qualified to perform the 521.21(a)(1) evaluation should not perform the evaluation (in lieu of informing affected licensees or purchasers in accordance with 521.21(b)) and determine that a report to the NRC is not required because the deviation does not appear to the unqualified vendor to be defect or because the failure to comply does not appear to be associated with a substantial hazard.
After discussing 10 CFR Part 21 responsibilities in detail with the YEI QA staff, the inspectors informed YEI that the failure to establish an adeouate procedure and having an inadequate posting at its Reading facility were violations of 10 CFR Part 21.
However, these failures constituted violations of minor significance and will be treated as a Non-Cited Violationr, consistent with Section IV of the NRC Enforcement Policy.
Subsequent to the i
Reading inspection, the YEI corporate staff drafted its revision to QAP 80.0 and the draft document was discussed during the Richmond inspection.
c.
Conclusion The inspectors concluded that YEI had not developed an adequate procedure to comply with the requirements of 10 CFR Part 21, and had not complied with 921.6 posting requirements at its Richmond facility. However, YEI had complied with the 921.6 posting requirements at its Richmond facility and has adequately revised QAP 80.0 to address the procedural requirements of 10 CFR 21.21(a) as well as including provisions to ensure compliance with 621.21(b).
3.2 10 CFR Part 21 - Informina Affected Licensees a.
Scope The inspectors reviewed an April 19, 1996, YEI letter that was sent to Southern California Edison (San Onofre Nuclear Generating Station) (SONGS) and Cleveland Electric Illuminating Company (Perry Nuclear Power Plant) (Perry) regarding GN type battery cells that were supplied.
The YEI author also provided a copy of the letter to the NRC staff for information.
The inspectors reviewed the letter to determine the adequacy of informing its customers of deviations or failures to comply.
b.
Observations and Findinas The YEI letter indicated that certain YEI GN type battery cells that were manufactured between October 1992 and December 1992 could potentially have less than YEI's publicized rated capacity of 8-hours.
The letter stated to the two licensees and NRC staff that only Perry and SONGS received GN type batteries manufactured in the suspect time period.
During the Richmond facility inspection, the inspectors reviewed the letter and associated documents including YEI-Richmond's 1992 " custom order status log" (status log) for the suspect time period.
The status log is an internal YEl-Richmond document that is used by YEI-Richmond quality control (QC) personnel and order entry personnel to maintain the status of all commercial and nuclear incoming battery cell orders that have special requirements 4
imposed.
The inspectors noted that the status log reflected that both licensees that received the YEI letter had GN type cells manufactured in the.
subject time period.-
However, the inspectors determined that there was a third licensee that received cells manufactured during this time period.. It was noted that Washington Public Power Supply System (WPPSS) had also received eight GN type safety-related battery cells.
Further review of YEI's records showed that the WPPSS Purchase Order (PO) 221020, dated October 1992, did in fact order eight GN battery' cells and imposed nuclear safety-related requirements on YEI.
YEI committed to informing WPPSS. of the deviation by July 25, 1997.
The inspectors determined'that YEI had failed to inform one of the affected licensees of the deviation in affected GN type battery cells within the time limit required by 10 CFR 21.21(b).
This-has been identified as Violation 99900359/97-01-01.
As discussed in Section 3.3 below, the inspectors determined that the GC and GN cells are similar in design and performance and their manufacturing process controls are nearly identical.
Therefore, the inspectors noted that GC cells may be susceptible to the deviation -discussed above. Although YEI has qualifir i only the GN cells for safety-related service, some-licensees buy the commen.1-grade GC cells and dedicate them.
The inspectors found that Florida Power Corporation (FPC) and New York Power Authority (NYPA) purchased GC-Type cells manufactured during the period in question.
FPC purchased 30 2GC-9 cells, assembled October 2,1992 (FPC P0 A7301166), and NYPA purchased 1
four GC-33 cells (P0 number undetermined), assembled November 23, 1992,-and tested December 11, 1992.
c.
Conclusion The inspectors concluded that YEI did not perform an adequate review of its manufacturing records for the time period in question.
Consequently, it did not identify the one other nuclear customer that received affected battery cells intended for safety-related applications. As.a result, WPPSS was not informed of the deviation, a potential substantial safety hazard.
3.3 Battery Manufacturina Process a.
Scope The inspectors observed processes in the various stages of manufacturing of YEI GN type battery cells.
The inspectors examined items in production, reviewed logs and other process records and. interviewed technicians to determine the adequacy and availability of the manufacturing instructions at the work stations and compliance with those instructions.
The inspectors also reviewed production QA activities.
b.
Observations
.The inspectors observed various phases of cell manufacturing, including oxide mill operation, grid casting and. assembly, flat plate manufacturing, paste
-i mixing, machine pasting, positive and negative plate processing (paste curing, j
5 J
l
~
a trimming, etc.), cell assembly and sealing, leak testing (15 GC-25 cells which two cells had post seal-to-cover leaks and were put aside for repair and
, of retesting), final formation of the cells (electrolyte filling and a specified sequencs:
of initial charging, discharging, recharging), and capacity discharge testing.
In general, manufacturing instructions were available and being followed.
No concerns were identified in this area. Operators were knowledgeable and c The inspectors observed QC personnel performing various prescribed routine hairline cracks, plate thicknessetchecks during the manufacturing process, incl were identified in this area.
'cose/ missing pellets, etc.
No concerns The inspectors noted that while the 4N (nuclear qualified Class IE) and GC (commercial) cells are generally similar in design and comparable in perforu nce, there are some differences, including the following:
The container material of GN is a type of polycarbonate.
The GC container materia' is a type of styrene.
The GN cover is polycarbonate; the GC cover is polyvinylchloride.
a The positive pastes of c'i and GC cells have different amounts of lead.
The outside negative p:nss of the GN are thinner than the negative GC plates.
The GC cells have not been qualified by YEI for seismic requirements.
The GN and GC plate separators are of different materials.
c.
Conclusy_is F
^
The inspectors cot.cluded that the YEI manufacturing process controls observed f
were effectively implemented.
3.4 Dita _)ification of YEI GN-Series Batteries for Class IE Service a.
Scope The inspector reviewed Wyle Laboratories Report 45001-1, Revision A, dated January 15, 1989, the environmental and seismic qualification report for YEI's plants and interviewed YEI's Naager of Engineering Suppor Batteries, who had witnessed the Wyle testing.
The inspector also reviewed the report prepared by Flight Dynamics, Inc., in which it documented the seismic qualification analysis for the YEI Series GN battery design change.
In the design change, the number of terminal posts was changed from two positive and two negative posts per cell to one positive and one negative per cell to conform to the current design of YEI's similar cells for non post nuclear applications.
6
4 b.
Observations Although Class lE batteries are typically not subject to exposure to the harsh environment of a design basis event such as loss-of-coolant accident or high energy line break (therefore not required to be qualified in accordance with 10 CFR 50.49), the Wyle test program was intended to qualify them for the expected extremes of normal service conditions (in accordance with the applicable general design criteria of 10 CFR Part 50, Appendix A) and subject the test specimen batteries to seismic testing at an end-of-life condition.
Accordingly, the various test specimen batteries underwent accelerated thermal ad radiation aging to the equivalents of 10,15 and 20 years of nuclear plant service (at Wyle, Huntsville) and the order of 10,000 rads Co-60 radiation exposure (at The Georgia Institute of Technology) before undergoing seismic testing at Wyle.
No deficiencies were noted in this report.
GN-Series batteries originally had two positive and two negative posts per cell (a total of eight posts in the usual 2GN or 2-cell-per-jar configuration) for ampacity reasons.
However, the non-nuclear line had larger single posts to save weight and cost without sacrificiag ampacity.
In order to qualify the improved design, YEI contracted Flight Dynamics, Inc., who used finite element analysis to show that (1) the stresses during design basis seismic excitation were actually lower in the larger single posts and (2) the single-post design was stronger than the 2-post design that had been qualified by the Wyle tests cited above.
No deficiencies were r:oted in this report.
c.
Conclusions The original qualification test program and subsequent design-change reconciliation analysis for the GN-Series batteries and racks appeared to have been conducted in accordance with applicable requirements and guidelines in effect at the time including General Design Criterion 2 of Appendix A to 10 CFR Part 50, Regulatory Guide 1.100, Institute of Electrical and Electronic fngineers (IEEE) Standard 323-1974 (Claas IE equipment qualification), IEEE 4
Std 344-1975 (seismic qualification of Class IE equipment), IEEE Std 450-1987 (Class lE battery qualification), and IEEE Std 535-1987 (Class lE battery seinic qualification). The inspectors had no concerns in this area.
3.5 Purchase Orders a.
Scope At the YEI Reading and Richmond facilities, the inspectors reviewed selected licensee purchase orders (P0s) for YEI GN type batteries and associated records to determine whether NRC licensees imposed the necessary and appropriate technical and quality requirements on YEI for the procurement of basic components, b.
Observations and Findinas The inspectors noted that each of the licensee P0s raviewed imposed the requirements of 10 CFR Part 50, Appendix B, and stated that 10 CFR Part 21 was applicable.
Each of the YEI customer files (called P0 packages) contained 7
~~
w i
documents such as test and manufacturing records, C0Cs, and Lab Test Assignment (LTA) sheets.
The LTA sheet is a form that is used as a manufacturing, test and quality assurance function traveler.
The LTA for each particular job contains all of the customer's specific nanufacturing and test i_
requirements.
It is generated by Richmond QC personnel, in conjunction with the order entry department.
The inspectors determined that all of the P0 packages reviewed contained appropriate quality and test requirements from the q
licensees except for one package discussed in Section 3.6 below.
c.
Conclusien The inspectors concluded that YEI's customer order packages were well maintained, retrievable and reflected the required test results.
With the exception of the test failures discussed in Section 3.6, no other anomalies were noted in this area.
~d Failure of Caoacity Test Discharae by San Onofre Cells a.
Scope The inspectors reviewed Southern California Edison (SCE) PO 6L225004, dated March 1, 1995 (with subsequent revisions in October 1995 and August 1996), and associated documentation for ten 2GN-15 replacement cells for the San Onofre Nuclear Generating Station (SONGS).
Also reviewed were test data for several
.other batteries with similar cells (Type 2GC) that had been capacity discharge tested at the 8-hour rate.
b.
Observations and Findinas The test data indicated that ten 2GN-15 battery cells tested in July 1996 failed to meet SCE's original testing requirements on four occasions.
The cells finally passed with acceptance criteria modified by SCE.
. Although the test failures were largely attributable to some weak cells (e.g.,
50001), YEI explained that they felt that the elevated temperatures (as high as 79'F) were a significant contributing factor to the poor performance of the weakest cells.
YEI explained that they had experienced difficulty in maintaining lower temperatures due to inadequate air conditioning in the test room and the July heat.
This was YEI's justification for conducting the test three more times after the initial failure, all with similar unsatisfactory
.results.
Originally, the 10 cells were to be discharged at the temperature-corrected 8-hour rate, maintaining ambient temperature as close to 77'F as possible, to an. average end voltage of 1.75 volts per cell [VPC] with a minimum of 90%
capacity per IEEE 450-1987.
The temperatures during the discharges conducted by YEI on July 1, 4, 11, and 17, 1996, varied between 74*-79'F.
However, in each test, the average cell voltage reached 1.75 VPC before 90% of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> (7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />, 12 minutes).
The worst results were from the July 17th test conducted at 79'F.
When the test was terminated at the 7-hour reading, the individual cell voltages ranged between 0.64-1.67 volts.
In this test at the highest temperatures, most of the cells suffered degraded performance; although, cell 8
^
50001 had exhibited the poorest performance consistently in the previous three test discharges.
Cell 50002 was also weak, but not as bad and the other eight cells had performed consistently better during the previous tests run at lower temperatures.
Thus, the data suggested that one or two cells were weak, but all the cells exhibited a sensitivity to high temperature.
Section 15, " Nonconforming Material, Parts, or Components," and Section 16,
" Corrective Action," of YEI's QA Manual required + Sat quality personnel prepare a material review report and disposition the nonconformances in accordance with the QA program. However, YEI did not generate any material review reports to document the deviation frem the licensee's P0 requirements and its disposition, as required by its QA procedures.
Instead, according to YEI, after the fourth test failure, the YEI Richmond facility QC staff contacted YEI corporate engineering, and YEI engineering contacted the SCE engineering staff. After' consultation between the YEI and SCE engineering staffs, SCE changed its testing requirements to 95% capacity at the 4-hour discharge rate.
The cells met this requirement during the fifth test and were shipped to SONGS.
c.
Conclusion The inspectors concluded that YEI had established measures to assure that conditions adverse to quality, such as test failures, and deviations are promptly identified and corrected, and that YEI had procedures to assure that nonconforming items would be reviewed and accepted, rejected, repaired or reworked in accordance witn documented procedures.
However, YEI failed to document and disposition the four test failures of the SONGS cells in accordance with its 10 CFR Part 50, Appendix B quality assurance program.
Accordingly, Nonconformance 99900359/97-01-02 with respect to Criterion XV of 10 CFR Part 50, Appendix B, was identified in this area.
3.7 Safety Imolications of Failure of SONGS Cells a.
Scope The fai M of the San Onofre cells is of technical concern with potentially generic u..plications.
In addition to concerns about the requirements for SONGS batteries, the inspectors were concerned that perhaps YEI's advertised 8-hour capability for the 2GN cells in its product literature was not always achievable, particularly under elevated ambient and cell temperature conditions.
To try to resolve concerns raised by the failure of the SONGS cells to pass their 8-hour capacity test discharge, the inspectors reviewed test data for several other groups of similar cells for other plants to determine whether there was any inherent difficulty in YEI 2GN cells meeting the manufacturer's published 8-hour discharge capability.
b.
Observations The 8-hour-rate capacity test data for several other batteries with similar cell types (2GC) with no anomalous cells and no elevated temperatures showed that all the cells exhibited greater than 90-percent capacity at the 8-hour rate.
Some, at the lowest temperatures, were above 100 percent.
- However, 9
=
i there were no data readily available for other 8-hour discharges at temperatures above 77'F.
Therefore, the inspectors could not rule out generic susceptibility to premature loss of capacity at elevated temperatures.
c.
Conclusion Based on the cell test data reviewed, the inspectors conc 1Lded that YEI cells of this type should be able to meet an 8-hour rate, 90-percent capacity requirement within the bounds of expected ranges of normel service conditions in nuclear plant battery rooms.
The inspectors further car.cluded that a significant factor contributing to the failures of the groap of 10 cells for SONGS was the poor performance of Cells 50001 and 50002.
However, the apparent susceptibility of the 2GN cell to premature loss of capacity at elevated temperatures may be indicative of an inherent weakness in some aspect of the design or manufacturing process.
For example, according to YEI, the 2GN cell is of a relatively low electrolyte volume design.
This feature causes the cell to exhibit a capacity-versus-discharge-rate profile typical of lead-acid cells up to about the 4-hour rate.
However, for longer discharges /
lower rates, the cell appears to suffer from electrolyte depletion and starts to exhibit reduced capacity ~in very long discharges.
The inspectors were concerned that should this become limiting, particularly at high temperatures, it could result in the unexpected inability of station batteries to maintain required voltage for the required time under certain design basis conditions.
In particular, elevated ambient temperatures due to loss of air conditioning during design basis events such as station blackout may impact the station batteries' ability to perform their safety function.
Accordingly, the inspectors strongly recommended that this be investigated and that should it be determined to be a deviation, affected licensees and purchasers should be informed in accordance with 10 CFR 21.21(b).
This issue was identified as Inspector Followup Item 99900358,359/97-01-05.
Instead of ordering a replacement for the weak cells, SCE revised the test acceptance criteria.
The inspectors were not able to determine at YEI what the basis for SCE's original specification was, whether the criterion of 95 percent at the 4-hour rate was appropriate, or whether the weak cells would adversely impact the performance of one of SONGS's Class IE station batteries.
3.8 YEI Commercial Grade Dedication Proaram GN-Series batteries are designed and manufactured under YEI's 10 CFR Part 50, Appendix B, QA program and accordingly, are supplied to NRC licensees as basic components as defined in 10 CFR 21.3.
VEI uses the provisions of its commer-cial grade dedication program as a systematic means to verify that purchased material and components used in the manufacture of Class IE GN-type battery cells are suitable for safety-related service.
In addition, the seismically qualified battery rack systems, manufactured to YEI specifications by the KIM Company, are purchased by YEI as commercial grade items, dedicated by YEI and supplied to NRC licensees as basic cc'nponents. An adequate and effectively implemented commercial grade dedication program, being an activity affecting quality, functions under the applicable controls of the vendor's 10 CFR Part b0, Appendix B, QA program.
However, it must meet, in particular, the re-quirements of Criterion III, " Design Control," and Criterion VII, " Control of
^
Purchased Material, Equipe nt, and Services," of 10 CFR Part 50, Appendix B.
10
a.
Scope a.1 Procedures and Technical Evaluation / Review for Suitability In order to evaluate the effectiveness of the YEI QA program and its implementation regarding controls applicable to the review for suitability of application and design verification in accordance with Criterion III, the inspector reviewed YEI Procedure QAP-70.0, " Dedication of Commercial Grade Items (CGI) for Nuclear Safety Related Applications," revision dated April 20, 1994.
The review included the attachments to QAP-70.0 which comprise the i
component technical evaluations and associated acceptance process sheets.
For l
reference, component part manufacturers' technical information, YEI design documents (principally drawings), and YEI procurement specifications, were also reviewed.
a.2 Acceptance / Procurement Specification Compliance To evaluate the effectiveness of the YEI QA program and its implementation regarding controls applicable to supplier selection and oualification, review of supplier documentation and examination of purchased material and components in accordance with Criterion VII for verification of satisfaction of procurement specification requirements, the inspectors reviewed acceptance
{
process sheets, incoming inspection reports (IIRs), supplier audits and commercial grade surveys, independent laboratory material analysis reports, procurement documents and supplier certificates of conformance.
The inspectors also interviewed technicians and QA/QC personnel and e..amined I
selected purchased component parts and materials.
l 1
b.
Observations and Findinas b.1 Procedures and Technical Evaluation / Review for Suitability l
The inspector determined that QAP 70.0 was not fully consistent with the requirements of Criteria III and VII of 10 CFR Part 50, Appendix B, and the provisions of Electric Power Research Institute (EPRI) Report NP-5652,
" Guideline for the Use of Commercial Grade Items in Nuclear Safety Related Applications (NCIG-07)." The procedures were also not consistent with certain provisions of NRC Generic Letter 89-02, " Actions to Improve the Dedication of Counterfeit and Fraudulently Marketed Products," issued March 21, 1989 and NRC j
GL 91-05, " Licensee Commercial-Grade Procurement and Dedication Programs,"
issued April 9,1991, in which the NRC promulgated clarifications of staff positions on key issues, later incorporated into the revision of 10 CFR Part i
21 that became effective November 1995.
QAP 70.0 did not contain the restrictions from NRC Generic Letter 89-02 on the l
use of Acceptance Methods 2 (commercial grade surveys) and 4 (product and
{
supplier performance history) alone of EPRI Report NP-5652.
However, the
{
inspector noted that QAP 70.0 and its attachments, which comprise the technical evaluation and acceptance worksheets for dedication of individual GF component parts and materials and for the batte"y rack components, did prescribe multiple acceptance methods for many items and relied predominantly on Method 1 (special tests and inspections) when a single method was employed.
11 I
l l
QAP-70.0 defined critical characteristics as in EPRI NP-5652, i.e., measurable quantities that when verified provide reasonable assurance that the item received is the item specified.
It then defined critical characteristics for design and acceptance as in EPRI NP-6406, " Technical Evaluation of Replacement Items (NCIG-11)." However, these definitions and the stated position (as in NP-6406) that critical characteristics for acceptance are a subset of critical characteristics for design are not consistent with the intent of GL 89-02, nor the explicit definition of and NRC staff position on critical characteristics as promulgated in GL 91-05, nor the definition of critical characteristics contained in the revision of 10 CFR Part 21 that became effective in 1995.
The NRC position is that critical characteristics when verified provide reasonable assurance that the item will perform its safety functions (not necessarily all design functions) and not fail in a manner detrimental to safety under all design basis conditions.
During the inspector's review of selected QAP 70.0 technical evaluation and acceptance process worksheets for the battery racks and various battery components and materials, the lists of critical characteristics for acceptance and their associated verification methods and acceptance criteria were not all complete and consistent with each other or with plant application requirements.
The technical basis or rationale for the selection of characteristics, verification methods and acceptance criteria was not apparent in some cases.
For example, QAP 70.0, No. 100-Series Attachments, the technical evaluation and acceptance process worksheets for the battery rack and components did not address the integrity of the welds in the fabricated rack ce ponents, nor the integrity of the bolted joints (e.g., stiffness, bolt torque, etc.) of installed racks.
In another instance, terminal post 0-ring material and cure date were omitted.
Since the battery racks must remain structurally sound, and the cell connections remain tight during a design basis earthquake in order to ensure the operability of the safety-related station batteries, weakness in the welds (or loose / broken connections) could lead to rack failure and battery failure and thus prevent the batteries from performing their safety functions.
In addition, the inspectors found that not all acceptance criteria were appropriate or correct, or consistent with design documents (drawings or bills of materials), or purchase specifications (which themselves were not always consistent with design documents), or expressed in technically correct terms.
One reason for this was because not all engineering design drawing changes were incorporated into purchase specifications, technical evaluations, or acceptance process sheet attachments to QAP-70, or into IIR forms.
The inspectors' sampling review indicated that for post seal caps, the wrong
- material, i.e., not in accordance with the latest revision of the design drawing, was given on the incoming inspection report form.
In another example, the incoming inspection form specified the wrong durometer hardness value for post-seal 0-rings.
Although this had been corrected by a pen-and-ir.k change on the IIR forms, the QAP-70 acceptance process sheet had a number from the previous drawing revision (D) and the P-spec had yet another number.
Having incorrect acceptance criteria may fail to detect components that are not capable of performing their safety functions under all application design i
basis service conditions.
12 l
l
[
4 b.2 Acceptance / Procurement Specification Compliance For those critical characteristics that were identified, not all were correctly specified or adequately verified.
The inspectors identified material as an example of inappropriate verification method and acceptance criteria that was seen throughout the selected job files reviewed.
Material was specified on most of the acceptance process sheets, but the verification method, visual inspection upon receipt, did not actually verify material.
The IIR forms (for which there was no procedure for preparation or for use and little training beyond 0JT) where material was identified as a critical characteristic, would simply say " material" under one of the critical characteristic column headings.
For example, IIRs for intercell connector fasteners, specifically, bolts, specified "SS-316," for the material and the blocks for each sample specimen would simply say "yes."
In effect, entering a yes just reflected the belief that the material was correct based on markings and other factors.
IIRs for post seal caps, showed that the material was similarly inappropriately verified.
The inspectors followed up on the question of intercell connector fastener material in more detail.
The material specified for intercell connector fasteners, expressed as "SS-316," was supposed to be ASTM Type 316 stainless steel.
YEI had Singleton Laboratories perform annual chemistry analyses on samples of various materials including these fasteners to confirm proper material.
However, YEI's implementation audits / commercial grade surveys of the suppliers of these fasteners (who were distributors, not the manufacturers), Threaded Screw Products (TSP), Inc., and PM Fasteners, Inc.,
were not performance based or item and critical characteristic specific.
They did not document objective quality evidence that the suppliers obtained valid, lot-traceable information (e.g., CMTRs) on material and fabrication from the fastener manufacturers, nor did they document objective quality evidence cf the suppliers' commercial quality controls to ensure that substandard or fraudulent material was not commingled.
Therefore, the audits were not usable as a basis for acceptance of the fasteners from lots other than the ones from which actual analysis samples were taken.
YEI also did not maintain lot traceability on the fasteners.
YEI did not know, for example, whether their suppliers obtained the fasteners from the same manufacturers (the markings to be verified were inadequately described) or whether the manufacturers had adequate material control, nor had YEI documented the history of the consistency of this attribute (material).
Finally, the audits were not consistent with the restrictions on the use of EPRI Method 2 in NRC GL 89-02 (nor was this mentioned in QAP-70.0).
Therefore, in view of these deficiencies, the significant instances of substandard or fraudulent fasteners on the market, many from Asia, in view of the Asian origin of the received material, and in view of the non-standard markings on the fasteners, the inspectors concluded that the yearly sample analysis was not adequate to ensure consistent material suitability.
The inspectors also noted that the C0Cs from TSP were not in accordance with the P0 requirements (B and N of QAP-500).
Although the signature blocks had the typed name of a person presumably in authority, it appeared that none of the C0Cs were signed by the named person, but rather by two different subordinates who signed the name of the designated person in authority's name instead of signing their own name with the annotation "for" the named person.
It was also not known whether the 13
actual signers had proper authority or whether the named person had ever reviewed or approved the C0Cs.
The fasteners were confirmed to be non-magnetic, consistent with-an austenitic stainless steel such as 316.
In addition, YEI milled a bolt taken from stock to check for indication of irregularities such as welded-on bolt heads.
No such indications were apparent by visual examination.
An example of a technically incorrect expression of a specification was the electrical resistance specification in the purchase specification for Amerace ACE-SIl. plate separators.
Incensistent with the separator manufacturer's technical information, the YEI purchase specification, acceptance process sheet, and the IIR form expressed this parameter in terms of ohms / square inch / mil of separator web thickness.
Separator resistance (as used in battery parlance) is actually conduction path length-specific resistance (resistivity normalized on conduction path cross sectional area) and is properly expressed in units of ohm-inch / mil of separator web thickness.
Although the separator resistance was specified using incorrect units, the inspector's calculation confirmed that the numerical values specified were consistent with the manufacturer's specifications.
c.
Conclusions The deficiencies in the YEI commercial grade dedication program description documents is considered a weakness in the YEI QA Program with respect to conformance to Criteria III and VII of 10 CFR Part 50, Appendix B.
In addition, the YEI definition of the term critical characteristic in QAP 70.0 was inconsistent with 10 CFR Part 21.
The inspectors concluded that contrary to these requirements, the measures established by YEI-Reading, and implemented by YEI Richmond, Kentucky, for review for suitability of application (CGI dedication procedures prescribed in QAP 70.0 and individual technical evaluations, acceptance process sheets, purchase specifications and prepared IIR forms) for purchased parts and materials to be used in the manufacture of Class IE station batteries for nuclear power plants did not meet the requirements of Criterion III of 10 CFR Part 50, Appendix B.
Accordingly, Nonconformance 99900358,359/97-01-03 was identified.
In addition, YEI's measures for verification that these purchased parts and materials met the procurement specifications did not meet the requirements of Criterion VII.of 10 CFR Part 50, Appendix B.
Accordingly, this was identified as part of Nonconformance 99900359/97-01-04.
3.9 Sucolier Ouality Audits a.
Scope The inspectors reviewed YEI audit and commercial grade survey procedures and checklists and also reviewed audits performed by YEI on its vendors.
The inspectors evaluated them to determine if YEI auditors verified critical characteristics of the items furnished by those suppliers, and if the audits were performance based.
14
l b.
Observations and Findinas According to the audit reports reviewed, YEI used checklists to perform the audit / surveys of Cobra Wire and Cable, Inc., Amerace-Microporous Products, Inc., I.E. Dupont De Nemours & Co, Threaded Screw Products, Inc., and KIM Engineering Company (KIM), who supplied cable, separators, acid, bolts & nuts, and battery racks respectively.
YEI documented adverse conditions in Audit Corrective Action Requests (ACARs) and requested suppliers to return the ACARs with the proposed actions to correct them.
YEI then closed the ACARs if the proposed corrective actions were acceptable.
The inspectors determined that the attributes in the audit checklist were common to all suppliers and covered a broad programmatic overview of the supplier rather than focusing on the specific item being supplied.
Furthermore, the audit did not attempt to verify the control of the quality of the critical characteristics of the specific item being manufactured.
For example, the audit of KIM failed to specifically verify the qualifications of the welders who welded the steel components of the rack and the quality control welding inspectors (QCWIs) who inspected the welds.
The audit did not reveal that the QCWIs were using a welding inspection checklist that contained all the attributes of an acceptable weld (such as, size, undercut, porosity, and length).
QAP 30.9, " Rack Welding requirements and Welder Qualifications," dated October 15, 1982, stated, in part, " Responsibility for the qualification and reexamination rests with Exide working with their suppliers as to requalification and certification of the welders." The YEI auditors did not verify this attribute.
YEI management acknowledged this weakness and committed to take adequate corrective action by developing supplemental item-specific checklists.
Similar concerns regarding the audit / survey of Threaded Screw Products is discussed in Section 2.8 above.
c.
[_qpel u sion The inspector's review of YEI's supplier quality audit and commercial grade survey procedures and checklists and selected supplier audit / survey reports revealed that they were a broad-brush, programmatic review, not performance based.
Such surveys / audits could provide a reasonable basis for preliminary qualification of a commercial grade supplier (e.g., placing the supplier on an approved commercial grade suppliers list), and may be useful in managing supplier quality resources, but they were not critical characteristic-specific and item-specific, and inconsistent with NRC GL 89-02.
The audits / surveys did not adequately verify that the suppliers' commercial quality programs were effectively implemented, and did not cover distributors' programs where l
applicable. The inspectors concluded that YEI's qualification of certain suppliers and examination of supplier documentation did not meet the requirements of Criterion VII of 10 CFR Part 50, Appendix 8.
Accordingly, this concern is identified as part of Nonconformance 99900358,359/97-01-04.
15
3.10 Seismic Racks a.
Scope To evaluate the requirements that YEI established to ensure that the seismic racks are manufactured to meet or exceed the requirements of those racks that met the seismic qualification tests, the inspectors reviewed QAP 30.8,
" Quality Assurance Control-Seismic Racks," dated November 16, 1994, QAP 30.9,
" Rack Welding requirements and Welder Qualifications," dated October 15, 1982, and YEI procurement specification P-Oll, Section 07, " Welding Steel Racks,"
dated October 1, 1982.
b.
Observations and Findinas QAP 30.8 referenced QAP 30.9 and P-Oll, Section 07.
QAP 30.9 referenced American Welding Society (AWS) Standard DI.1, as the applicable standard for welding the racks and specified the following:
Filler material to be AWS A5.18 Z 70S-3 or equivalent Welded material to be ASTM A-36 of specified minimum yield strength Welding to be done in horizontal position Weld profiles to be in accordance with AWS DI.1, Sections 2.7 and 3.6 Weld Procedure Specification and Welder Qualification to be in accordance with AWS DI.1, Parts B and C Weld quality as required by AWS Dl.1, Paragraphs 5.10.3 and 5.11.2 Visual inspection in accordance with AWS 01.1, Paragraph 5.6.3 Weld size and location as specified on Drawing MC-83860, " Frame - Steel
- 2 Step (G)," Revision E, dated October 3, 1985 The inspectors noted that Drawing MC-83860 had no weld dimension tolerances and observed that the engineering specifications were scattered among various documents and not consolidated in the drawing to facilitate QC inspection of the welds to the applicable requirements.
YEI committed to revise the drawing which is frequently used by welders as well as inspectors to ensure that all welds meet or exceed the quality of the qualified specimen.
c.
Conclusion The engineering specifications for the weld and assembly of the racks were adequate, but were found to be scattered in various places instead of being consolidated in one place.
The lack of colocated specifications and the audits or surveys of the KIM Company not verifying welder qualifications were weaknesses in YEI's QA control of special processes which YEI committed to strengthen.
16
,'u e
a.
i 3.11 Internal Audits a.
Scope In order to determine.if QC performed internal audits to verify the effectiveness of the quality assurance program, the inspectors reviewed Quality Procedures Manual (QPM) 4.17, " Internal Quality Audits," dated October 7, 1996, as well as selected internal audit reports.
I b.
' Observations and Findinos According to the records, the internal audit performed by QA on April 3,1997, to verify compliance with QPM 4.3.1, " Contract Review" dated July 10, 1995, identified 19 nonconformances.
The audit conducted March 31, 1997, to verify-compliance with QPM 4.18, " Training," dated March 13, 1997, identified several nonconformances; three of them repeated. The inspectors noted that corrective action for the repeat nonconformances was being implemented, but full compliance had not yet been achieved.
The inspectors determined that corrective action requests (CARS) were written to identify nonconformances in I
i all instances and were sent to the cognizant manager to determine the root cause of the problem, and document the corrective action taken or recommend the proposed corrective action.
c.
Conclusion i
The inspectors concluded that plant QA auditors performed internal audits in i.
accor6ance with procedures to verify that quality activities comply with the piant.ed arrangements and to determine the effectiveness of the quality system, audits were legible and retrievable, and corrective action being taken was being monitored.
No concerns were identified in this area.
3.11 Trainina i
a.
Sq_ogg The inspectors reviewed selected training records of craftsmen to determine if they had received training for the activities they ware performing.
b.
Observations and Findinas The inspectors reviewed the training folders of five craftsmen working in the pasting, oxide, burning, and gluing areas.
The folders contained train %g record sheets that documented the date and subject of the training and the name of the instructor.
The sheets were signed by attendees, acknowledging the training received.
The training records were legible, retrievable, and complete.
c.
Conclusion The inspectors determined that personnel were trained in the areas in which they were working.
17
o juo o
4 PERSONS'CONTACTEQ I.C. Baeringer, Vice. President, Engineering-R.P. Bender, Vice President, Quality Assurance and Procurement C. Claypool, Quality Control Technician J. Hall, Quality Control E. Simpson, Quality. Control L. Rickman, Plant Manager B.P. Lightner, Manager, Supplier Quality M.A. Patel, Manager, Engineering Support, Large Stationary Batteries S.J. Weik' Manager, Design Engineering and Document Control L.R. West, Manager, Quality Assurance, Richmond ITEMS OPENED. CLOSED. AND DISCUSSED l
Opened 99900358,359/97-01-01 VIO Informing Affected Licensees of Deviation
- 99900358,359/97-01-02 NON Documenting /Dispositioning Test Failures 99900358,359/97-01-03 NON Review for Suitability of Application 99900358,359/97-01-04 NON Product Verification / Supplier Audits / Surveys i
Closed None.
No prior open items.
Discussed 99900358,359/97-01-05 IFI Elevated temperature effect on GC and GN cells 18
. -.