ML20058C818
| ML20058C818 | |
| Person / Time | |
|---|---|
| Site: | LaSalle  |
| Issue date: | 07/08/1982 |
| From: | Hansing K COMMONWEALTH EDISON CO. |
| To: | |
| Shared Package | |
| ML20058C714 | List: |
| References | |
| NUDOCS 8207260452 | |
| Download: ML20058C818 (25) | |
Text
Attachment H Q.P. FORM 18-1.2 Commonwealth Oh Edison Company Dm M-f QUALITY ASSURANCE MANUAL AUDIT REPORT G.O. Audit of Morrison Construction Company LaSalle Project Type Audit:
/
/ Program Audit
/
/ Product Inspection Point
/ _/ Records
/ y/Special To:
J. Hamilton, Morrison Construction Company - Project Manager Project LaSalle Visit Date 6/29 - 7/81eport Date 7/8/82 1982 System N/A Component Identification N/A Material Description N/A Vendor Morrison Construction Co.
Location LaSalle Subcontractor N/A Location N/A Contacts See Report P.O. No.
181110 Spec. No.
J-2530 Recommended Inspections:
6 mos 3 mos 1 mo Other:
As Scheduled Notes:
See Report Auditor f
Date h
Lee:r. Auditor A K. Ji Hansing ieRey /C))./:amf%
Rev Date 7- $ ~ b'A
/'
Director of QA (Engr-Constr) cc:
Manacer of QA Site Constr. Supt. or Prod.
't,
Engr.
g 4' J Site Quality Assurance Project Manager p,
wa - _ 2 Project Engineering Mgr.
(List others as required)
Manager of Projects hhhh ((
Lead Auditor 8207260452 820719 PDR ADOCK 05000373 1
G PDR
Attachment H AUDIT P.EPORT FOR s
M0 PRISON CONSRUCTION COMPA. Y N
IASAllE PROJECT.
JULY 8,1982 ~
PURPOSE The purpose of this audit was to evaluate Morrison Construction Company's (MCCo) calibration practices and procedures, various quality control records, records control and the Quality Control review functions. Additionally, the audit, inspection and surveillance activities in the area of MOCo calibration were evaluated to detemine if required reviews were performed and to assess the credibility of their reviews and associated documentation.
SCOPE This audit examined the foliowing areas:
1.
GCo audits of MCCo calibration activities 2.
GCo auditor qualifications 3.
Credibility of CECO audits 4.
GCo surveillances of MCCo calibration activities 5.
Authorized Nuclear Inspector's involvement in MCCo calibration activities 6.
MCCo audits of MCCo calibration activities 7.
NRC involvement in MCCo calibration activities 8.
GCo or MCCo NCR's generated which address calibration activities 9.
Requirements and performance of Quality Control reviews
- 10. Dispositioning of out-of-calibration or damaged calibrated tools and equipment.
- 11. A broad scope review of other documentation activities to assess whether the types of documentation problems identified in the torque wrench calibration activities existed in other areas
- 12. Training, skills, and qualifications of MCCo personnel performing calibration functions
- 13. MCCo review of calibrations perfomed by others for them
- 14. Records control
- 15. Accountability of records
- 16. Performance of calibrations
- 17. Calibration procedure adequacy
- 18. Storage and naintenance of calibrated tools and equipment
- 19. A review of all calibration records not recently reviewed by the NRC as part of the investigation.
Attachment H Page 2 AUDIT TEAM The G.O. QA audit team consisted of the following personnel:
Principal Team Members D. A. Brown Auditor QA Supervisor-Braidwood
^
S. M. Jaquez Auditor QA Engineer-Braidwood S. J. Reuteke Auditor QA Engineer-LaSalle R. D. Vine Auditor QA Engineer-LaSalle R. E. Waninski Auditor QA Engineer-LaSalle K. J. Hansing Lead Auditor QA Supervisor-Byron G. F. Marcus Management Observer Director Quality Assurance (Engr /Constr)
Others G. M. Maksimuk Auditor-Data Taker QA Engineer-LaSalle A. M. Monta1+.o Auditor-Data Taker QA Inspector-LaSalle E. A. Kram Auditor-Data Taker QA Engineer-Braidwood R. F. Smeets Observer QA Engineer-LaSalle M. J. Wendell Observer QA Engineer-LaSalle Auditing activities conducted by the data takers and the LaSalle personnel were performed under the direction of a principal team member and were reviewed by an off-site member of the audit team.
PERSONNEL CONTACTED The following personnel were contacted during the course of this audit:
M. Wherry MCCo QC Supervisor D. Kanakares MCCo QC Inspector D. Kozlowsky MCCo QC Inspector P. Granby MCCo QC Inspector B. Hamilton MCCo QC Inspector B. Angell MCCo QC Inspector R. McClosky MCCo QC Inspector T. Harrington MCCo (Former) QC Inspector K. Hamilton MCCo Project Manager B. Bal:
MCCo Head Doc. Clerk J. Terry MCCo QC Inspector J. Bitner MCCo QC Inspector D. Shamblin CECO Proj. Constr. Super.
A. Patak Continental Insurance MCCo. ANI W. Caldwell Hartford Insurance CECO's ANI i
1
Attachment H Page 3 BIRANCE PEETING An entrance meeting was conducted on June 29, 1982 with the following personnel in attendance:
- R. Waninski CECO QA Auditor
- S. M. Jaquez GCo QA Auditor
D. A. Brown CECO QA Auditor
- K. J. Hamilton MCCo Project Manager
- M. Wherry MCCo QC Supervisor
- W. E. Vahle GCo PCD Lead Mechancial Engineer R. A. Braun CECO QA QA Supervisor R. D. Vine GCo QA Auditor
- D. J. Skoza Ceco PCD Engineer D. J. Kanakares MCCo QC Inspector G. F. Marcus CECO QA - G0 Management Observer The scope of the audit was discussed at this time.
denotes present at exit meeting.
EXIT FEETING An exit meeting was conducted on 7/8/82. The audit results were discussed and acknowledged by those present. The following personnel were present at the exit meeting in addition to the personnel indicated above:
A. R. Huffman MCCo QA Engineer R. T. Rose GCo PCD Lead Struct. Eng.
J. J. Maley CECO Manager of Projects
Attachment H Page 4 OVERAll AUDIT ASSESSMENT It is the conclusion of this audit team, based on the extensive review of MCCo quality records, that the documentation problems originally identified by the NRC are limited to calibration records.
It was further concluded that the duplicating of some signatures and data did not appear to be an attempt to falsify quality records but rather was a time saving measure used to handle the numerous, repetitious calibration records as they viewed the calibration functions as a go, no-go type check.
Relative to the adequacy of the surveillance and audit activities, our evaluation indicates that many qualified individuals had examined this area and that without being specifically attentive to looking for record alterations they would not have identified this as a quality issue. Examination of the overall surveillance and audit process, however, does indicate that our auditing methodology involving large numbers of similar records needs to be changed to provide that larger samples be taken to achieve a better assessment. As a result, Quality Assurance has committed to implement this change.
Evaluation of the reports where torque wrenches were found to be out of calibration or had questionable calibration data or records were deemed to have no adverse impact on plant construction. This is based on the records reviewed and the results of independent inspections of bolt torquing by the Independent Testing Agency involving Morrison. The inspection, conducted by CECO.'s Independent Testing Agency, examined 6,756 Morrison installations between 1977 and June 1982. Only 81 of the 6,756 installations were rejected for low torque values during the 6 year period. The reject rate of less than 1.2% supports the conclusion that torque wrench calibration was not a problem and helps explain, at least in part, why this calibration documentation matter had not surfaced.
In addition, examination by Quality Assurance of follow-up checks conducted by Project Construction of other calibration areas has not identified any cases where the installed condition of Unit #1 equipment appears questionable.
e
Attachment H Page 5 SENARY 1.
Assessment of Audit Questions 1-12 In order to ascertain if there were any actions which could have discovered the " deficiencies" in Morrison calibration records which are the subject of this audit, the following areas were investigated.
- 1) Adequacy of site QA audits 6 surveillances
- 2) Involvement of CECO 4 MOCo ANI's
- 3) Trends of applicable CECO 6 MCCo NOl's
- 4) Adequacy of MCCo internal auditing
- 5) Past NRC site inspection activities
- 6) Adequacy of CECO G.O. QA audits The effort was directed at determining if this problem went undetected due to oversight by many parties or a widespread problem not adequately addressed.
The results in these areas can be generalized as follows:
(1) MCCo's ANI and CECO's ANI did not document evidence of examining calibration reports as this was their normal practice in the past.
(2) Region III Incpections (57 in 198161982) did not uncover any problems with MCCo quality control records. Five of the fif ty-seven sitt visits involved review of MOCo quality records.
One of these visits involved an NRC investigation of an allegation of lack of control of quality records.
(3) No CECO NCR's have had to be issued for concerns relating to MOCo calibration records. Therefore, no trends were overlooked.
(4) No MCCo NCR's have had to be issued for concerns relating to calibration records. Therefore, no trends were overlooked.
(5) Morrison internal audits of calibration activities have been extensive and did not identify any calibration record problems.
(6) CECO Site QA surveillance acitivites consisting of 27 calibration surveillances, which specifically covered thirteen calibration records since 1979, were found to have adequate coverage and frequency.
(7) Twenty-four Site CECO QA audits were done on schedule with qualified auditors and recorded objective evidence is sufficient to support the conclusions of those audits. Of the audits conducted since 1979, nine covered site calibration activities.
l l
l l
f Attachment H Page 6 (8) G.O. audits addressed the areas of Morrison calibration activities and did not discover similar calibration documentation problems.
(9) Site CECO QA examined 64 MOCo calibration reports during its surveillances and audits since 1979.
(10) None of the Morrison calibration documents found deficient in the NRC investigation were specifically examined in any CECO audit or surveillance.
In summation, many qualified individuals examined the calibration docunents as well as many other MCCo quality documents in accordance with quality auditing and surveillance plans; however, by the nature of the sampling process inherent in the audit and surveillance process they did not specifically discover the calibration record deficiencies.
Furthermore, without being sensitized to look for record alterations or repetition either by allegations, specific training and experience or as a result of previous problems, it is doubtful that the MCCo calibration document type problems would be identified in the normal audit and surveillance process.
2.
Assessment of audit questions 13-16 In order to determine the adequacy and accuracy of Quality Control reviews of quality records, the following record types were examined:
a.
Calibration records b.
Material procurement documents c.
Material receipt records d.
Installation documentation / travelers e.
NDE reports f.
NCR's g.
Test reports h.
Procedures and work instructions An evaluation was made of the adequacy of the Quality Control reviews l
for the above listed record types by examining the Quality Assurance manual and procedure requirements and by interviewing the Quality Control I
personnel who perform the reviews.
It was found that the requirements and meaning of the Quality Control I
reviews were not clearly delineated in the procedures for several of the record types. As a result of interviewing the MCCo Quality Control personnel perforning the actual reviews, it appears that these Quality Control personnel are knowledgeable in how to do the reviews and overall are doing adequate technical reviews. The lack of delineated requirements for Quality Control reviews is identified as a deficiency in Attachment A of this report.
i l
l l
l
Attachment H Page "i To verify accuracy of Quality Control reviews, a sample of each above listed record type was examined. With the exception of calibration records all of the record types reviewed were found complete and acceptable. The deficiency relating to calibration records is identified in Attachement A of this report.
To determine proper disposition of out-of-calibration or damaged calibrated tools and equipment, MCCo NCR's were reviewed and compared to the MOCo Deleted Instrument Log and Hold Tag Log entries involving calibrated tools and equipment. Through this review it was determined that MCCo.'s Quality Assurance Manual requirement, Section 10.10, was not being consistently followed in that nonconformances were not generated each time a calibrated tool or piece of equipment was suspect and put on hold. The need for issuance of an NCR for suspect equipment became a requirement on October 8,1979. MCCo Procedure PC-31 does not include this requirement and thus appears to be a contributory factor in the inconsistent issuance of NCR's for suspect calibration tools and equipment. This deficiency is identified in Attachment A of this report.
3.
Assessment of Audit Questions 17-22 A review of all MCCo calibration reports, not previously reviewed by NRC, was performed to determine the types and extent of documentation problems. This review identified a generic problem relating to reproduction of signatures and data and is identified in Attachment A of this report.
Also, in an effort to determine if the problems found in the calibration area were present in other MCCo Quality Control documentation, a review of the following documents was performed:
a) Receiving inspections b) Weld data, and traveler packages c) As-Built drawings with checklists (small bore) t 1
l
Attachment H Page 8 d) Component support checklists e) Mechanical revision directives f) Na's g) N m Log h) Final line walk reports
- 1) Pressure test reports j) A9E Code Data Reports (CECO Supplied) k) NDE Reports
- 1) Quality Control surveillances m) On-Site audits n) Purchase requisitions o) Welder qualifications p) Quality Control inspector qualifications q) Calibration personnel certifications This extensive review found only isolated instances of the use of white-out, incomplete information and stamping of Quality Control approval signatures. Also, the isolated cases of white-out identified were judged to be corrective actions and not an attempt to falsify.
Furthermore, the use of white-out on quality documents was not specifically prohibited at the site until February 1982 when a letter to this effect was sent to all site contractors as a result of a recent NRC penalty assessment against the Zimmer Project. As a result, it is the conclusion of this audit that the documentation problems only involve the calibration area and the quality control document review activities of a specific Quality Control Inspector. Also, one case was identified during the review of Quality Control Inspector qualifications where the certification of an inspector was not signed but otherwise appeared in order. This deficiency is identified in Attachment A to this report.
Since most of the problems identified in the calibration area were traceable to one Quality Control individual, an expanded review was performed of other records processed by that person. A selection of records from the following record types reviewed by that MCCo Quality Control Inspector were examined:
l l
a) Receiving and storage inspections b) NCR's c)
Installation records d) Receiving reports e) Heat treatment records
(
f) Purchase requisitions g) Pressure test memoranda 6 test results
Attachment H Page 9 e
This examination found no repetitious occurrences of the documentation problems noted in' the calibration area. This further supports the audit conclusion that the generic documentation problems are isolated in the area of calibrations.
4.
Assessment of Audit Questions 23-26 A review of training and qualification records for the mi11 wrights and fitters performing calibrations indicated that all were properly trained and the training was documented. The training appeared adequate and further the required skills were within craf t capability. Eye examinations of Quality Control Inspectors performing calibration activities were-reviewed to assure they had required vision capabilities and all were found acceptable. Additionally, the millwrights and fitters performing calibrations were eye tested for near and far vision acuity and all passed. An examination of the areas where calibration is performed was conducted to assure that adequate lighting was present; all areas were found acceptable.
A review of MCCo's handling of calibrations performed by the Operating Station Maintenance Department indicated that MCCo does not have adequate guidelines to assure review of test results for acceptability of the calibration. This deficiency is identified in Attachment A to this report.
5.
Assessment of Audit Questions 27-33 MCCo has established and inplemented measures to control access to the documentation vault by using a listing of personnnel authorized to remove documents from the vault and requiring logging of documents being taken in or out and identification of the person involved. Also, since June 29, 1982, a security guard has been posted at the vault to countersign issuance of documentation, monitor the review of such issued documentation to assure no alterations are made and assure that Unit #1 and common documentation with Unit 2 is only reviewed in the presence of the guard. Additionally, all Unit #1 documentation in the vault is secured with locks on the file cabinets. This special security arrangement is to be maintained until conclusion of the NRC investigation.
Prior to the actions taken on June 29, 1982, the CECO Project Construction Superintendent had verbally communicated to the MCCo Project Manager, on June 24, 1982 that actions should be taken to insure that no documentation would be altered. This requirement was communicated to the Quality Control personnel verbally on June 24, 1982 and again on June 25, 1982 in the form of written correspondence from the MCCo Quality Control Supervisor to the Quality Control personnel. During the week of June 21, 1982 when the MCCo Quality Control Inspector involved in the calibration documentation problem became aware of the fact that duplicated signatures were unacceptable on calibration reports, he decided to go back and initial those calibration reports he had issued with his signature duplicated. This. initialing of signature's which occurred for about two hours ended on June 24, 1982.
It was the auditor's opinion that the addition of the initials was done to correct an onission as opposed to falsifying records.
l
_ 1 Page 10 In the area of records accountability this auditor determined that MCCo.'s logging systems are adequate to verify accountability of records. However, there was no documented evidence that MCCo has performed reviews to determine if documents are missing. Also, current procedures do not specifically provide instructions for replacing lost records. This deficiency is identified in Attachment A to this report.
6.
Asessment of Audit Questions 34-38 A review of the storage and maintenance of calibrated equipment and tools noted that they are kept in an orderly manner and in locked cabinets. Observance of millwrights and fitters in the performance of calibrations indicated they were knowledgeable and competent in their jobs. One procedural deviation was noted with the calibration of torque wrenches and is identified in Attachment A of this report.
The following procedural deficiencies were identified regarding calibrations:
Acceptance criteria and calibration points are not stated a.
for each type of tool or instrument.
b.
Not all calibrated equipment is addressed in procedure PC-31.
The specific deficiencies are identified in Attachment A of this report.,
For the majority of the calibrations witnessed, the procedure and actual performance of the calibrations was determined to be acceptable.
Upon interviewing personnel performing calibration and reviewing calibration histories it was concluded that, in instances where the actual test data was extremely close to the established test reading, the i
actual values may not have been recorded in all cases; rather, the generic test point values were recorded.
In other words, a go, no-go check was performed. The interviews did, however, indicate that out-of-tolerance calibration readings were identified. The deficiency relating to recording of actual data is identified in Attachment A to this report.
1 S
I 1
Attachment H Page 11
Response
Morrison Construction Company is requested to respond in writing to the Findings and Observations listed in Attachment A to this report by July 23, 1982. The response must include the following:
1.
Action taken to correct the deficiency 2.
Action taken to prevent recurrence 3.
Data for full compliance Please address your response to:
Commonwealth Edison Company G. F. Marcus Director of Quality Assurance (Eng./Constr.)
P.O. Box 767 Chicago, IL 60690 With copies to:
Commonwealth Edison Company W. J. Shewski Manager of Quality Assurance P.O. Box 767 Chicago, IL 60690 and Commonwealth Edison Company K. J. Hansing QA Supervisor - Byron Station Byron, IL 61010
Attachment H ATTACINENT A FINDING #1 Contrary.to 10CFR50 Appendix B Criteria XVII the Quality Control final review and acceptance of calibration reports was not adequately implemented to assure complete and acceptable calibration records.
Discussion An examination of completed and accepted calibration reports identifled the following problems:
a)
Wrench type was not indicated on torque wrench #10894 calibration reports dated March 16, April 3 and September 25, 1981.
b)
Quality Control review of torque wrench #97366 calibrated on July 31,1981 was not documented.
c)
Error measurements were not indicated on the report for torque wrench #97366 calibrated on January 2, 1981. The error values, however, were attainable from information on the report and found acceptable.
d)
Test data was not recorded for pressure gauge #PG-31 calibrated on June 8, 1982.
FINDING #2 Contrary to MCCo Quality Assurance Manual Section 10.10, nonconformance reports have not been generated for disposition of each calibrated tool or instrument found suspect.
Discussion A review of MCCo's Deleted Instrument Log, Hold Tag Log and NCR's indicated the following torque wrenches as damaged or out of calibration but records of NCR's for each was not available:
a)
Torque wrench #4292 - Hold Tag #1106 - Torque wrench was scrapped instead.
b)
Torque wrenches #15758 and #10468 - hold tag #1223 - Torque wrenches naaded to be repaired.
(The Hold Tag Log showed iten was repaired and the Hold Tag removed.)
c)
Torque wrench #E40468 - hold Tag #1171 - Torque wrench was damaged while being used.
(Hold Tag Log showed item was repaired and the Hold Tag removed.)
Attachment H Page 2 Attachment A Other instances of hold tags being issued on calibrated torque wrenches or torque screwdriver without references to applicable N G's are as follows:
Hold Tag s # 1131, 1132, 1133, 1134, 1108, 1129, 1092, 1103.
Furthermore, a review of procedure PC-31 Rev. 3 dated July, 1977 indicates that the requirement of Section 10.10 is not addressed and thus could be a contributing factor in the inconsistent implementation of NG's for suspect calibrated tools and equipment.
FINDING #3 Contrary to 10CFR50 Appendix B Criteria XII, MCCo has not established and implemented adequate guidelines for receipt inspection review of calibration reports as to acceptability of calibrations performed by outside agencies.
Discussion Five calibrations were accepted for the torque wrench tester where the tester was out of calibration. CECO NCR #599 has been issued for engineering disposition.
FINDING #4 Contrary to 10CFR50 Appendix B, Criteria XII and V, MOCo Procedure PC-31 has the following deficiencies:
a)
Not all calibrated tools 6 equipment used by MCCo are covered by this procedure.
b)
The procedure does not address acceptance criteria for all the calibrated tools and equipment and c)
The procedure does not always address the calibration method and specifically the points of calibration where applicable.
Discussion a)
A review of calibration equipment indicates that some tools and equipment have not been included in procedure PC-31. These items are:
- 1) Deadweight tester
- 2) Torque wrench tester
- 3) Densitometer
Attachment H i
Page 3 Attachment A f
b)
The following items, although listed in MCCo procedure PC-31, did not have specific calibration acceptance criteria listed (ie. acceptance tolerances). The auditor was told that the manufacturers recommendations were used as the basis for acceptance.
i
- 1) Gauge blocks
- 2) Multi point recorder
- 3) Micrometers
- 4) Linear measuring devices
- 5) Optical level
- 6) Ameter
- 7) Standard microneter lengths
- 8) Standard verniers In addition, the following list of items were found to have acceptance criteria discrepancies or lack clarity.
9)
PC-31 states that dial indicators only need free movement of the plunger, but calibration Certification fom PC-61 indicates tools should be within 0.001" of a standard.
- 10) PC-31 indicates pressure gauges are to be calibrated within
+ 5%. While witnessing pressure gauge calibration demonstration, MCCo personnel were noted to be divided as to whether this means 5% of full scale or each incremental calibration value.
c), While PC-31C was noted to be very specific about data recording for torque wrenches, the following inconsistencies were noted in other areas:
1)
Micrometer (S/N QC-5) was tested on June 4, 1982 at six settings, whereas on March 4, 1982 it was tested at thirteen settings.
l 2)
Test settings for pressure gauges are listed on Calibration Certification form PC-61, but are not indicated in l
procedure PC-31.
I FINDING #5 Contrary to 10CFR50 Appendix B, Criteria XII and XVII, it was an l
accepted MCCo 3ractice not to record the actual test data on calibration records when tle data was extremely close to the standard value used.
Discussion Through interviews with the personnel performing and witnessing the calibration activities and a review of the historical records, it becane evident that MCCo did not always record the actual test data readings when they were within acceptable limits. The generic test point values were recorded. A go, no-go check was performed. The interviews did conclude, however, that out of tolerance calibration readings were being identified.
Attachment H Page 4 Attachment A FINDING #6 Contrary to 10CFR50 Appendix B, Criteria XII and XVII, Morrison Construction Co. is not adequately controlling calibration documentation for measuring and test equipment.
' Discussion A review of all calibration records on file at Morrison was conducted to deternine the extent of documentation deficiencies. The deficiencies involved (1) the frequent use of reproductions of calibration report forms already signed by the cognizant Quality Control Inspector and Witnesses to record calibration data, (2) the incompleteness of data, dates and signatures on calibration reports and (3) the use of reproduction of calibration data on consecutive reports.
FINDING #7 Contrary to 10CFR Appendix B, Criteria IX and XVII and Morrison procedure PC-41 Rev. O, three (3) voltage / amperage (V/A) readings recorded on Morrison Quality Control surveillances were found to be out of procedure tolerances and there was no evidence of corrective action.
In addition, there were instances of incomplete and/or incorrect information on the surveillance forms, such as lack of proper procedure revision and V/A ranges listed on the form which were not consistent with the procedure ranges.
Discussion a) The following reports had the V/A readings out of procedure tolerances.
Report Date Reading Procedure Range Observed Procedure 4/24/81 Amps60-250 55 amps P1-3LS Rev. 7 6/1,2/78 Amps60-250 50 amps P1-11LS Rev. 7 12/2/77 Amps60-110 125 amps P1-20LS Rev. 4 The audit team subsequently reviewed the three above cases with a Sargent 6 Lundy Welding Engineer. This discussion included a review of the associated weld procedures, the materials being welded, and the weld inspections performed. Based on this review, it did not appear to this SSL Engineer that a quality concern exists.
Attachment H Page 5 Attachment A b) In addition, inaccurate recording of surveillance information was identified. Out of 343 surveillance reports reviewed,104 had incorrect or no procedure revision identified on the report form and 95 had V/A ranges listed which were not consistent with the ranges provided in the applicable procedure.
OBSERVATION #1 Quality Control final review requirements on quality records are not clearly defined in MCCo procedures or site instructions.
Discussion The following record types receive a final review by MOCo Quality Control personnel but the meaning and requirements of this review are not delineated:
a)
Calibration records b)
Installation documentation / travelers c)
NDE reports d)
Test reports Additionally, the final Quality Control review of the Weld Data Record is performed, but documentation to indicate acceptance, other than entering selected information from it on the Status Log, is not provided.
OBSERVATION #2 MCCO has not performed reviews of records to determine if docunentation is missing. Additionally, current procedures do not provide instructions for replacing lost records.
Discussion MCCo's systems of logging docunentation appear adequate as a means of accounting for documentation but no evidence was available to indicate that MCCo had performed reviews for missing documentation. MCCo also has l
not established steps to be taken to replace documents discovered missing.
Attachment H Page 6 Attachment A OBSERVATION #3 Mi11 wrights were observed calibrating torque wrenches with some variation from procedure PC-31.
Discussion While observing millwright and pipefitter demonstrations of the calibration of various tools, it was noted that " clicker type" torque wrenches were manually calibrated. A review of procedure PC-31 indicates that these wrenches are to be calibrated using the mechanical lead device which is built into the torque wrench tester rather than be pulled by hand.
The following canments are relevant to this observation:
1)
The pipefitters and mi11 wrights performing the various calibration functions appear to be competent.
2)
Although use of the mechanical lead device would seem to make the recording of calibration data more accurate, the results obtained by manually calibrating the " clicker type" wrench were not substantially different.
3)
The mechanical device on the torque wrench tester can not accommodate large (500 ft-lbs) torque wrenches.
OBSERVATION #4 MCCo Tool / Instruments Inventory listing is not current.
Discussion Gauge blocks (S/N bel-9) are not listed on MCCo Tool / Instruments Inventory List. Also, as a comment, the list should be updated as several hand written entries were noted.
{
l OBSERVATION d5 A Quality Control Inspector certification was not signed by the MOCo certifying int.ividual.
Discussion A review of certifications for all Quality Control Inspecters identified the omission of the Certification signature for J. R. Ragan, Level II Limited. Other than the missing signature, the information on the certification form and back up documentation appeared to be in order.
I
Attachment H LASALLE COUNTY UNIT I & II REPORT OF:
ACTION TAEDT BY PROJECT CONSTRUCTION AS A RESULT OFNRCANDC.E.Co./Q.A.CONCERNSRELATIVETO MORRISON CONSTRUCTION COMPANY CALIBRATION RECORDS.
DATE:
JULY 8, 1982 REPORTED 3Y:
N DAN SK0ZA O
PROJECT CONSnuedON DIGINEER REVIEWED BY:
wb VARREN VAHLE PROJECT CONSTRUCTION LEAD Y2CHANICAL DIGDIEER l} ig APPROVED BY:
}
eg. rater / -
hA!AGER OF PROJECTS
Attachment H SUWARY This report seavizes the confirmatory measures performed by Project Construction to verify the adequacy of the field work done by Perrison Construction Company which required the use of calibrated torque vrenches and hydro test pressure Eauges. Our review sub-stantiates our conclusion that bolts were properly tightened and pressure systens were properly measured. Exceptions that were found represent small percentages of the sample. It is not evident that these exceptions are traceable to calibration deficiencies.
All exceptions discovered have been corrected or are otherwise addressed in this report. Therefore, we believe that the plant construction is adequate and satisfactory.
e
~ ~ -
Attachment H s
REPORT I.
TORQUE VRENCH curnnCATIONS:
Of the approximately 6,000 MCCO torque wrench certifications the following Minor and Major catorgories were identified.
A.
Specific Concerns Deemed Minor 1.
Torque wrench type misidentified or not stated.
2.
Math errors on certification but wrench within tolerance.
3 Error calculations not perfomed but wrench within tolerance.
h.
Vissing QC signatures or report date.
B.
Specific Concerns Deerad Major 1.
Single zeroxed report - data and all signatures.
2.
Group xeroxed report - data and all signatures (1-6weekperiod,1-12weekperiod,1-3weekperiod).
3 Missing tester or witness signature.
h.
Inco=plete or uncertain data at isolated data points.
Si. Missing reports.
C.
Torque Vrench Out-Of Tolerance Concerns Certifications identified with out-of-tolerance readings at isolated data points and had been accepted by MCCO Q.C.
D.
Generic Concerns 1.
Numerous series of consecutive reports with identical original test data indicating actual test data was notrecordedbutago/nogotestwasutilizedand nomini values recorded when the data point was deemed within calibration tolerance.
2.
Numerous cases of Q.C. acceptance stamp xeroxed onto calibration data sheet already signed. Only date of Q.C. review and page nu=ber was affixed to indicate Q.C. acceptance review.
l Attachment H The specific concerns deemed minor required no remedial field checks of the work done with the subject wrenches. In all cases the concern is a paper problem. Other infomation existed on the certification to correct the concern, or a calculation was easily perfomed to correct the concem, or the Q.C. review was performed on the calibration certification.
The specific findings deemed major and the out of tolerance wrenches required ascertaining the safety related work perfomed with the wrench in question and field checking the torque for acceptability. Q,hanscrewithapproximatelyJj_2,CEA'sandbolting 2
on 2 valves were required to be rechecked. The results of the re-checks indicated 1h CEA bolts on 9 hangers were found lower than the specified test torque. These bolts will be retorqued or re-placed per LS-CEA requirements. The first h of these hanger as, se:blies were recalculated and found that they would function with-out the subject CEA's. The remaining corrective action will be ce=pleted by July 12, 1982.
The basis for accepting the installed work given the generic findings are the following:
A.
ICO general calibration procedure PC-31 only requires a 6 month frequency for torque wrenches utilized on all work except CEA's.
CEA work required weekly or monthly calibra-tions of torque wrenches. Calibration certifications with actual data exist to meet the 6 month frequency intervals.
B.
It has been shown that torque wrench accuracy is not icportant for CEA work. The primary reason for specifying a torque value in CEA work is to assure an auditable method of ascertaining that the CEA is set in the concrete. Once the anchor has been shown to be set, torque tightness of the CEA becomes irrelevant. Thisbeingthecase,thego/
no go calibration test utilized should be sufficient to ascertain that the torque wrench is accurate enough to set the CEA. An independant check performed by ConAm Inspection Agency with their own calibrated torque wrenches on a mini-mum of one anchor per hanger assemblies but at least one out of each ten expansion anchors installed in that assembly.
showed no excessive rejection of CEA's throughout the work history. The rejection has been extremely low; on the order of 1% of the bolts tested.
e
I Attachment H C.
Evidence exists that torque wrenches were discarded or returned to the manufacturer when calibration data showed the wrench out-of-calibration.
D.
Evidence exists that the torque wrench calibration certi-fication did receive a Q.C. review even though the Q.C.
stacp and signature were zeroxed. The date of review and page number were original writings.
In su= mary, we have concluded that the torque wrenches were sufficiently accuzste to assure the installed work meets its fune-tional requirenents.
II.. TORQUE VRENCH TESTER CALIBRATION:
During the NRC review a concern was stated that the tester utilized to calibrate the torque wrenches was out-of-tolerance on specific scale readings. Co=monwealth Edison Co=pany Nonconformance Report 599 documented the calibration discrepancies. Preliminary reviews have indicated the following:
A.
Variables such as the tester condition and operator variance probably caused the erroneous readings in the ranges identified.
B.
The torque tester or torque wrench calibration tolerances are not the primary concern in torque operations. Tolerances can be relaxed in most situations with assurance the installed bolt condition is acceptable.
On the above basis, no checks of the installed work has been initiated.
III. HYDRO PRESSURE GAUGE CALIERATION CERTIFICATIONS:
Approximately 1000 MCCO Hydro Test Pressure Gauge Calibration Certifications for 100 gauges were reviewed by C.E.Co. Q.A. auditors.
The known concerns are the following:
e Attachment H A.
Specific Concerns 1
Partial or no calibration data listed.
2.
Nowitnessand/orQ.C. signatures.
3 Ilate. of calibration in error.
15 Out-of-tolerance readings accepted by Q.C.
5.
Calibration missing.
6.
Tester and Q.C. signature not original.
2.
Generic Concerns 1
Actual test data was not recorded but go/no go test utilized and nominal values recorded when the data point was deemed within calibration tolerance.
2.
Uitness and Q.C. signatures xeroxed onto calibration data sheet. No objective evidence exists on calibra-tion data sheet indicating tests were witnessed.and Q.C. reviewed.
We specific concerns can be resolved without re=edial field checks.
The questionable gauges were not utilized at all, or not utilized in the ranges with the problem data.
Se Generic concerns require several different explanations and actions to resolve them:
1 Variations in industry instrument standards and dial reading techniques which include paralax effects tend to vary the readings taken during instrument calibrations.
In some cases, time was not taken to log the actual readings if it was apparent that the values were within the specified tolerances.
In these cases, the nomini settings were apparently logged. Reviews of the calibra-tion files for each instrument and successive calibrations of the same instrument have shown only several instances of instruments out of tolerance and only within a small part of their range. These reviews substantiate that the readings taken were of sufficient accuracy to use as a basis for the certifications. Additionally, records indicate 35 hydro test pressure gauges out of 107 vere discarded during the course of work when calibrations could not be cocpleted within specifications.
/
(
Attachment H 2.
The establishment of validity of the zeroxed witness and Q.C. signature can be done through the Rydro Test Pressure Gauge Usage and Calibration Log and by virtue of the fact that cause calibration is a two(2) man operation. For each gauge, a log is kept identifying the dates the instrument was calibrated and the test number corresponding to that date. The Q.C. inspector that witnessed the test then signs the pressure gauge log indicating the test was completed. 'Lhis was done with original signatures on a 100% basis. The testing apparatus requires two people to operate, one to pump the machine and read the pressure gauge and another to assure the dead weights are free floating. This has been done historically by a Pipefitter and Q.C. inspector.
On the above basis, it was not necessary to establish a program to recheck previous pressure tests.
The containment valve structural bolting review has been co=pleted.
- 17. All bolts on the safety related valves have been torqued to within the manufactures' recommended values.
l
______ _ _ _ _ - _ - _ _ _ _ _ _ _ - _ _