IR 05000456/1986041

Insp Repts 50-456/86-41 & 50-457/86-31 on 860729-870326.No Violations Noted.Major Areas Inspected:Activities Re Allegations,Qa Implementing Procedures & Quality Records

ML20215H111
Person / Time
Site:	Braidwood
Issue date:	04/13/1987
From:	Gardner R, Westberg R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20215H107	List: IR 05000456/1986041 ML20215H106
References
50-456-86-41, 50-457-86-31, NUDOCS 8704200075
Download: ML20215H111 (13)

Text

.. _ _ _ _ _ _ _ - _ _ _ _ _ _ _ .

.

.

U.S. NUCLEAR REGULATORY COMMISSION

REGION III

Reports No. 50-456/86041(DRS); 50-457/86031(DRS) ,

Docket Nos. 50-456; 50-457 Licenses No. NPF-59; CPPR-133 Licensee: Commonwealth Edison Company Post Office Box 767 Chicago, IL 60690 Facility Name: Braidwood Station, Units 1 and 2 Inspection At: Braidwood Site, Braidwood, Illinois Inspection Condu ted: July 29 through March 26, 1987 h &

Inspector: Rolf A. Westberg /3 dl

""

5pp/m Approved By: Ronald N. Gardner, Chief /3 07 Plant Systems Section Date Inspection Summary Inspection on July 29 through March 26, 1987 (Reports No. 50-456/86041(DRS);

No. 50-457/86031(DRS))

Areas Inspected: Special safety inspection of activities with regard to review of allegations and resultant review of QA implementing procedures (51061B) and quality records (510658).

Results: No violations or deviations were identifie .

8704200075 870413 gDR ADOOK 05000456 PDR

_ _ . _ _ _ _ _ _ _ - - _ .

r

.

.

DETAILS 1. Persons Contacted Commonwealth Edison Company (CECO)

• E. Fitzpatrick, Station Manager

• H. Shamblin, Assistant Project Manager

• H. Barnes, Regulatory Assurance Supervisor

• D. Kyrouac, Operations QA Superintendent

• L. E. Davis, Assistant Superintendent

• P. C. Zolan, Quality Assurance Construction L. W. Raney, Nuclear Safety M. R. Doughorty, Project Construction M. Takaki, Regulatory Assurance L. M. Kline, Regulatory Assurance U.S. NRC

• T. M. Tongue, Senior Resident Inspector (Ops)

• J. Kropp, Senior Resident Inspector Construction

• T. E. Taylor, Resident Inspector (0ps)

,

Other personnel were contacted as a matter of routine during the inspectio * Indicates those attending the exit meeting on March 26, 1987.

2. Allegation Followup (0 pen) RIII-83-A-0059: A L. K. Comstock Company (LKC) employee contacted the NRC regarding his concerns at Braidwoo On July 22, 1986, the Alleger was interviewed by the NRC. This interview resulted in the identification of 11 technical issues to be reviewed. The purpose of this report is to address these technical issue (1) Relative to the Alleger's concerns, the inspector reviewed the following documents:

(a) LKC Procedures 1 No. 4.2.1, " Document Control," Revision No. 4.11.2, " Nonconforming Items," Revision No. 4.11.2, " Inspection Correction Report," Revision No. 4.3.1, " Safety Related Conduit Installation,"

dated March 20, 198 r

.

.

-

5 No. 4.8.1, " Inspection of Class IE Safety-Related Conduit Installations,'! dated August 11, 198 ! No. 4.3.5, " Safety-Related Cable Pan Installation,"

dated May 27, 198 No. 4.8.5, " Inspection of Class 1E Safety-Related Cable Pan Installation," dated October 2,198 g No. 4.3.12, " Conduit,and Cable Pan Hangers and Auxiliary Steel Installation," dated March 17, 198 No. 4.3.11, " Stud Welding Procedure," dated July 11, 198 (b) Ceco Procedures:

1 Quality Requirement (QR) No. 2.0, " Quality Assurance Program," Revision 1 Quality Procedure (QP) No. 3-1, " Design Control,"

-Revision QP No. 3-2, " Design Change Control," Revision 1 Quality Assurance Department Memorandum No. 9, " Unit Concept Surveillance Inspection," Revision (c) CECO Memorandum, " Subject: Dismantling Permanently Installed Equipment," dated August 28, 198 (d) CECO Surveillance Reports:

1 No. 6376, "Laminations of Electrical Hanger Base Plates," dated September 11, 198 No. 4735, " Document Review of CECO NCR No. 6076; 50.55(e)," dated August 14, 1985.

!

3 No. 4410, "BCAP Concerns Identified in the Review of l the LKC Document Review Program,"' dated May 9, 1985.

L 4 No. 5980, " Document the Resolution and Closure of Deficiencies Noted in LKC Document Review Program,"

dated June 9, 1986.

! (e) Sargent & Lundy (S&L) Specifications:

l

[ No. F/L 2804-01, " Reactor Containment Electrical Penetration Assemblies," Amendment 5, dated October 3, 1983.

(

,

'

[

r

.

-

2 Form 1868A, " Standard Specification for Nitrogen Gas," dated December 8, 197 (f) CECO Audit Schedules:

1 Supplemental Audit Schedule," ANSI Standards," dated May 5, 197 Unit Concept Inspection Schedule for 1983 dated January 28, 198 (2) The Alleger's specific concerns and the inspector's conclusions follow:

(a) Concern No.1 - The containment base is mislocated by three to six feet. The A11eger based this concern on the fact that an inspection report indicated hardware was missing from the location where it was supposed to be installed, but it was located three to six feet awa The inspector's review of S&L Drawing No. M-10, " Basement Floor at El. 364'-0"," Revision J, indicated that there were two center lines in the containment commonly used during construction; center of the containment and center of the reactor. The center line of the reactor is three feet North of the center line of the containment so the polar crane can reach certain areas of the containmen If a Quality Control (QC) inspector used the wrong center line for an inspection, the hardware would appear to be mislocated. The inspector also reviewed the results of a survey made on June 10, 198 This survey, using the two remaining monuments from the original survey, indicated that the apparent error in the chord distance between the containment radius (R) columns was 3/16 of an inch and that the angles between R points had an apparent error of 0.027 degrees. Since the center point of the containment to the i R columns is 68 feet, this gave an accuracy of 1/4250.

l

'

This indicated that the containment base is not mislocated. Since the original survey was made using seven monuments, turning direct 90 degree angles and using electronic measuring equipment and this 1985 survey was made using two monuments while plumbobbing and taping the traverses, the measured error is within the acceptable

,

range.

l I

The inspector concluded that there was no basis in fact ( for this concern.

!

I I

l-

r

.

.

(b) Concern No. 2 - The pressure gauges on the electrical penetrations were not calibrated to maintain accuracy and the service or maintenance on the gauges was not stated or published. The Alleger became concerned about this issue when the original Bunker Ramo penetrations were replaced by Conax penetrations and he was asked to write installation procedure Conax penetrations are configured in the form of a tubular canister slightly shorter than the containment structure nozzle into which it is installed. Ceramic terminal bushing assemblies t e provided at both the outboard and inboard containment ends for each feed through. Where the conductor feed throt]hs exit the terminal bushings, they are secured and sealed with stainless steel compression fittings. The seal for the terminal bushings to the header at both the outboard and inboard ends of the penetration is formed with a Viton o-rin The area between the o-ring seals and the space within the canister act as a pressure chamber which is used to monitor the leakage rate, via a pressure gauge. The penetrations were

'

shipped pressurized with 15 psig of dry nitrogen for b leakage integrity monitoring by the pressure gauge located on the periphery of the header plat The inspector's review of Conax Maintenance and

-

Installation Manual No. IPS-370 and S&L Specification No. F/L 2804-01 indicated that there was no requirement

'

for calibration, service, or maintenance of the pressure gauges. The purpose of the pressure gauge was to monitor seal integrity visually (Go or No Go). A penetration with suspected seal failure would have to be tested for excessive leak rate using more accurate measuring equipmen IPS-370, Paragraph 6.0, defines leakage as a leak rate exceeding 1 x 10-2 cubic centimeters per second. Since this leak rate cannot be determined by using the gauge on the penetration, there is no need to calibrate or service the gaug Further, once the penetrations are permanently installed, they are connected to a nitrogen manifold at 25 + 2 psig and subjected to periodic leak rate testin The inspector concluded that there was no basis for this Concer (c) Concern No. 3 - The surface finish on the penetrations was not the industry standard of a 32 microfinish (32 RMS) and the sealing surfaces were not controlled. The Alleger voiced a concern that in his experience the surface finish was not acceptable for industry standards, especially for military standard ,

.

The inspector's review of this concern indicated that this concern pertained to the uninstalled Unit 2 penetrations only. Further, there is not just one industry standard for surface finish. ANSI B46.1, 1976, defines ten standards ranging from 250 RMS (rough machined parts) to 1RMS (highly polished parts). However, the acceptable roughness of a surface depends upon the use to which the part will be put. The manufacturer (Conax)

specified a 32 RMS finish for the penetration header plates and the penetration flanges that were welded to the containment structure nozzles. Since the aperture seal is formed between the penetration header plate and the flange with two concentric Viton o-rings, the remainder of the machined mating surfaces form a static seal onl For this type of usage, ANSI B46.1, 1976 recommends a 125 RMS finis In general, 32 RMS is three times smoother than 125 RM The inspector's review also indicated that between April 20, 1984 and June 5, 1985, the sealing surfaces of the flanges were inspected by the mechanical contractor, Phillips Getschow Co. (PGCo), and S&L to determine if they still met the 32 RMS surface finish requirement. These inspections resulted in Field Change Orders (FCOs) No. 2 SPARE 6984 through No. 2 SPARE 7030 which repaired any flanges that were corroded or nicked that did not meet the 32 RMS finish. This in place machining was accomplished by Continental Field Machinin The inspector concluded that this concern was partially substantiated in that some flanges subsequently did not meet the 32 RMS finish requirement; however, the actions taken corrected this situation prior to the installation of the penetrations. Therefore, this concern was not safety significan (d) Concern No. 4 - We had a great deal of concern on the lack of traceability for the reels for wire and cabling. There was a 25% rejection rate when we inspected the file The inspector's review of this concern indicated that the Alleger was not opening a new concern relative to traceabilit Rather,.he was referring to documentation problems discovered during the LKC document review progra Historically, the documentation problem started back in 1979 when the original electrical contractor (Ernst) left the sit LKC, who was in a joint venture with Ernst, then took over the job using the Ernst QA program. In 1981 and 1982, audit findings by CECO pointed to problems

._. _ -_ _ -.

.

with the QA records. This caused LKC to undertake a records review program in October of 1982. When the Alleger left the site in late 1983, the records review was still in progres l In 1984, CECO notified the NRC via 10 CFR 50.55(e) that the records were considered a significant construction deficiency. On February 6, 1985, CECO placed an administrative hold on the LKC document review progra This was the result of concerns raised by the Braidwood Construction Assessment Program (BCAP). As part of BCAP, the Review of Significant Corrective Action Programs (RSCAP) identified the following deficiencies:

,

1 No objective evidence of the document review.

. 2 The third level of review by the supervisor was never

,

performe Out of a sample of 120 records reviewed, 60 were unacceptabl In March of 1985, a new records review was started. This review, which was completed in April of 1986, required 212 field inspections to be redone to establish traceabilit The NRC has previously reviewed and approved the results of the BCAP in Inspection Reports No. 50-456/86003(DRP);

No. 50-457/86003(DRP). In addition, the inspector's review of 37 cable pulling record packages in Inspection Reports l No. 50-456/86036(DRS); No. 50-457/86028(DRS) did not '

indicate any problems with cable reel traceabilit The inspector concluded that this concern had been adequately addresse (e) Concern No. 5 - The most significant event we had in I

hardware was that one of our QA auditors reported that we were in violation of the ANSI N45.2 Daughter Standards on the bending of the bolts legally and that according to ANSI are supposed to satisfy the process for the day.

l The bolts the Alleger was referring to were stud welded i bolts. At the time this item was reported, CECO took the position that the ANSI Daughter Standards (N45.2.XX) did not appl i ANSI Standards are standard good work practices recommended for the industry. They are not an NRC l requirement; however, they become a requirement if they are a commitment of the QA program. In this case in 1983, CECO was committed to all the daughter standards and LKC l

t i

7 l l

_

.

was committed to none of them. CECO Project Construction Department (PCD) personnel were correct when they said that the standards did not apply to LKC's contract. This did not mean that the standards did not apply to the work. Criteria I of 10 CFR 50, Appendix B allows Ceco to delegate the work of executing any part of the QA program to their contractors, but they still retain responsibility for i In this case, it meant that if CECO QA or QC provided the assurance that the standards were being met, the contractor would not have to. Examination of CECO QA supplemental audit schedules indicated that CECO had been auditing compliance to the daughter standards since May of 197 Relative to the bending of the stud bolts, it is not a requirement of the ANSI daughter standards; however, it is a requirement of ANSI /AWS D1.1-1983, " Structural Welding Code." Paragraph 7.7.1.1 of the AWS Code requires the destructive bending of the first two studs that are welded to qualify the welding process for that days production welding. It also states that the test welds can be made on a separate piece of material similar to the production member in thickness and propertie The inspector reviewed LKC Procedure No. 4.3.11, " Stud Welding Procedure," dated July 11, 1980. This review indicated that Paragraph 3.2.7 of the procedure required, at the start of each day's welding, each welding unit to be used to weld two studs to a separate test plate in the same general position (flat, vertical, averhead, sloping)

and of similar thickness. It further reluired that after being allowed to cool, the test studs be bent to an angle of approximately 30 degrees by striking the stud with a hamme Review of inspection records in the 1983 time frame indicated that this was a required verification on the inspection checklist The inspector concluded that there was no basis for this I Concern, s

(f) Concern No. 6 - One of my inspectors made an inspection report identifying three areas of delaminations. When interviewed, the Alleger stated that the delaminations were in the main support plates, but did not identify where they were or what they supported.

i The inspector determined that the delamination concern was

,

the result of a CECO Unit Concept Surveillance Inspection that was corducted in June of 1983. This inspection was

'

conducted to visually inspect the field cut end of electrical hanger base 91.tes for laminations. The scope

+

of the inspection includcd 284 randomly selected base l 8

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _

..

.

.. ..

.

plates located in Containment 1, Containment 2, and the Auxiliary Building. Of the 284 plates inspected, laminations were found in 12 conduit support base plates in Containment 1 only. As a result of the generic implications of this inspection, LKC was requested to inspect and evaluate all of the electrical hanger base plates in Containment I that were not previously inspecte In August of 1986, CECO performed a surveillance to close out the original Unit Concept inspection. At that time, LKC was not able to produce inspection reports or other objective evidence to indicate-that the requested inspection had been performe As a result of an agreement between QA, PCD, and Project Field Engineering (PFE), an inspection of electrical hanger base plates in Containment I was conducted by S& ;

This inspection used a random number generator to select samples from the population of supports. Of 65 base plates inspected, no laminations were identifie The inspector reviewed the results of the S&L inspection and concluded that the A11eger had a valid concern in 1983; however, the results of the inspection conducted in 1986 indicated that the concern had no safety significanc (g) Concern No. 7 - The Alleger used the termir. ology

" falsified records" regarding incomplete Installation Completion Reports. When interviewed, the Alleger defined falsification as stating that work was complete when, in fact, it was not. He also stated that he was being judged on his ability to inspect in a timely manner and that wasted visits to areas that were not complete affected his timelines In 1981 and early 1982, when there were approximately ten QC inspectors and 50 welders, there was no official notification of completion of vork. QC inspections were completed based on the QC inspectors awareness of work in his area. In mid 1982 and early 1983, as the number of welders increased and production picked up, Ceco QA demanded greater accountability. By a letter from the corporate QA Manager, completed work was requested to be inspected within 30 days. As a result of this letter, the conduit, cable pan, and Hanger Installation Procedures No. 4.3.1, No. 4.3.5, and No. 4.3.12 respectively, all contained installation report forms by June of 1982. They were a method of notifying QC of work that was ready for inspectio _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _

-. - .-. -

.

The inspector's examined Regulatory Guide 1.88, which endorses ANSI N45.2.9, " Collection, Storage, and Maintenance of Nuclear Power Plant Quality Assurance Records." This standard defines quality records as those records which furnish documentary evidence of the quality of items and of activities affecting qualit Since installation report forms do not meet this definition, they are not quality records and as such they are outside the NRC's pervie t The inspector concluded that no quality records were falsified and that this concern had no safety significanc The Alleger also stated that the Unit Concept Inspection Program was instituted because the LKC Production Manager was reporting areas of completion that were not complet The Unit Concept Inspection Program was established by directive of the CECO Corporate Quality Assurance Manager in January of 1983. It was intended as an independent overinspection of completed work and it was imposed on LaSalle, Byron, and Braidwood. It was conducted in accordance with an approved procedure and the inspections were scheduled a year in advance. All site contractors were subject to these inspections.

The inspector reviewed the unit concept procedure, the yearly schedule, and a sample of completed inspections and concluded that there was no basis for this concer (h) Concern No. 8 - The Alleger wrote a memorandum to LKC construction management regarding uncontrolled removals of previously installed equipment. Attached as an exhibit was a Unit Concept Inspection Report that identified a missing hanger. The A11eger also stated that the generic deficiency, unauthorized removals, was previously documented by LKC auditors and CEC The Inspector's review of this concern indicated that at the time of the allegation the Alleger had a legitimate concern. Hardware was being removed both through the use of a nonproceduralized system called " Avoid Verbal Orders" (AV0) and some hardware was being removed without any authorization. However, as the A11eger stated, this fact was recognized early in construction by CECO PCD and about

the same time as the Alleger by Ceco Q In August 1981 the Site Project Superintendent issued a memorandum to all site contractors regarding dismantling

'

of permanently installed equipment. This memorandum i

informed the contractors that anyone dismantling equipment

- -- - - - - - - . -. - - - _ - _ -. - -. _ - - .

.

without prior authorization would be immediately dismissed and not subject to rehire. This memorandum also outlined the method for removing equipment temporarily to facilitate construction. That method was to contact PCD by speed memo and they would in turn make the arrangements with the original installer to dismantle the conflicting equipmen Interviews with cognizant site personnel indicated that, in 1982, after the Construction Appraisal Team Inspection at Byron (CAT I), the CECO site QA Staff investigated certain aspects of construction at Braidwood to ascertain whether the deficiencies identified at Byron also existed at Braidwood. One of the items investigated was the use of nonproceduralized communications by the electrical contractor at Byre Since the AVO system was used for removal, reinstallation, and rework of previously installed hardware, its use was suspect; however, its use was continue In March of 1984, NRC Inspection Reports No. 50-456/lM006; No. 50-457/84006 identified an unresolved item relative to the removal and reinstallation of completed Category I hardware. This item led to the following actions by the licensee: On March 29, 1984, a new LKC Procedure No. 4.3.24, was issued to facilitate rework of completed Category I hardwar . The use of AV0s was discontinued for installation related activities by a letter to LKC from the Project Construction Superintendent, dated March 30, 198 . On March 30, 1984, LKC NCR No. 1996 was issue It identified a generic problem of field rework via AV0s without issuing installation reports for reinspectio . On March 31, 1984, CECO QA put a mandatory hold point for QA concurrence on all electrical installation l reworks (QA HP/WP-4).

l The above actions constituted a correction action program to assure that all items removed, reinstalled, and reworked by LKC craft personnel via AVDs had all the necessary documentation and QC inspection The inspector concluded that this allegation was substantiated, in that, unauthorized removals of electrical hardware were performed prior to March 29, 1984;

.

however, the Licensee had taken effective corrective actions to resolve that practice. Further, the unresolved item associated in part with this item has been reviewed by DRP and addressed in Inspection Reports No. 456/87007(DRP);

No. 457/87006(DRP). Therefore, this item is considered close (1) Concern No. 9 - The major designer (S&L) used duplicate numbers on two different drawings, both with different configurations. The Alleger gave an in process -

(incomplete) LKC Inspection Correction Report (ICR)

No. 2670 as an example of what he described as a generic problem with the electrical hanger drawing The Inspector determined that when ICR No. 2670 was written on June 18, 1983, it documented duplicate hanger numbers with different configurations on drawing No. 20E-0-3337A, Revision S, and No. 20E-0-3337C03, Revision . Review of this ICR during this inspection indicated that when it was dispositioned on July 5, 1983, it was closed based on the fact that Revision P to drawing No. 20E-0-3337C03 corrected the drafting error that had duplicated the hanger numbe Research of the LKC ICR and NCR files did not indicate any generic problems with duplicate numbers. The Inspector concluded that duplication of hanger numbers was not a generic problem and that this concern had been adequately addressed by the LKC ICR syste (j) . Concern No. 10 - Tracking completion of inspections was difficult due to S&L drawing changes. This concern was documented in a memorandum to the production manager in which he related that the S&L revisions to drawings obscure QC records for items that were inspected to previous revision The Inspector reviewed two out of the three examples of this concern given by the Alleger.

. Example 1 - On drawing No. 20E-0-3301A, Revision J, a l hanger is identified as CC-84. On Revision V, the hanger is identified as CC-119 per FCR No. F-364 The hanger is still CC-119 on Revision W, but is changed back to CC-84(BR) per FCR No. L-140 The Inspector reviewed the referenced drawing revisions and the two FCRs. This review indicated that there was a hanger with two different numbers, but one hanger (CC-119) was at Byron and the other (CC-84) was at Braidwood. When the configuration of

l

.

the CC-84 hanger at Byron was changed, it was given a new number, CC-119, on FCR F-3645. The "F" identifier in the number identifies it as a Byron FC Further, when the hanger was changed back to CC-84(BR) on FCR L-1403, the "L" identifier in the FCR identifies it as a Braidwood FCR. The "8R" in the hanger number also identifies it as a Braidwood only numbe . Example 2 - On Drawing No. 20E-0-3301A, Revision W, a hanger is identified as CC-102. On Revision Y it-is changed to CC-131(BR) per FCR No. L-128 The Inspector determined that this example was the same as Example The FCR changed the hanger number for Braidwood onl The Inspector concluded that while tracking of completed inspections may have been complicated by the changes to the drawings by S&L, the changes were accomplished in accordance with the CECO approved Procedures, No. QP 3-1,

" Design Control," dated May 12, 1983 and No. QP 3-2,

" Design Change Control," dated July 8,1983; therefore, there is no basis for this concer (k) Concern 11 - The Alleger was concerned that the reinspection of the files was completed adequately and fairl The Inspector determined that there was a legitimate basis for this concern. When the Alleger left the site, the records review program was still in progress. However, subsequently BCAP (RSCAP) and a 10 CFR 50.55(e)

notification reviewed the records and found them acceptable. The NRC assessed the corrective action programs in Inspection Reports No. 50-456/86003(DRP);

No. 50-457/86003(DRP) and found them acceptabl The Inspector concluded that this concern had been adequately addresse . Exit Interview The inspector met with licensee and contractor representatives denoted in Paragraph I during and at the conclusion of the inspection on March 26, 1987. The inspector summarized the scope and results of the inspection and discussed the likely content of this inspection repor The licensee acknowledged the information and did not indicate that any of the information disclosed during the inspection could be considared proprietary in nature.

.

! 13 t