ML20198C711
Text
-
1.
Allegation Category: QA/QC 8, As-Built 2.
Allegation Number: AQ-135 3.
Characterization:
It is alleged that the post construction verification program (PCVP) was conducted in such a manner that it failed to adequately process the results of quality control (QC) inspections as required by 10 CFR Part 50, Appendix B.
It is alleged that the scope and depth of the program were inadequate, that reviews of identified deficiencies were in-adequate, and that management of program results was inadequate.
The PCVP was alleged to be:
(1) a " blitz inspection" consisting of an
" instant review and an instant fix," (2) of narrow scope, encompassing only a few cable trays, (3) an inspection program with no requirement to identify problems which were fixed on-the-spot, and (4) a program that was stopped prematurely in February 1984.
The alleger maintained that the PCVP was a " violation of law" that lacked data on engineering review of identified problems and a method of tracking and trending deffciencies.
The alleger asstrted that the PCVP was "out of control" and that the facility constructor, Brown and Root, Inc. (B&R), was " turning over a mess" to the owner, Texas Utilities Electric Company (TUEC).
4.
Assessment of Safety Significance:
The NRC Technical Review Team (TRT) found that in FSAR Section 8.1.4.3, TUEC committed to implementing Institute of Electrical and Electronics Engineers, Inc. (IEEE) standard 336-1971, " Installation, Inspection, and Testing Requirements for Instru-mentation and Electric Equipment During the Construction of Nuclear Power Generating Stations." Section 6 of this standard requires that post-construction verification consisting of the inspection of installed equipment and systems shall be inspected to verify the following:
(1) that equipment and materials have not sustained damage during install-ation, (2) that good and proper workmanship has prevailed, (3) that the installation has been made in accordance with specified requirements, (4) that all nonconforming items have been satisfactorily resolved, (5) that appropriate protective measures are applied for lay-up after installation, and (6) that all temporary conditions such as jumpers, bypass lines, and temporary set points have been clearly identified so that subsequent restoration can be ascertained prior to placing the items in service.
The TRT found through interviews with the managers, supervisors, and inspec-tion personnel associated with the PCVP that TUEC chose to implement IEEE-336 with a series of post-construction inspections of instrumentation, electri-cal equipment, and raceway.
These post-construction inspections followed inprocess and final inspections and were made prior to a plant area being turned over to the startup testing organization. Therefore, the post-construction inspections reexamined previously inspected equipment and raceway.
i e605230041 840512 0 299 PDR f DIA GARDE 05-59 PDM
TUEC's written description of the PCVP was contained exclusively in admi-nistrative and implementing procedures (References 7 through 10).
The TRT found no higher-tiered post-construction inspection plan to describe the scope, control, and management of the PCVP.
The TRT reviewed these pro-cedures as a part of its. examination of this allegation.
The PCVP was exclusively an electrical inspection activity to satisfy IEEE-336. TUEC managers and supervisors involved in the PCVP indicated to the TRT that other post-construction inspection activities were con-ducted in a similar manner, i.e., they occurred after final inspection.
These other activities included:
(1) the American Society of Mechanical Engineer's (ASME) NA stamping verification program conducted by B&R, (2) post-construction inspections of Hilti anchor bolt separation, and (3) post-construction verification of baseplates, grouting, and Richmond inserts.
TUEC managers also indicated that startup testing personnel made additional room and system walkdowns prior to testing, as did plant operations per-sonnel prior to TUEC's formal acceptance of these roc;ns and systems.
These walkdowns were not considered to be inspections, and were not neces-sarily conducted by personnel certified as inspectors in accordance with American National Standards Institute (ANSI)/ASME N45.2.6-1978; however, these walkdowns served to identify and control the correction of any re-maining identifiable deficiencies.
Since the PCVP was related to electrical inspections and since the alleger's concerns were also related to electrical inspections, the TRT concentrated its review on those post-construction inspections implementing IEEE-336 and on the final walkdown and accept.ance by TUEC plant operations.
Based on its review of TUEC's written description of the PCVP, the TRT found two principal programmatic weaknesses.
These were:
(1) lack of a program plan for the PCVP, and (2) lack of written guidance on the type of level of deficiency which required issuance of a particular corrective action report.
TUEC complied with IEEE-336 by implementing procedures QI-QP-11.3-40,
" Class IE Electrical Post-Construction Verification," and QI-QP-11.8-14,
" Instrumentation System Turnover." Nonetheless, the first weakness indi-cated the lack of a management program to define the scope of the post-construction verification program.
The lack of a program plan could have led to the perception of inadequate scope, as maintained by the alleger.
The second weakness in the PCVP was considered to be significant by the TRT and to have broad generic implications'related to a lack of defined thresholds for triggering corrective actions.
In ascending corrective action systems available to TUEC personnel as a p, order, the art of the PCVP and other inspection activities at CPSES were: (1) line items on punchlists, deficiency reports (DRs), and deficiency lists (DLs), which were attached to inspection reports (irs); (2) unsatisfactory irs, (3) nonconformance reports (NCRs); (4) corrective action reports and 0-300
L J
corrective acti.on requests (CARS); and (5) reports submitted to the NRC in accordance with 10 CFR Part 21 and 10 CFR Part 50.55(e).
During the review of TUEC's PCVP, the TRT found no guidance other than
-general definitions of how to initiate a higher-tiered corrective action system for an identified problein.. In particular, the TRT found minimal guidance concerninr, the types of levels of unsatisfactory IR attributes that required ger.eration of an NCR.
In TUEC procedure CP-QP-18.0, the guidance on in tiation of NCRs from irs was limited to the following note in Section 3.2:
" Note:
Deficiencies for items which have been accepted by final inspecticn (unless an IRN [ item removal notice] has been issued) as delineated by Quality Instructions, or deficiencies for which
'use-as-is' or non-standard repair disposition is required, shall be docuriented in accordance with Reference 1-A."
Reference 1-A was TUEC procedure CP-QP-16.0, "Nonconformances." The guidance in this procedure was limited to the definition of noncon-formance as, "a deficiency in characteristic, documentation, or procedure which renders the quality of an item unacceptable or indeterminate."
During interviews with PCVP personnel, the TRT found a nearly universal understanding within the PCVP that an NCR was required for those deficien-cies for which an engineering evaluation was necessary.
In addition, most personnel interviewed ' stated that an NCR was initiated for those deficien-cies that'could not be corrected using normal craft procedures.
- However, as discussed above, only minimal guidance for NCR initiation was contained in TUEC procedures CP-QP-16.0 and CP-QP-18.0.
Evidently, this minimal guidance was insufficient because inspectors listed identical deficiencies
(.
on NCRs in some cases and on DLs, only, in others.
The TRT considers this issue to be significant for the following reason.
Deficiencies listed on NCRs were subject to evaluation by Engineering,
. Quality Engineering, and management, but deficiencies that were listed only on DLs or irs did.not receive the same level of evaluation.
Specific deficiencies listed on DLs or irs may have been corrected individually; however, the absence of engineering and management review could have resulted in failure to identify and resolve the more generic concern.
For further discussion of NCRs, see QA/QC Category 50.
For a discussion of hardware deficiencies that were not identified and resolved by the quality program, see QA/QC Category 8, allegation AQ-50.
A further indication of the lack of effectiveness of some of TUEC's pro-cedures on nonconformances appeared in an error in Revision 20 to CP-QP-16.0.
Section 3.2 of this procedure described the processing of irs.
The third paragraph of Revision 20 to this procedure, issued August 9, 1984, stated:
0-301 u
E F
-When all unsatisfactory conditions have been corrected and reinspected and accepted, or when an NCR or superseding IR has been prepared for the unsatisfactory condition (s) per-approved procedures / instruction, the QC Inspector shall Block 16). The closed IR shall then be forwarded to the PPRV by-the IR clerk.
Evidently, some portion of the first sentence was missing from line 4.
The TRT noted this error on October 2,1984, and discussed it with TUEC management, who confirmed the error. They also noted that the same error' had been present in Revision 19.
Such an error in a procedure was not surprising; however, its uncorrected existence through two revisions issued two months apart caused doubts regarding the usefulness of this procedure among the personnel who generated and processed irs.
The TRT reviewed TUEC's program for trending nonconformances which were generated as a result of the PCVP.
The trending program was described in CP-QP-15.7 and CP-QP-17.0, and the results were documented in quarterly trending reports issued by TUEC's site QA organization.
The TRT found that the description of analytical considerations of the trending program was weak in that the procedures listed above failed to include:
(1) a discussion of data normalization to account for the effects of varying i
construction or inspection activities, (2) guidance on the characteriza-tion of nonconformances, (3) methods for determining fundamental underly-ing causes of nonconformances, and (4) guidance for using the trend ana-lysis results to initiate CARS.
l The allegation that the PCVP was not trended was incorrect; however, the quality and uses of this trending were weak. The TRT found that the quarterly trend reports contained little more than gross statistical categorizations of different types of deficiencies in the form of tables and bar graphs.
A typical categorization contained in the fourth quarter 1983 trend report identified a total of 1168 deficiencies (through the PCVP) with the following breakdown:
Work Incomplete /Not per Requirements 778 Procedure Violation (Upgraded to NCR) 0 Poor Workmanship / Improper Hardware 8
Send Radius Violations O
Color Code /ID Violations 288 Damaged Equipment / Cables 94 Raceway Unacceptable for Pull 0
Hilti Bolt Violation /No Sealant 0
The TRT did not find assessments of the data or discussions of any deter-mination of the root causes of significant nonconforming conditions; however, the quarterly trend reports did contain a tabulation of deficien-cies identified by the PCVP.
0-302
~
,,_.--y
The TRT examined TUEC's program for final acceptance of systems and areas by plant operations.
This program was described in STA-802 and was designed to take place after all final QC inspections and all post-construction verification were completed, and following testing by the startup organization. The program included a review of master data base items, a review of test results and outstanding deficiencies, a verifica-tion of the accuracy and adequacy of drawings needed to support operations, and system and area walkdowns.
Walkdowns were conducted by operators, instrument and control technicians, electrical and mechanical maintenance personnel, radiation protection per-sonnel, and plant engineers from the plant operations organization. TUEC personnel stated that walkdowns were not intended to serve as a final con-struction QC inspection, but were intended to take advantage of the per-spective of plant operations personnel to identify those items requiring resolution as a part of TUEC's final acceptance of the facility.
Items identified during the STA-802 walkdowns included hardware condition pro-blems, problems with equipment access for operability and maintainability, and potential problems from a radiation protection standpoint.
Items identified as a result of the STA-802 program were documented, tracked, corrected, and reviewed by managers and supervisors.
The TRT found that the STA-802 program went beyond regulatory requirements and was an effect-ive mechanism for ensuring the operability of the facility prior to fuel load.
To evaluate the effectiveness of the PCVP and to determine the merits of the allegation, the TRT reviewed:
(1) 22 irs associated with the PCVP; (2) attachments to the irs, including punchlists, post-construction deficiency reports and lists, electrical equipment punchlists, and deficiency reports; (3) selected NCRs; (4) quarterly trending reports for 1983 and 1984; and (5) CARS written during 1983 and 1984.
As a result of its review of the PCVP documentation, the TRT observed that irs were sometimes generated later than the attached ors, punchlists and other attachments. Two examples'of irs with this characteristic were numbered PCV-1-0028616 and PCV-1-0016954. TUEC personnel stated that during the early stages of the PCVP, irs were not always initiated at the start of a post-construction inspection for a particular room.
In some cases irs were generated after completion of a room inspection.
In the latter instances, the post-construction inspection was repeated for those rooms. TUEC's practice at the time of the TRT review was to initiate an IR at the start of the inspection.
In all cases examined by the TRT, irs were closed after all unsatisfactory attributes had been corrected and satisfactorily reinspected by QC.
Because the unsatisfactory IR was the administrative vehicle for tracking correction, reinspection, and closeout of identified deficiencies, the lack of timely generation of some irs lends credence to the allegation that the PCVP discrepancies were not always tracked.
However, the TRT review indicated that TUEC strengthened the program to include timely initiation of irs.
Furthermore, in November 1983, TUEC shifted tracking 0-303
i r
esponsibilities for unsatisfactory irs to the Paper Flow Group (PFG).
r JThe TRT's review indicated that the tracking of irs was satisfactory.
A TRT review of the PFG administrative control system showed that the PFG was not only tracking unsatisfactory irs, but was also tracking the cor-rection,' reinspection, and closeout of individual deficiencies using the computerized master data base (MDB).
The TRT determined that the MOB et the time of the TRT review was used as a mtnagement tool;that was updated continually.
The TRT observed that there was no documented QC supervisory review of irs' and attachments.
The TRT found that the IF form did not contain blanks
-for a QC lead inspector or supervisor to dccument his or her review wi.th a signature and date.
In some cases, individual supervisors documented a review by signing in an open space at the bottom of the IR form. TUEC's IR procedure, CP-QP-18.0, did not describe supervisory review of irs.
The TRT found that during the 19 revisions to QI-QP-11.3-40, 8 different-DR and punchlist forms were promulgated to list the individual deficiencies and to serve as attachments to an IR.
None of the forms attached to the various versions of the procedure contained blanks for signatures or dates to denote a QC supervisory review.
However, some ~of. the printed DR forms used by TUEC did contain blanks for QC supervisory' review. The TRT deter-mined through a review of records that these signature blanks were not always used. Procedure CP-QP-18.0 did not address QC supervisory review
~
of irs; however, TUEC QC supervisors stated that they reviewed the irs'and attachments as part of their normal practice.
The.TRT considered TUEC's lack of a formal mechanism to document QC super-visory. review of irs to be a weakness that became more significant when coupled with the lack of guidance of the level of a deficiency that would require. initiating an NCR.
The lack of a strict requirement to conduct and document QC supervisory review of irs may have contributed to the incon-sistent reporting of identical deficiencies on NCRs in some cases but only as punchlist deficiencies in others.
The TRT observed the lack of a consistent administrative control system to ensure that all required attachments were part of a given IR. During a review of TUEC's written program, the TRT found no description of a method which ensured that all required punchlists and DRs remained attached to an IR.
Some individual inspection groups sequentially numbered each attach-ment and denoted the total number of attachments on the IR, but other irs reviewed by the TRT contained several attachments, each labeled "page 1 of 1," with no indication of the total number of DR or punchlist pages. TUEC supervisors stated that the responsibility for completeness of an IR pack-age rested with the inspector generating the IR.
TUEC's later practice of entering individual deficiencies into the MDB tended to mitigate the adverse consequences of this weakness.
The TRT observed an inconsistency between the post-construction deficiency report (PCDR) and electric equipment deficiency report (EEDR).
Revision 19 0-304
to QI-QP-11.3-40 included both an attached PCDR form with a blank for a QC reinspection signature and date and an attached EEDR form without such a blank. The PCDR was used to document raceway deficiencies, and the EEDR was used to document electrical equipment deficiencies.
The TRT was unable to determine any reason for this inconsistency.
The TRT found that NCRs generated during a post-construction inspection were not always listed on the IR.
Such NCRs were listed with the parti-cular deficiency on the DR or punchlist, but these were not cross-referenced to the appropriate block (block 15) on the IR. This cross-reference'was a specific procedural requirement of QI-QP-11.3-40, Revision 18; however, not one of the 22 PCVP irs reviewed by the TRT included this cross-reference.
The TRT also found that some DR forms were used that were not described in an approved procedure. Three of the irs examined by the TRT had forms attached entitled " Post-Construction Verification Deficiency Report" (emphasis added) that were not included with any of the 19 revisions to QI-QP-11.3-40.
In addition, the TRT found that inappropriate attributes were listed on an IR and an incorrect DR was used.
IR E-1-0015374 listed attributes applicable to cables and raceways which were marked as unsatisfactory.
The attached DRs were intended for raceway, but described deficiencies associated with a single piece of electrical equipment (an emergency light fixture) rather than those associated with raceway.
The TRT noted that there were unexplained line-outs on DRs; 26 deficiencies on DRs attached to IR PCV-1-0034919 were lined-out and no indication of the reason for the line-out was provided.
The TRT could not determine whether the items had been incorrectly recorded, if entries had been duplicated, or if there was some other plausible reason.
Individually, the weaknesses found by the TRT have little significance; however, collectively, they point to a program that, although subject to site supervisory review and evaluation, suffered from a lack of administra-(
tive precision.
For the reasons discussed earlier, the TRT considered the following weaknesses to be more significant, particularly when coupled l
with the programmatic weakness regarding the lack of guidance for initia-l tion of NCRs.
The TRT reviewed deficiencies listed on irs and found a number of similar l
deficiencies for which an NCR was initiated in some cases, but not in others. Examples of these deficiencies are as follows:
l cables not properly trained into a n.otor control center o
l lack of torquing of limit switch cover fasteners o
lack of color coding on conduit and junction boxes o
o missing identification l
o broken flexible conduit l
0-305
,--n
The TRT also found some deficiencies listed on attahements to irs which appeared to require an engineering evaluation or repair or rework using other than normal construction procedures. However, these were not listed with an NCR number.
Examples of such deficiencies are as follows:
rusted terminal points o
twisted lugs o
o broken terminal blocks a bent transformer top o
corrosion in a cable tray o
5.
Conclusion and Staff Positions: The TRT found much of this allegation to be without basis.
The PCVP was not a " blitz inspection" consisting of an
" instant review and an instant fix" that was stopped prematurely in 1984.
Rather, the PCVP was an ongoing program beginning in 1982 and continuing for Unit 1 in October 1984.
(The PCVP for. Unit 2 had not yet begun.) The TRT could find no evidence that the PCVP failed to document problems which were identified by QC but were corrected by the craft personnel on-the-spot.
The PCVP was not limited in scope to "only a few cable trays"; however, the TRT attributed the alleger's perception of a limited program to the lack of a program plan for the PCVP and due to the limited regulatory basis con-tained in IEEE-336.
Contrary to the allegation, TUEC's PCVP was not "a violation of law."
No particular violation of law was identified by the alleger.
In fact, the PCVP met TUEC's commitments to IEEE-336.
The PCVP included tracking and trending of all identified deficiencies including line items on DRs and punchlists, unsatisfactory irs, NCRs, and CARS.
However, the TRT did find TUEC's trending program to be weak.
The alleger's assertion that the PCVP lacked an engineering review of identified problems was essentially correct, since only NCRs and separa-tion violations received engineering review. The vast majority of defi-ciencies identified during the PCVP were minor and inappropriate for engineering review, for example, chipped paint and loose cabinet fasteners.
The alleger's description of the PCVP being "out of control" and B&R
" turning over a mess" to TUEC was exaggerated.
However, the TRT did find that the PCVP suffered from a number of administrative weaknesses that collectively detracted from a smooth-running program.
Based on its review of the PCVP and related programs, the TRT did not find that B&R was
" turning over a mess" to TUEC. Although the PCVP contained weaknesses, it was impitmented and met the requirements of IEEE-336.
In addition, TUEC's final acceptance by plant operations appears to have been thorough. well-managed, and effectively implemented. The final acceptance program, which followed the PCVP and testing, was specifically designed to prevent TUEC's acceptance of "a mess."
In summary, the TRT concludes that the PCVP meets the requirements of IEEE-336. Most aspects of the allegation were not substantiated and had no generic implications. However, the allegation of inadequate review of 0-306
~
identified deficiencies led the TRT to determine a programmatic weakness involving the lack of guidance on the level of deficiency needed to initiate an NCR.
This finding has generic implications for TUEC's other inspection and corrective action programs for the design and construction of CPSES.
Review of this allegation also led the TRT to conclude that TUEC's program for trending nonconformances was weak.
8 On November 8, 1984, the TRT had an interview with the alleger.
The TRT's assessment of the allegation was discussed with the alleger. At that time, the alleger did not identify any further concerns or allegations.
7.
Attachments:
None.
y/
Reference Documents:
1.
Comanche Peak Steam and Electric Station (CPSES) FSAR, Section 8.0,
" Electric Power."
2.
Texas Utilities Generating Company (TUGCO-TUEC's operating division and wholly owned subsidiary) Quality Assurance Procedure CP-QP-15.0, Revision 4, " Tagging System."
3.
CP-QP-15.7, Revision 2, " Tracking of Audit Reports / Corrective Action Reports."
4.
CP-QP-16.0, Revisions 12 and 14, "Nonconformances."
5.
CP-Q7-17.0, Revision 4, " Corrective Action."
6.
CP-QP-18.0, Revision 20, " Inspection Report."
l 7.
QI-QP-3-29, Revision 16, " Verify Electrical Separation."
l 8.
QI-QP-3-29.1, Revision 16, " Verify Electrical Separation."
)
9.
TUGC0 Quality Assurance Instruction QI-QP 11.3-40, Revisions 0 through 19, " Class IE Electrical Post-Construction Verification."
l 10.
QI-QP-11.8-14, Revision 3, " Instrumentation System Turnover."
11.
TUGC0 Startup Administrative Procedure CP-SAP-3, Revision 12,
" Custody, Transfer of Station Components."
12.
CP-SAP-16, Revision 8, " Test Deficiency and Nonconformance Reporting."
13.
CP-SAP-20, Revision 2, " Guidelines for System Walkdown Inspections."
14.
CPSES Station Administrai.tve Procedure STA-802, Revision 2, " Final Acceptance of Station System:, Structures, and Equipment."
8.
This statement prepared by:
ma b d i
l C/Haughn gate TRT Tech
' ewer Reviewed by:
OMst/CW-G4teW bd H. Livermore, Group Leader Approved by:
V. Noonan, Date Project Director 0-307 k
~
(