ML20008C254
| ML20008C254 | |
| Person / Time | |
|---|---|
| Issue date: | 01/26/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20008C251 | List: |
| References | |
| REF-QA-99901323 99901323-97-01, 99901323-97-1, NUDOCS 9801290217 | |
| Download: ML20008C254 (15) | |
Text
-
?
y e
^
i l
U.S. NUCLEAR REGULATORY COMMISSION
. OFFICE OF NUCLEAR REACTOR REGULATION i
1 l.
1 r
Report No:
99901323/97-01 Organization:
Westlectric Castings, incorporated f
2040 Camfield Avenue City of Commerce, Califomia 90040 i
Contact:
Andrew Arechiga, Quality Assurance Manager (213) 722-8000 i
Nuclear Industry Manufacturer of low carbon and stainless Activity:
steel sand castings such as impellers, valve bodies and pump replacement parts.
1 l
Dates:
November 17 - 19,1997 1
I l
inspectors:
Joseph J. Petrosino, NRR l
l Donald G. Naujock, NRR Approved by:
Robert A. Gramm, Chief Quality Assurance and Safety Assessment Section l
Quality Assurance, Vendor Inspection and l
Maintenance Branch i
l Division of Reactor Controls and Human Factors l'~
Office of Nuclear Reactor Regulation i
l i
9001290217 990126
~
PDR GA999 ENVe****
l 99901323 PDR l
j 1
INSPECTION
SUMMARY
During this inspection, the NRC inspectors reviewed the implementation of selected portions of l
the Westlectric Castings, incorporated, quality assurance (QA) program, and reviewed activities l
associated with its manufacture and supply of low carbon and stainless steel sand castings to l
the nuclearindustry.
l The inspection bases were:
Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing e
Plants," to Part 50 of Title 10 of the Code of Federal Reoulations (Appendix B) l 10 CFR Part 21," Reporting of Defects and Noncompliance" During this inspection, a violation of NRC requirements was identified and is discussed in Section 3.1 of this report. Additionally, severalinstances where Westlectric Castings, incorporated (Westlectric), failed to conform to NRC requirements contractually imposed upon them by sub-tier suppliers for NRC licensees were identified. These nonconformances are discussed in Sections 3.2, and 3.4 of this report.
r 2
STATUS OF PREVIOUS INSPECTION FINDINGS l
This was the first NRC inspection of the Westlectric facility.
3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 10 CFR Part 21 Prooram a.
Inspection Scone The NRC inspectors reviewed Wes'.lectric's procedure for reporting in accordance with 10 CFR Part 21: QA Procedure QA104, "Part 21 - Reporting of Defects and 1
Noncompliance," Revision A, dated, January 26,1996. The NRC Inspectors also observed and reviewed Westlectric's 10 CFR Part 21 posting.
b.
Observations and Fino nos l
The procedure did not address evaluation, notification or associated time constraint requirements of $21.21, " Notification of failure to comply or existence of a defect and its i
evaluation," of 10 CFR Part 21. Instead, it focused on delineating specific requirements for Westlectric employees regarding intemal processes to identify, document and transmit product deviations to a Westlectric Part 21 Committee member. For example, the
" reporting" section of the procedure had three paragraphs indicating how an employee was required to fill out the Westlectric deviation documentation form.
The NRC Inspectors conhted discussions with the QA Manager to outline the salient points of the Part 2* regula 7 that are required to be addressed and included in a Part 21 procedure. Failure to have adequate procedures to require that evaluations or reporting of deviations is performed as required in 10 CFR 21.21(a), constitutes a violation of NRC requ rements. The QA Manager committed to revising and issuing the j
2 I
I
l
(
corrected Part 21 procedure within 120 days of the November 19,1997, exit meeting.
Violation 99901323/97-01-01 was identified in this area.
The NRC Inspectors determined that Westlectric's posting documents were in compliance with the Part 21 regulation and that it was conspicuously displayed in multiple locations at Westlectric's facility. However, it was noted that the inadequate Part 21 procedure was integral with the posting. Therefore, the QA Manager stated that he would also revise the posting document to include Westlectric's new Part 21 procedure when it was revised and issued.
Potential 10 CFR Part 21 lssu.2 As discussed in Section 3.4 below, a Westlectric practice was identified by the NRC Inspectors regarding the method employed for " correcting" the chemical composition analysis heat results obtained from Westlectric's spectrometer readings. As stated in S21.3 of Par 121, a deviation means a departure from the technical requirements included in a procurement document. Since this matter may represent a deviation of the procurement documents, as of November 19,1997, Westlectric was reviewing the circumstances to determine if its customers needed to be informed in accordance with
$21.21(b) of Part 21.
c.
Conclusions The NRC Intpectors concluded that Westlectric's procedure adopted to implement the provisions of 10 CFR Part 21 was not adequately established to ensure that evaluations i
were performed in accordance with 921.21, " Notification of failure to comply or existence of a defect and its evaluation."
3.2 Quality Assurance Proaram a.
Insoection Scoce The NRC Inspectors reviewed Westlectric's QA Manual (QAM), Revision K, dated October 14,1996, and associated procedures and records to assess the Westlectric r, 'ality program to determine whether it adequately addressed applicable Appendix B inquirements.
b.
Observations and Findinas Quality Assurance Manual The NRC Inspectors determined that Appendix B requirements were contractually imposed on Westlectric i'. approximately 1991 by Pacific Pump (currently ingersoll-Dresser Pump, Inc. (IDP) and Bryon Jackson Pump Division (BW/IP). It was noted that Westlectric's QA manual (QAM) did not address whether it did or how it would meet the applicable Appendix B criteria necessary to provide adequate confidence that its components supplied for safety-related use would perform satisfactorily in service. That is, the QAM did not adequately address how the applicable requirements of the 18 criteria of Appendix B were to be satisfied by the Westlectric QA program. The introduction of the QAM stated that the manualis a description of the procedures followed to maintain l
l i
j k
quahty standards required to produce castings in accordance with Military Specification-i inspecibn (MIL-l)-45208, " Inspection System Requirements." It also stated that the
'j manus sets the calibration system requirements of Military Specification-Calibration l
(MIL-
'l62A, " Calibration System Requirements." The QAM also stated that all l
measuse,nents related to product conformance are traceable to the National Bureau of j
Standards (NBS), and that the manual establishes the quality system and procedures i
required by the company.
The NRC inspectors noted that the QAM contains general instructions and procedures i
such as casting traceability, pattom maintenance control, corrective action, control of non-conforming castings, calibration procedures, processing test bars and test reports, i
shipping procedures, sand control, heat treat process, and metal control.
.i Accuraev of QAM Procedures i
The NRC inspectors also selectively assessed the Criterion I aspect of requiring that the authority and duties of persons and organizations performing activities affecting the j
safety-telated functions of nuclear power plant components be clearly established and j
delineated b writing. After reviewing specific requirements of selected procedures, the -
NRC inspectors identified several areas where it determined that the procedural steps i
and requirements were not clearty or accurately established or delineated in writing. For example, the NRC Inspectors found that the QAM general policy s3ction did not accurately reflect the QA personnel and QA department structure that was in existence for several years. The QAM's general policy section states that all of the Westlectric inspectors, including non-destructive examination (NDE) level 11 inspectors are under the direct supervision of the "QA Assistant." However, the NRC Inspectors determined that -
as of November 19,1997:
1 Although Westlectric's QAM states that all Westlectric visual and nondestructive l
e examination (NDE) inspectors are under the direct supervision of the "QA Assistant,"
Westlectric has not had a QA Assistant since approximately 1991.
Until recently the visual inspector worked for and reported to production department personnel. Currently, there are two visualinspectors.
Until recently Westlectric's QA department consisted of only the QA Manager.
e Currently, the Westlectric QA department consists of the QA Manager and both visualinspectors.
Westlectric has had only one NDE certified inspector since approximately 1991, and e
the NDE Inspector does not report to or take direction from the QA Department.
Although the Level 11 Inspector also holds the positicn of " Chief Inspector," the e
authority and responsibility of the Chief Inspector were not delineated in QAM Section 1.0, " General Policy," 3.0, " Quality Control Department Operation anu Duties," 8.0, "in-Process inspection," or 10.0, " Final Inspection."
Although both Westlectric visual inspectors report to the QA Manager, they also take L
inspection activity direction from the Chief Inspector / Level 11 NDE Inspector. Thus, their independence from production is not assured.
l' :
l l
The NRC Inspectors identified that the Level 11 NDE Inspector is also designated as the Cleaning Room Foreman (discussed in weld rod control procedure), and the Cleaning Room Supervisor (discussed in heat treat process). The NRC inspectors determined that the employee that holds each of these positions reports to, and is under the responsibility of the Operations Manager, and does not report to or take inspection direction from the QA Manager. Discussions with the Westlectric staff and management indicated that this relationship had not been reviewed or approved by the QA organization to determine whether sufficient independence from cost and schedule when opposed to safety considerations existed.
The NRC inspectors identified that the QAM description of the reporting relationship of the visual and NDE inspectors is inconsistent with actual practices. Nonconformance 99901323/97-01-02 has been identified in this area.
Insoection Stamos The NRC Inspectors noted during review of PO packages that Westlectric's original certified material test reports (CMTRs) were not hand signed. It was noted that the approving official's signature was stamped instead of being signed by the approving official, the QA Manager. The QA Manager stated that he had a signature starrp for his use (mostly on CMTRs). During a discussion between the NRC Inspectors and QA Manager, the shipping supervisor brought in a stack of CMTRs to sign (stamp), and the QA manager stamped the records and retumed them. The NRC Inspectors noted that the QA manager kept his signature stamp in his desk drawer. When asked whether anyone else was issued signature or inspection stamps, the NRC inspectors were informed that the Level 11 NDE Inspector / Chief Inspector had also been issued a signature stamp for stamping NDE quality records. Subsequently, the NRC Inspectors requested the Chief Inspector to retrieve his signature stamp and it was noted that the Chief inspectors signature stamp was stored in a locked cabinet in the Cleaning room office.
The Chief inspector stated that the main reason that he had a signature stamp was because many of the Westlectric documents were triplicates. The NRC Inspectors obtained a blank Westlectric CMTR form and noted that the form was made of carbon-I copy type paper, which would allow a signed signature to be reproduced on tne second j
and third pages of the triplicate form.
l i
The NRC Inspectors noted that paragraph 1.4 of Section 3.0, " Quality Control Department Operation and Duties Procedure," of Westlectric's QAM requires the QA Manager to maintain a file in his office of all inspection stamps issued to workers, and Criterion V,
" Instructions, Procedures, and Drawings," of Appendix B requires activities affecting quality to be accomplished in accordance with documented precedures. The NRC Inspectors review of this area revealed that the two signature stamps for the QA Manager and Chief Inspector were not indicated as being issued even though they are used for quality record approval. Nonconformance 99901323/97-01-03 was identified in this area.
c.
Conclusions The Inspectors found that Westlectric's QAM did not accurately describe its QA J
department personnel and associated duties. Further, the NRC inspecice concluded that the QA Managers and Level ll NDE inspectors signature stamps wen 'ut identified as being issued in the QA Managers inspection stamp file.
_._ _~
l i>-
i 3.3 :
Welding Control
-1 7
a.
Inspechon Scoce 1
j i
The NRC Inspectors conducted discussions with Westlectric staff regarding its welding.
program including welding processes, observed weld rod control and issuance, and reviewed certification and qualification of personnel. The NRC Inspectors noted that l
l Westlectric's welding program control is outlined in a " Welding Control Procedure,". that is.
i
. part of Section 22.0, " Control of Special Processes," of Westlectric's QAM. There were
.l no orders for nuclear applications being processed at the time of the inspections involving i
weld maps or other special controls. Therefore, the NRC Inspectors were not able to observe or witness implementation of the welding control process.
b.
Observations and Findinas
~ Paragraph 1.1.1 of Sechon 22.0 of Westlectric's QAM states in part, that the welding.
program will meet Section IX, " Welding and Brazing Qualifications," of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code. The NRC '
inspectors reviewed welding procedure specifications (WPS), procedures qualification reports (PQR), and welder qualifications. Westloctric has 20 different WPSs, each with a unique PQR, and currently has 11 welders that are qualified to at least one weld procedure. The NRC Inspectors determined that the welding records are well maintained and records are kept for each welder to indicate when last a specific procedure was used during each 3 month interval. This satisfies the requiremants of American Society for Testing and Materials Standard (ASTM) A 488, " Standard Practice for Steel Castings, Welding, Qualification of Procedures and Personnel," which in tum, satisfies the 6 month interval required by subparagraph QW-322.1, " Expiration of Qualifications," of Section IX, ASME Code.
Paragraph 5.1.5 in the weld control procedure stated that major weld repair shall be -
l documented by generating a weld map if required by specification or customer purchase order (PO) requirements. The NRC Inspectors observed examples of weld maps in shop traveler recoid package 128674, dated February 11,1997, for ingersoll Dresser Pumps (lDP), PO 91797, for an intermediate cover, grade CA6NM, (Heat 976A).
c.
Conclusions t
The NRC inspectors concluded that Westlectric maintained good records of its welder qualification and certification program and also generated and maintained satisfactory records to allow traceability of individual welders to the applicable welding specifications and procedures to maintain certification.
3.4 Soediemeter-Chemical Analysis a.
Inspechon Scope The NRC Inspectors reviewed the current and historical chemical analysis records generated from Westlectric's Thermo Jarrell Ash spectrometer. The purpose was to assess chemical analysis accuracy (by verifying calibration), ensure that the results were
- l., _
accurately documented on the applicable certified material test reports (CMTRs), and verify compliance to Westlectric and customer requirements.
b.
Observations and Findinas Spectrometer Calibration - First Step i
Westlectric performs chemical analysis with a Thermo Jarrell Ash (TJA) 181/81 optical emission spectrometer, Model 12621600, Serial Number 41883. The TJA recommended calibration consisted of two steps. The first step required creating a curve-set by developing curves for each element that the machine is capable of assessing. Each curve is a plot of light intensities emitted from electrical arcing against metal standards of different chemical concentrations and the certified chemical concentrations for these standards. Each curve-set is a family of curves for the chemical elements to be analyzed. For example, carbon steelis one unique curve-set, while stainless steelis a different set of curves. The standards are normally purchased from recognized sources including the National Institute of Standards and Technology (NIST).
The NRC Inspectors requested to see the calibration records and traceability back to NIST for Westlectric curve-sets identified as: LOWB (ASTM-A-216), SER3M (ASTM-A-743, CF3M), and SERIES 4M (ASTM-A-743, CA-CNM). The Westlectric staff was unable to provide curve-set calibration records for the spectrometer, and did not have curve-set traceability to NIST. The NRC inspectors and QA Manager phoned the spectrometer service representative and were informed that the curve-sets were retrievable from the computer files along with the identity of the standards.
Spectrometer Standardization - Second Step The second recommended Spectrometer manufacturer's step in calibration is the standardization of the curve-set. The repositioning of the curves in a particular curve-set is called standardization. Standardization of the curve-set is necessary because the spectrometer is sensitive to atmospheric effects, equipment wear, and equipment cleanliness. In order to maintain a high level of accuracy, repeatability, and reproducibility, the curve-set must be standardized each day before use, and more frequently if necessary. The NRC Inspectors determined that Westlectric used a one-point technique for standardization. The one-point technique locks a particular alloy's curve-set's chemical composition parameters to the associated NIST standard. This technique is a fast and effective technique for analyzing specimens with a chemical composition that is similar to that particular standard.
A common weakness in the one-point technique is that the further the test specimen's analysis is from the locked chemical composition value, the larger the potential error in analyses. For the curve-sets LOWB, SER3M, and SERIE4M, Westlectric used NIST Standards 1261A,1155, and C-1289 respectively. The NRC Inspectors identified that Westlectric did not establish specific documented acceptance or rejection limits on the analysis range for each element affected by the one-point standardization for each curve-set.
' One curve-set consists of multiple individual or unique curves. ~.
l u
Since Westlectric did not have comprehensive spectrometer operational procedures, as discussed further in this Section, the spectrometer technicians de
- eloped the concept of looking at the standard deviation for each element to determine
icceptability of the calibration. One of the technicians (who is the production melte aformed the NRC inspectors that low standard deviations were acceptable and t standard deviations j
were rejectable. The operator could not place a number on s would be considered I
the maximum acceptable standard deviation values. The NRL aspectors consider the lack of written procedures for the acceptance of the standardization calibration and the absence of limits to the chemical range for each element as a deficiency in Westlectric's spectrometer control.
Since the NRC inspector was not able to obtain evidence of the accuracy of the curve-sets, and since Westlectric was not able to demonstrate tracewility of the curve-sets back to NIST standards, the NRC Inspectors were not able to verify that all 1
measurements related to product conformance are traceable to NIST as required.
1 Additionally, the NRC inspectors were not able to verify that Westlectric's spectrometer meets the calibration system requirements that Westlectric had stated, that is MIL-C-STD-45662A requirements. The NRC Inspectors determined that Westlectric did not establish adequate measures to controlits spectrometer operation and maintenance.
Nonconformance 99901323/97-01-04 was identified in this area.
As a result of this identified deficiency, a departure from the technical requirements, as defined in S21.3 of 10 CFR Part 21, included in a procurement document may have occurred as discussed within Sections 3.1 and 3.4 of this report.
Soectrometer Procedures The NRC Inspectors found that production personnel that used the spectrometer for its heat verifications used a written procedure for a standardization calibration but it was not controlled within Westlectric's QA program. The procedure was found to be untitled, was not dated, and was not approved or signed by the person (s) who developed the procedure. Additionally, the procedure for a standardization calibration did not set limits on the analp range for each element affected by the one-point standardization.
Therefore, the !diC inspectors determined that Westlectric did not assure that the spectrometer was property control'ed, calibrated, and adjusted to maintain its accuracy.
This is another example of Nonconformance 99901323/97-01-04.
Chemistrv Verification Metal Control Procedure, Section 11.0, states that heats shall be comprised of select scrap, iron ore, processed alloys, and remelt-material. It also requires that a minimum of three chemical analysis chechs be made on the spectrometer for arc fumace heats prior to pouring castings. The three checks for the arc fumace heats are performed at the beginning, middle and end of the process, specifically: (a) melt-down, (b) preliminary, and (c) final. The preliminary analysis checks all chemical elements to allow calculation of the necessary alloy additions to the heat as well as assuring that a vigorous carbon boil is occurring. The final analysis verifies that all alloy additions were properly added and that the analysis meets industry and customer specifications.
The Westlectric person that assures that each heat is comprised of the correct amounts of materialis the metter. The metter is also responsible for the outcome of each of the three checks by verifying on the spectrometer that each heat, which he is making and monitoring, meets the specifications.
Criterion I,
- Organization," of Appendix B, state that the QA functions are those of assuring that an appropriate program is established and effectively executed, and verifying, such as by checking, auditing, and inspection, that activities affecting safety-related items have been correctly performed.
The NRC Inspectors noted that Westlectric's QAM did not specifically allow or disallow its metter from verifying his own work. It merely stated that the chemical analysis testing shall be performed by a trained laboratory technician. The inspoetors also noted that Westlectric's in process procedure (Section 8.0), and final inspection procedure (Section 11.0) did not address " inspection" of heat chemistry, even though the chemical composition is very important to the safety-related component.
Evaluatino Correction of Chemical Comoosition Analyses Effects immediately after standardizing the curve-sets, Westlectric runs ihe same NIST standard as if it was an ordinary heat specimen. The values from this run are compared to the NIST standard's CMTR values, and the differences for each major element are noted on a daily adjustment work sheet. The operator told the NRC Inspectors that the daily adjustment work sheet is used to change the final chemical analysis of a heat that might otherwise be out of specification. The rationale is that the standardization process is not accurate because the spectrometer does not provide identical results with the values shown on the NIST standard's CMTR (neglecting any tolerances).
Therefore, Westlectric would either add or subtract the daily adjustments to the actual values from the spectrometer p; inter. The NRC Inspectors observed that the operator adjusted the chemical results while making out the " Daily Chemical Analysis Report. 2 After the operator performed the corrections to the daily chemical analysis report, it was given to the QA Manager. The QA Manager then takes the daHy chemical analysis report
{
that has been corrected and manually inputs the data into Westlectric's computer network for the documentation and issuance of applicable CMTRs for each heat. The NRC Inspectors observed an example where the shipping department supervisor called up a specific heat on the network and printed the CMTR in the shipping department. The NRC Inspectors were informed by the QA Manager that he is not typica!!y involved in generating the chemical analysis and has limited knowledge regarding the operation of the spectrometer.
During a subsequent telephone conversation with the TJA spectrometer representative, the representative stated to the NRC Inspectors and Westlectric's QA Manager that there is no need to perform any correction or adjustment of the spectrometer values because all necessary adjustments are automatically performed during the standardization process. This was not known by the Westlectric personnel.
2 The daily chemical analysis report was merely a pre-printed form where the corrected values would be documented and handed to the QA Manager.
.t I
l As a result of Westlectric's adjustments of its spectrometer results, Westlectric was documenting incorrect spectrometer values and consequently, was issuing CMT& that L
were not representative of the actual chemical composition of the casting.
t l
Section 11.0,
- Metal Control," states that the chemical analysis verifies that all alloy additions were property added and that the analysis meets customer specifications. To determine the effects from Westlectric using this manual adjustment technique, the NRC I
inspectors reviewed five heats that were applied on purchase orders identified for nuclear l
safety-related applications. The five heats were 109B,125B, U195, U354, and 976A.
Table 1 tabulates the chemistry before and after adjustment of the actual spectrometer's I
chemical analysis and the CMTR chemistry.
l Totdo 1 L
Heat Chemcal Analyest Babe & ARet Manuel Correcten/%ustment I
rMises are in WeigM Percent) l Heat No.
C Mn Sa P
S Cr Ni Mo l Cu I
Muste(P 976A 0 06 0 71 0.52 0 029 0 011 12.27 4 25 0 62 l-No 975A 0 03 0 67 0 52 0 028 0 011 12 08 4 23 0.53 Yes j
yp% [R
$Qf M
(fR QR M gg@MW.
\\
125A 0052 1.23 0 61 0 037 0 021 18 71 10.50 2.92 0.26 No j
125A 0 03 1.30 0 84 0 038 0.020 18 81 10.25 2 92 Yes N$ffh)s l
,h g fb E 109B 0076 0 67 0 67 0 027 0 022 12.27 3 70 0 56
' Ni 109B 0 03 0 67 0 67 0 027 0.022 12.27 3 70 0 56 Yeg 1
3 hShh hi
/
h bi
. h hh U195 0.216 0 85 0.50 0 021 0 038 0.34 0.21 0 06 0 05 -
No U195 0.21 0 88 0 50 0020 0.031 0.34 0.21 0.06 0 05 Yes i
' Yl b
%! N ~ O U354 0.28 0 92 0.56 0 019 0.022 0 18 0.18 0.04 0 04 No l 0.04 U354 0 28 0 90 0.57 0 f123 0 023 0 18 0.18 0 04 Yes Table 2 tabulates the heat numbers with customer purchase orderinformation. From Table 1, the carbon in heat 109B is 0.076 before adjustment and 0.03 on the CMTR after adjustment. The 0.076 exceeded the ingersoll-Dresser Pump Company purchase order number 074528 specification ASTM-A743 CA6NM which has a 0.06 maximum carbon.
The NRC Inspectors asked Westlectric to standardize the spectrometer and run a recheck of the initial test specimen which came back as 0.05 carbon. Based on the ffects that the manual adjustment can have on the final chemical analysis, the NRC Inspectors informed Westlectric that they were required to review this matter in accordance with $21.21 of 10 CFR Part 21.
l c
l t _
(:
l Table 2 Heat Number and Customer Purchase Order Wormaton Heat Traveler Customer PO PO Matenal Part No.
Date Spec.
No.
l 970A 128674 (DP 91797 1/31/97 ASTM A 743 M6535 CA6NM 125A 133255 BWAP 21W50517 8/29/97 ASTM A 351 24296 1302*A CF3M 21997 I
130257 '
23968 109B 130120 IDP 95695 7/28/97 ASTM A 743 M6535 CA6NM U195 128220 ASTM U195 126220 BW/IP W433933 11/13/96 A 216 72579 Westlectric started to review the circumstances surrounding the matter to determine if it was a deviation, and whether it needed to either evaluate the issue or inform the customer. Subsequent to the inspection, Westlectric informed the NRC staff that it had informed an applicable customer pursuant to $21.21(b) of 10 CFR Part 21, so that it could cause an evaluation to be performed, and it was still reviewing the remaining issues to determine if additional deviations existed.
The NRC Inspectors determined that the adjustment of the spectrometer's chemical analysis results without an established program to control the special process and without a program to assure that appropriately tri.ned personnel operated the spectrometer were indicative of an inadequately controlled special process. Nonconformance 99901323/97-01-05 was identified in this area.
Westlectric QA 11.1.5.1 states that chemical analysis testing shall be performed by a trained lab technician. Westlectric said that their melt shop superintendent has been trained by TJA on the operation of the spectrometer. However, the melt shop super-intendent was unavailable during the inspection. The NRC inspectors were told that the melt shop superintendent gave on-the-job training to the individuals operating the spectrometer. The NRC Inspectors observed a metter and a recent'y assigned lab tech operating the spectrometer. The melter told the NRC Inspector that when problems occur he would check the argon pressure, push the spectrometer reset button, adjust the lens (mirror), clean the arc chamber, and/or restandardize the spectrometer. If none of these actions clear up the problem, he calls the melt shop superintendent. The lab tech said he calls the melter if he has problems. The NRC Inspectors noted that some speci-mens had multiple cracks across the testing surface and some of the bums overlapped each other. The NRC Inspectors noted that although either of these conditions can cause incorrect or erroneous readings the Westlectric personnel were not aware of the impact of these conditions. Additionally, the NRC Inspectors determined that neither the metter nor laboratory technician knew how to control movement in the mercury meter during the mirror adjustment prior to standardization. It appeared to the NRC Inspectors that the individuals operating the spectrometer received limited guidance on operating the l
l spectrometer and specimen preparation. Except for the standardization procedure, the spectrometer is operated with no other written guidance. Nonconformance 99901323/97-01-05 was identified in this area.
i Soectrometer Technicians l
The NRC Inspectors conducted discussions with and observed two production personnel that normally test heat samples taken from the Westlectric arc and induction fumaces during the melting / combining process. Both employees typically operate the spectro-meter; therefore, the NRC Inspector's discussion encompassed the operation, calibration and principles of the spectrometer's chemical analyses. The QA Manager was involved in all of the discussions. The NRC Inspector determined that the two production personnel were regularly assigt.ed to operate the spectrometer and verify the chemical analyses of the heats. One of the employee's title was " melter," and the other was the production department's " laboratory technician." Both employees work and report to production department management. The metteris a production department employee that is responsible for measuring and putting the correct amount of raw material (alloy additives) into the fumaces to bring the heat within the chemical analysis before the pouring of the specific castings.
The NRC Inspectors were informed by the Westlectric QA Manager that the two Westlectric personnel had received indoctrination on the operation of the spectrcmeter, but their responses indicated that the training was limited and was not overly comprehen-sive. The NRC Inspectors noted that the Westlectric personnel were not very familiar with the calibration and operating principles of the spectrometer. Although the NRC inspectors were informed that both employees had received training for the operation of l
the spectrometer, records to indicate the training was satisfactorily completed were not provided during the inspection. The NRC inspecto s could not verify that both production department employees were appropriately trained for the job activity. Westlectric pro-vided certification records for the employees subsequent to the inspection; however, those records were not specific enough to determ;ne the appropriateness of training.
Although Appendix B requires that special processes are controlled and accomplished by qualified personnel using qualified pror 9dures in 6 ccordance with applicable codes, standards, specifications, criteria, and other special :equirements, Westlectric did not i
establish adequate procedures or appropriately trined personnel to control its spectro-meter chemical analyses' operation. This is anoiner example of Nonconformance 99901323/97-01-05.
c.
Conclusions
{
The NRC inspector was not able to verify that "all measurements related to product conformance regarding its spectrometer are traceable to NIST, and that Westlectric's spectrometer meets the calibration system requirements specified in MIL-C-STD-45662A." Westlectric was not able to produce records to indicate that it had been adequately controlling, calibrating and adjusting its spectrometer within documented
' The contents of one fumace batch that is blended to a predetermined chemical composition j
is commonly referred to as a " heat." Therefore, each particular fumace batch will be identified with a different heat number.
t,
i
parameters, and that its operators were appropriately trained. Given the extent of the concems identified by the NRC Inspectors the adequacy of the quality oversight functions for the spectrometer operation is questionable.
3.5 Traceability of Castinos
- a. Inspection Scoot The NRC Inspector reviewed procedure 19.0, " Casting Traceability " to identify the Westlectric requirements that have been established and to determine whether manufacturing has been in compliance with the requirements.
- b. Observations and Findinos Paragraph 1.2 of Procedure 19.0 states that heat / code numbers shall be affixed to pattems with raised aluminum letters or pressed into molding sand with letters attached to a handle. The NRC inspectors observed three 8" x 8" valve body castings from heat U818 which showed that the production department was not in strict compliance with its procedure. The NRC Inspectors observed that the first three digits of the heat number cast into the body appearea satisfactory, but the fourth digit was ground off and replaced with a stamped letter. Westlectric stated that the heat number sequence is determined by the molding foreman because he has to insert the heat number into the mold several days before pouring.
Occasionally, process control modifications such as, smaller quantitie-s for a specific heat, or a chemical composition analysis that was incorrect will cause a change in the anticipated production and heat numbers. Instead of making up new molds with the correct heat number, Westlectric casts the existing molds, grinds off the wrong portion of the cast heat number, and stamps the correct heat numberin its place.
1 The NRC Inspectors were informed that another exception to the procedure is that on occasion, the cast heat number is illegible and Westlectric will grind off the illegible j
portion and stamp it. The NRC Inspectors confirmed that even though these practices are commonly performed, they are not delineated in the Westlectric casting traceability procedure. Although this process is not in conformance with an Appendix B QA program, the NRC inspector notes that a nuclear safety related component would te handled somewhat different. That is, a Westlectric shop traveler would be used on nuclear orders i
and the shop traveler, which accompanies the component throughout the rnanufacturing process, requires that the heat be documented. As a result, if the heat was illegible, the i
traveler could provide some assurance of the correct heat.
c.
Conclusions The NRC Inspectors concluded that although these manufacturing practices may be necessary to prevent costly rework, they are not in complia.nce with the documented Westlectric and Appendix B requirements. This practice could be a factor for con-j sideration if the component was destined to be " dedicated," in accordance with i
10 CFR Part 21, because traceability to a heat may be indeterminate if numbers are modified from original cast numbers. This issue is an additional example of Nonconformance 99901323/97-01-03.
5.;
i j
i
- 3.6 ~ Nondestruchve Examinshon i
" a. Inspechon Scone 1
- The NRC Inspectors revie,wed Westlectric's Procedure's 1.0, " General Policy," and 22.0,1
" Control of Special Procosses." Procedure 1.0 outlined Westlectric's QA department policies and responsibi' sties and Procedure 22.0 addressed the control of welding, heat treat and nondestructive examination (NDE) to assess the adequacy of Westlectric's NDE control.
- b. Observations and Findinos Paragraph 1.1.1 of Procedure 22,0 stated that the QA manager is responsible for the welding program which will meet Section IX of the ASME Boiler and Pressure Vessel Code. Paragraph 2.6 of Procedure 1.0 stated that allinspectors, including magnetic particle testing (MT) and dye penetrant testing (PT) Level 11 who are under the direct supervision of the QA Assistant are responsible for the proper execution of required NDE.
They must inspect all castings in accordance with customer requirements and standards used in the casting industry.
As discussed above, the NRC Inspectors determined that Westlectric has three -
inspectors total, two of whom are only visualinspectors certified to MSS-SP-55, and the I
third is characterized as a Level ll NDE. The two visual inspectors are not trained to ASME Code. Instead, they use the visual acceptance and rejection criteria illustrated in MSS-SP-55. The NRC Inspector observed examples of rejected material that had been.
identified by the visual inspectors such as, coup-drag misalignment, interrupted / cold pour, 1
porosity, and holes.
The NRC Inspectors requested Westlectric's NDE personnel certification' and qualification records. The NRC inspectors were only provided with the Chief Inspector's records,.
which indicated that he was certified as a_ Level 11 MT Inspector, certification dated February 21,1997. The certification stated it was in accordance with SNT-TC-1A,.
"American Society for Nondestructive Testing Recommended Practice," but did not list the applicable edition. The Westlectric staff stated that it would review the matter.
The NRC Inspectors were also informed that a Level lli NDE Inspector from Sun-Ray Testing intomational, incorporated, Downey, Califomia, developed Westlectnc's NDE training program, performed training, maintained the personnel certifications and provided certifications to Westlectric personnel as necessary. The NRC Inspector requested to see the employer's written practices covering all phases of certification including training as required by SNT-TC-1 A. At the time of the exit meeting, Westlectric had not made this information available for review. Unresolved item 99901323/97-01-06 was identified in thisarea.
i c.
Conclusions Westlectric was unable to provide the procedural controls and documentation associated with NDE personnel certification practices.
l,,-
s 3.7 Entrance and Exit '.tc.Gni;is in the entrance meeting on November 17,1997, the NRC Inspectors discussed the scope of the inspection, outlined the areas to be inspected, and established interfaces with Westlectric management. In the exit meeting on November 19,1997, the NRC Inspectors discussed their findings and concems.-
- 4.
PERSONS CONTACTED J.R. Heine :
President R.L Ogden Operations Manager A. Arechiga QA Manager -
G. Kusumi Sales D. O'Sullivan Sales Manager M. Gutierrez Clean Room Foreman / Chief Inspector S. Fericean Shipping Supervisor J. Lietzau inside Sales / Lab Ta$nician R. Young-Core / Molding Supervisor 5.
ITEMS OPENED, CLOSED, AND DISCUSSED 99901323/97-01-01 VIO Inadequate Part 21 Procedure 99901323/97-01-02 NON QA Organization and Program 99901323/97-01-03 NON Documented Instructions 99901323/97-01-04 NON Control of M&TE and QA Records 99901323/97-01-05 NON Control of Specal Processes 99901323/97-01-06 URI Unresolved item - NDE Program l
l i
i i