IR 05000456/1987015

Safety Insp Repts 50-456/87-15 & 50-457/87-15 on 870415- 0730.Violations Noted.Major Areas Inspected:Resolution of Previous Insp Findings,Deficiency Repts,Part 21 Repts, Allegation & ASME N-5 Code Data Rept Sys

ML20237G666
Person / Time
Site:	Braidwood
Issue date:	08/11/1987
From:	Danielson D, Jeffrey Jacobson, Liu W, Schapker J, James Smith NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20237G594	List: IR 05000456/1987015 ML20237G590 ML20237G599
References
50-456-87-15, 50-457-87-15, NUDOCS 8708140146
Download: ML20237G666 (16)

Text

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

-

.. ... .

, .

..

v.

U.S.' NUCLEAR REGULATORY COMMISSION

- REGION III f

'.Repurts No. 50-456/87015(DRS); 50-457/87015(DRS)

~ Docket Nos. 50-456; 50-457 Licenses.No..NPF-72; CPPR-133 Licensee: Commonwealth Edison Company Post Office Box 767

' Chicago, IL' 60690. ,

I Facility Name: Braidwood Station, Units 1 and 2 Inspection At: Braidwood. Site, Braidwood,. Illinois Inspection Conducted: May 15, 16, 22, 23, 29, 30, May 6, 7, 19, 20, June 3, 4, 11,.12, 24,. July 14, 15, 28, 29, and 30, 1987 AT1F '

Inspectors: . M. Jacobson "h7 Date

-

.' F. Smith Bh/ /s7 y Date- '

d J F. Sch pker. 8 N//7 Ddte W L /0 7 Date U

Approved By:

~

D. H. Danielson, Chief "h7 Materials and Processes Section Date

!

Inspection Summary Inspection on April 15, 16, 22, 23, 29, 30, May 6, 7, 19, 20, June 3, 4, 11, f[' 12, 24, July 14,15, 28, 29, and 30,1987 (Reports No. 50-456/87015(DRS);

50-457/87015(DRS))

Areas Inspected: Unannounced safety inspection of the resolution of previous inspection findings (92701 and 92702); of 50.55(e) deficiency reports (99020);

of Part 21 reports (92716); of an allegation (99014); of-the ASME N-5 Code Data Report System (49055 and 49065); condensate discharge piping failure (93702);

and of an RCS pressure transient (93702).

'Results: Of the seven areas inspected, one violation was identified for failure to take adequate corrective action - Paragraph 5. ,

a

ff$

-G.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ -

r

'

.

.

.

DETAILS

!

1. Persons Contacted  !

!

Commonwealth Edison Company (CECO)

• T. J. Maiman, Vice President

• E. E. Fitzpatrick, Station Manager M. J. Wallace, Manager of Projects

• W. E. Vahle, Project Manager T. E. Quaka, Site QA Superintendent

• C. W. Schroeder, Station Services Superintendent

• K. Kofron, Production Superintendent D. L. Shan4blin, Assistant Project Manager

• P. L. Barnes, Regulatory Assurance Supervisor

• M. Takaki, Quality Control Supervisor

• C. Bedford, Regulatory Assurance Operating Experience Group J. K. Jasnosz, Technical Staff AR/PR. Coordinator E. R. Wendorf, PFE Mechanical Supervisor L. M. Kline, Regulatory Assurance F. Lotarski, Project Construction

• B. Tanouye, Project Construction The inspectors also contacted and interviewed other licensee and 4 contractor employees. i i

• Denotes those attending the exit meeting at the Braidwood Station on July 30, 1987.

2. Licensee Action on Previous Inspection Findings a. (Closed) Violation (456/84044-08): NCR 39, which was issued in April 1979, identified weld deficiencies in electri:al struts and hanger assemblies supplied by Systems Control (SC). The supporting documentation attached to the NCR indicated that 90% of the welds identified on the NCR were unacceptable in accordance with AWS D1.1-75 (Structural Welding Code). The welds were considered unacceptable due to improper dimensions and weld quality. The " Corrective Action" block on the NCR was marked "N/A" and contained a statement indicating that the welds were acceptable. However, there was nothing in the NCR package to support this determination. Consequently, a Notice of Violation (456/84044-08; 457/84040-08) was issued.

The NRC inspector reviewed the following documented chronology of events:

(1) Material Receiving Report No. 4408 placed the shipment ,

'

on hold due to questionable weld quality (April 23, 1979).

(2) NCR 39 issued to document welding deficiencies (April 25, 1979).

'

____________________A

_

.,

'

. .

'

.

.

(3) Pittsburgh Testing Laboratory (PTL) inspection of SC material reports high weld reject rate-(April 23, 1979).

.(4) Reinspection of welds by PTL, S&L, CECO, and SC (May 2, 1979).

PTL reports same high reject rate complicated by weld spatter.

. 1

~(5). Ceco. Construction indicates in a May 8, 1979, memorandum to l LKC that the May 2, 1979, reinspection reported that welds !

are within tolerances and directs LKC to remove spatter before installing and to report to Construction any deficient welds.

Though NCR 39 Corrective _ Action block at this point stated

' N/A, Disposition block reflects the May 8,1979, . memo regarding removal of spatter and inspection of welds before installing hangers.

(6) LKC memo to Construction, dated May 25, 1979, reports that receipt inspection of Shipment 131 and 133 found similar welding deficiencies.

(7) CECO performs QA Surveillance 1195 on June 4 and 5, 1979, of, SC shop and issues stop work order until changes in production control and weld inspection are.made. Followup inspection on June 19 and 20, 1979, authorizes; start of work with a PTL

' inspector to remain at SC shop for ten days to observe weld inspection practice. .Also, on June 19, 1979, hangers from.

several SC shipments are returned to SC for repair of welds.

(8) CECO,.NCR 451 is . issued December 21, 1982, to document that a large percentage of SC supplied. hangers did not conform to AWS D1.1-75 with regard to weld quality. The NCR was dispositioned by performing a random sample inspection of-80 hangers. S&L performed a design evaluation based on the results of the weld inspections-and reported that all were l . adequate. S&L stated that this sample provided a 95/95 confidence / reliability that all SC fabricated hangers were acceptable (February 17, 1984).

(9) S&L letter to Ceco, dated May 9, 198A, states that the return

.

welds at the top and connection of horizontal strut members l were missing in many cases. S&L recommends that a weld l configuration reinspection be performed to verify actual weld size and length.

(10) CECO issued NCRs 708 and 709 to document reinspection of SC hangers for weld configuration.

The NRC inspector reviewed CECO NCR 708 (Unit 1) and related documentation. This NCR included a 100% inspection of all cable tray hanger connections involving in excess of 30,000 connections and a 100% inspection of all 90 fittings totaling 932 welds. ,

In addition, a sample of ladder rack rung to side channel welds, !

T-fittings, and instrument racks were inspected. No design

s  ;

{* .

.

l- ,

.

,

.

significant discrepancies'were found during the ladder rack, 90 fitting and instrument rack inspections. Any cable tray hanger connection missing a portion of a weld was evaluated and repaired as required to restore design margin. The weld quality of cable tray hangers had previously been resolved by CECO NCR L-451.

In addition to the reviews outlined above, the NRC inspector visually inspected in excess of 150 cable tray hanger connections and found them acceptable.

b. (0 pen) Open Item (456/87004-01): Diesel generator rocker arm failures (reference 10 CFR 50.55(e) 457/87001-EE, Paragraph 3.c.

below). In addition, both the 1A and 1B diesels have successfully completed a 100 bcur proving run to demonstrate rocker arm reliability.

c. (Closed) Unresolved Item (456/85007-05; 457/85007-05): Conformance with installation requirements. The licensee committed to installation of electrical penetrations in accordance with IEEE-317-1972, which required the use of ASME Section III. However, the licensee took exception to the application of ASME Section III to certain portions of the installation.

The bolted connection between the header plate and the weld neck flange of the electrical penetration was the subject of a letter (Westermeier to Cleff), dated April 24, 1981, and ECN No. 36330, dated March 26, 1987. Based on the contents of these documents,.

the installation and inspection of the bolted portion of the penetration need not be performed by contractors qualified in accordance with the ASME Boiler and Pressure Vessel Code. Conax installation instructions for the penetrations were added to the requirements by the ECN.

The field weld joining the weld neck flange to the penetration nozzle was made by Phillips Getschow in accordance with ASME Boiler and Pressure Vessel Code,Section III. The NRC inspector confirmed this by review of Drawings S-1032, Revision AG, dated June 20, 1986, and S-936, Revision 2, dated June 16, 1986. The mechanical assembly of the penetration was performed by L. K. Comstock in accordance with the manufacturer's instructions as authorized by the ECN and the welding was performed by a qualified contractor (Phillips Getschow)

in accordance with the requirements of ASME Boiler and Pressure Vessel Code,Section III.

3. Licensee Action on 10 CFR 50.55(e) Items a. (Closed) 50.55(e) Item (456/84005-EE; 457/84005-EE): Inconsistencies in G. K. Newberg's (GKN) structural steel installation documentation.

These inconsistencies included changed heat numbers on Structural Steel Installation Travelers (SSIT), changed weld lengths recorded on weld inspection requests, and the use of alternate fabrication

--

'

. .

-

.

'

.

i details without proper documentation. The changes were lined out, initialed, and dated; however, no explanations for the changes were noted.

Corrective action included a review of all SSIT packages for i

correctness. The generic resolution for many packages consisted of reviewing supporting documentation to verify that the changes were correct and/or reinspection of the installation. For those items which could not be resolved in this manner, specific or generic NCRs were generated. Approximately 7800 SSIT packages were reviewed and all documentation deficiencies were resolved. The NRC inspector reviewed a sample of SSIT packages, GKN Procedures 29, 29-1, and 29-1-1 and found the corrective action to be acceptable.

The review of SSIT packages performed as corrective action for this item was also utilized to close 456/84015-EE; 457/84-15-EE discussed in Paragraph 3.c. below, b. (Closed) 50.55(e) Item (456/84011-EE; 457/84011-EE): Failure of steam generator hydraulic snubbers. NRC Inspection Reports No. 50-454/83020; No. 50-455/83017 and No. 50-454/84008; No. 50-455/84006 discuss the Region III concerns relative to these snubbers including the adequacy of the qualification testing performed by the Boeing Company. To resolve these concerns, the licensee contracted ITT Grinnell to perform additional testing.

During initial testing conducted by ITT Grinnell, some snubbers failed to meet the bleed rate criteria in tension due to excessive leakage past the high pressure rod. seals. Failure of the snubbers to meet test specification requirements indicates that the snubbers tested would not resist design loads during a Design Basis Event, thus exceeding the ASME allowable stress limits of the Reactor Coolant System. To remedy this situation, ITT Grinnell was directed by CECO to design an alternate front and rear seal assembly for the Boeing snubbers. The qualification test procedure used for the redesigned snubbers has been reviewed by Region III Staff. All redesigned snubbers have been tested and qualified per S&L requirements. Each snubber was tested for spring rate, lock-up velocity, bleed rate, stroke, and internal resistance. The NRC inspector reviewed the <

qualification test data for Snubber No. 41-0035-25 and found it l acceptable.

c. (Closed) 50.55(e) Item (456/84015-EE; 457/84015-EE): Deficiencies in the Gust K. Newberg (GKN) (structural contractor) welding program.

Documentation deficiencies were identified in certain Structural Steel Installation Traveler (SSIT). packages involving Flux Core Arc Welding (FCAW) and Shielded Metal Arc Welding (SMAW). The following deficiencies were identified:

(1) GKN Engineering specified Welding Procedure Specifications (WPS) which were not approved. l

b

-.

m ,. ,

.

.

'

.

'

(2) Engineering specified WPS's which did not apply to the particular weld-joint detail.

(3) Engineering did not specify WPS's for all weld joints.

(4) GKN production utilized unapproved WPS's.

(5) Production utilized improper WPS's.

(6) Production listed WPS's on the SSIT's which, in fact, were not utilized.

(7) Production failed to list all WPS's used on some installations.

(8) Production utilized WPS's not specified by. Engineering.

To correct the deficiencies concerning. unapproved WPS's, they were incorporated in GKNs Welding Procedures, Sections 26 and 43, and approved for use. The specification or use of improper welding procedures did not result in any hardware deficiencies. All procedures utilized were AWS prequalified and the procedures in question were not significantly different in technical content from those which should have been utilized. In virtually all instances, the WPS's utilized resulted in technically acceptable welds. The remaining discrepancies were related to documentation and were corrected.during the SSIT review described in Paragraph 3.a above.

The NRC inspector reviewed.the resolution of a sample of each of the discrepancies-described above and found them to be acceptable, d. (Closed) 50.55(e) (45ti/84019-EE): Insufficient crush strength of energy absorbing material (EAM) used in pipe whip restraints.

During testing of the Byron Station EAM, it was found that some material exhibited an average dynamic crush strength less than that specified. The tests were mostly conducted on specimens from new material obtained from the supplier Hexcel, not specimens from the EAM core blocks originally supplied to Byron Station. Further investigation indicated that the variation in strength may have been due to problems associated with an earlier brazing process utilized by Hexcel. It was further determined that the EAM supplied to the Byron, Braidwood, and LaSalle projects could be affected by the j questionable fabrication process.

The technical evaluation of the pipe whip restraint design and EAM l is addressed in NRC Inspection Report Nos. 50-454/84051; 50-455/84035. '

To resolve the EAM crush strength variability issue it was decided to replace all EAM with new material properly qualified by improved production tests. The NRC inspector reviewed the installation documentation and visually field verified the installation of the new material for the following pipe whip restraints:

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -

p

'

,.. .

'

.

1WR-FWR-1 1WR-FWR-9 IWR-FWR-10 IWR-RC3-4'

IWR-RY-4-1 All whip restraints examined were found to have the proper EAM installed.

e. (0 pen) 50.55(e) (457/87001-EE): Diesel generator rocker arm failures.

The failure of rocker arms on the 2A and 28 diesels is discussed in NRC Reports No. 50-456/87004; No. 50-457/87003. The licensee and Cooper-Bessemer (C-B), the diesel manufacturer have performed metallurgical evaluations to determine the cause of failure.

C-B has determined that the 2A diesel failure on February 4, 1987, was caused by seized crossheads. As corrective action, the crosshead to guide clearance has been increased and a 100-hour test run has been completed. C-8 advised that the 2A rocker arm failure on l

February 13, 1987, and the 2B rocker arm failure on March 29, '

1987, remain unexplained at this time.

CECO has conducted an evaluation of the February 13, 1987, and March.23, 1987, failures and has focused on the tensile strength of the rocker material. The tensile strength of the failed exhaust rockers were found to be less than 18,000 psi, while the tensile strength required by specification is 40,000 psi. It has been postulated that failure of.the exhaust rocker causes increased loading of the intake rocker resulting in subsequent failure.

CECO has devised a methodology to relate Brinell hardness of the rocker to tensile strength using portable hardness test equipment.

At this time, Ceco has concluded that exhaust rocker arms exhibiting a hardness relating to less than a 20,000 psi tensile strength should be replaced. Further, in an effort to improve confidence, CECO has committed to replace rockers exhibiting less than a 30,000 psi tensile strength, upon parts availability.

To date, all exhaust rockers exhibiting less than 20,000 psi tensile strength have been replaced at the Byron Station. A similar effort will be performed at the Braidwood Station upon receipt of replacement parts.

4. Licensee Action on 10 CFR Part 21 Item (Closed) Part 21 Item (456/84007-PP): Insufficient crush strength of energy absorbing material (EAM) used in pipe whip restraints.

This item is addressed in Paragraph 3.d of this inspection report as 10 CFR 50.55(e) (456/84019-EE).

5. Allegation Followup I

(Closed) Allegation RIII-87-A-0033: This report documents the receipt of and followup on allegations made by "A Concerned Worker" at the l Braidwood Station. These allegations are summarized as follows:

'

I -- -- - - - - - - - - - - - - - - - - - _

- _ _ _ _ _ - - _ _ - _ - _ - _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ - _ - _ _ -

-

.s

, .

.

V. ,

5("

.

a '. Phillips-Getschow (PGCo) Welder No. 204 welded pipe support hangers

'

in a downward progression i,n violation of the approved. welding procedure.

b. The PGCo QC inspectors are not qualified. They lack the experience and training to be qualified'and they accepted the improper welding

<

described in (a) above.

^

c. PGCo management was aware of the improperly. welded hangers and did not correct it nor report it to Commonwealth Edison Company.

d. - Unauthorized repairs were made.to some hanger supports.

NRC Review a. Downhill Welding Progression The NRC inspector's review of this allegation disclosed that Welder No. 204 had welded pipe support hanger No. 1AB18005G in a downward progression. This condition had been reported by the QC inspectors, who subsequently issued Nonconformance Report 3379, dated January 8,

1985. As a result of this deviation from the approved procedure, the welder was terminated on January 7, 1985. Corrective action consisted of removing and scrapping the steel angle on which the weld was made and replacing it with a new piece which was properly welded. As a result of the current allegation, all safety-related hangers welded by Welder No. 204 were identified and.a sample of this population was visually examined for welds with downward progression. . When evidence of additional downhill welding was found in the initial sample of the welder's work, the licensee expanded- the sample size to 100L The licensee'also initiated a search of records and interviews of inspection personnel to determine the names of other welders who had been found to weld in the downhill progression.

~ Visual examination of the first welders work on 62 safety-related hangers disclosed the presence of downhill welding on seven hangers.

Another six hangers contained welds which would have been made in the vertical downward progression if they were made in place, but which could have been made in another position, if the assembly was made as a bench weld. Because the welder in question had demonstrated a preference for welding in the downward progression, the NRC inspector took the position that this welder might also prefer to weld downhill on a bench weld. In accordance with this conservative assumption these six welds were also considered to be downhill welds.

J Downhill welding by a second welder was the subject of NCR No. 3757, dated February 2, 1985. The corrective action for this NCR was to remove the downhill welds on the subject hanger, reweld property and retrain the welder. As a result of the current allegation all

l J

!

!

!

t_.__.

_

.- ;.

'

.

..

.32 safety-related' hangers which he welded were reinspected, where !

possible, for downhill welding progression. Visual examination !

disclosed the presence of downhill welding on three hangers with'

an additional three hangers assumed to be downhill welded due to

' inaccessibility.

A third welder was rumored to have welded downhill but none of .

'

!the six sleeves on which he welded could have been welded downhill.

A fourth welder who was alleged to have performed downhill welds worked on 65 hangers. The licensee. interviewed this welder, v.isually examined a random sample of ten of his hangers and found no evidence uf downhill welding. Based on these results it was concluded that the welder had not welded downhill.on this job.

In addition, the Braidwood NRC Senior Resident Inspector for construction interviewed the welder on this issue. The welder indicated that at no time did he weld downhill nor did he have knowledge of any other downhill welding.

The total population of hangers on this job with downhill and possible downhill welds was determined to be 19. For each of these hangers,.the licensee performed analytical calculations which demonstrated either that the downhill welds were not essential to the integrity of the hanger or that the contribution of the hanger could be eliminated without overloading adjacent hangers. An NRC Region III inspector reviewed a sa- le of ten of these calculations for methodology and correctness. .4 confirmed that the calculations were appropriate and correct. These calculations show that the systems supported by hangers containing downhill welds will be unaffected even if all of the downhill welds are considored to offer no contribution to the strength of the hangers.

In an effort to determine the quality level of the downhill welds, the licensee destructively evaluated a hanger which contained several downhill welds. Three welds on the hanger were evaluated by visual examination, magnetic particle examinations, fillet weld break tests, and longitudinal and transverse macro sections. The conclusions of the evaluating laboratory were as follows:

(1) Downhill welding progression was confirmed.

(2) All welds exhibited good penetration, absence of slag inclusions and absence of lack of fusion.

(3) All weld sections showed a high level of quality and soundness.

The NRC inspector reviewed the results of this evaluation, summarized in Taussig Report No. 74975 dated June 23, 1987, and concurred with the test methods and conclusions. These welds appear to be capable of contributing substantial strength to the hangers.

l -.

1

< -___-

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

.. .

-

.

l

.*

l l

l b. Inspector Qualification The requirements for education, experience, physical requirements, and training for PGCo inspectors are outlined in ANSI N45.2.6 and NRC Regulatory Guide 1.58. The PGCo requirements also follow the requirements of Commonwealth Edison's PM-12. Procedure PM-12 requires cont ~" ors who perform inspection activities to verify the education and experience of inspectors via letter or telephone before certification. Randomly selected sample records of inspector qualifications were reviewed in detail by the NRC inspector and found to comply with the training and testing requirements for Level II personnel outlined in Table 2 of QAP/BR-QCT-20.15, Revision 17 (which complies with CECO PM-12). The training and testing records of each inspector on the job is approved by Commonwealth Edison before he is permitted to perform independent inspection.

In addition to the current review of PGCo inspector qualifications, the results of an earlier review were also examined. The " Quality Control Inspector Reinspection Program" (QCIRP), initiated by CECO in the Fall of 1984, included 2968 inspection reports covering reinspection of 6436 elements. The overall contractor performance reported in that program met the 95% conformance threshold established for QCIRP.

To preclude the possibility of any PGCo welding inspector inadvertently accepting a vertical downhill weld in the future, the licensee completed training in this area for all PGCo welding inspectors by the end of July 1987.

c. PGCo Management's Role in Correction and Reporting of Improperly Welded Hangers, As previously indicated, the improper welding progression on hangers had been reported by PGCo QC Inspection on Nonconformance Reports.

Thus, PGCo Management was aware of the conditions, reported it via the issuance of NCR's, and recommended corrective action. The NRC inspector interviewed several members of PGCo Management and found that NCR's had been issued for the known instances of downhill welding, and corrective action had been taken.

d. Unauthorized Repairs on Hanger Supports The administrative rules governing welding at Braidwood allow welders to correct their welding deficiencies without docunientation prior to final QC inspection. Because of this arrange.ent, it is not possible to identify an ' unauthorized" repair.

In lieu of a search for " unauthorized" welds, the NRC inspector reviewed welds on approximately 30 hangers chosen at random for defective local repairs. No evidence of local repairs was found on any of them. If repairs were performed on these velds, they are of

.

- _ _ _ _ _- _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ __ _ _ _ _ - _ - _ _ - _ _ - _ _ _ _ _ _

.. .

.

1.-

the type applied to build up the weld cross section over the full length of the weld and were of a quality level comparable with that of the original weld. Based on the results of this observation there appears to be no reason to expect the presence of defective repair welds on the remainder of the hangers.

Conclusion a. The allegation concerning the downhill welds was substantiated in that vertical downhill welds had been performed and accepted. The licensee perfonned an investigation which disclosed that 19 hangers might contain downhill welds. Analytical calculations have been performed to confirm that the remaining welds and adjacent hangers can absorb the loads which would result if all of the questionable welds were removed. This is a technically acceptable resolution of this issue.

b. The allegation concerning the lack of proper qualification of QC inspectors was not substantiated. The mechanical contractors QC welding inspectors were found to be properly qualified. In addition, the inspectors were to be provided further training in inspecting for downhill welding to preclude future acceptance of such welds on their jobs.

c. The allegation concerning PGCo management was not substantiated.

The NRC inspector interviewed PGCo QC management and found no evidence that they were aware of any downhill welding on hangers other than those reported on the three NRC's which were processed by them.

d. The allegation regarding unauthorized repairs could not be substantiated. No unacceptab's repairs were found while investigating hangers for downnill welds. No records are kept 1 of in-process weld repairs, so an " unauthorized" repair cannot be identified.

The corrective actions to the original NCRs disclosing downhill welding ,

are not considered to be responsive. The corrective action for the NCRs should have anticipated more extensive incidences of downhill welding and prescribed inspection of additional hangers. However, the appearance of NCR No. 4259, disclosing downhill welding by welder No. 204 over two months after his discharge obviously confirmed the presence of one additional l incident of downhill welding and should have eliminated any doubt ac to I whether or not inspection for more downhill welding in the hangers was appropriate.

Instruction for proper control of this situation is provided in 10 CFR 50, Appendix B, Criterion XVI, Corrective Action. This states, in part,

" Measures shall be established to assure that conditions adverse to quality, such as . . . Nonconformances are promptly identified and corrected."

Failure c,f the licensoe to promptly identify the full extent of the

_ _ _ _ _ -

r

.. .

-

.

.-

adverse condition deferred ccrrection until the problem was highlighted by investigation for this allegation. Failure to perform adequate and prompt corrective action of a nonconforming condition is an example of a violation of 10 CFR 50, Appendix B, Criterion XVI (456/87015-01(DRS)).

6. Condensate Discharge Piping Failure On April 16, 1987, The Unit 1 "D" condensate pump discharge line experienced a 1-2 gpm leak. Visual observations identified at least two cracks of approximately 7" and 3" in length respectively. The cracks were located in the 18" diameter pump discharge line at its junction with the 30" diameter header. The failures appeared to be located in the heat affected zone of the weld. The cracks were subsequently grou1d out and weld repaired. Based on the location and orientation of the cracks, the licensee determined that the failures were caused by fatigue enduced by mechanical vibration. A similar failure occurred earlier in the year at the intersection of the 30" diameter header and a 2" diameter drain line.

S&L has completed an evaluation of the piping and support arrangement in the vicinity of the failed weld. Pipe supports were field verified to be installed in accordance with the latest design drawings. The piping analysis indicates that the stress levels during normal oparating loads are well below the ANSI B31.1 Code allowables.

S&L has determined that the Braidwood pipe and support configurations are the same as those at the Byron Station. No similar failures have been reported at Byron.

Vibration measurements were taken during operation and calculations were performed which demonstrate that stresses resulting from the measured vibration were reasonable. Additionally, vibration measurements were taken both with and without a temporary support installed on the header.

It was found that vibration levels were not significantly reduced by the addition of this support.

The licensee has elected to install strain gage instrumentation on this line in an attempt to quantify the vibration during various levels of plant operation. The review of this vibration data upon completion is considered an Open Item (456/87015-02(DRS)).

No violations or deviations were identified.

7. RCS Pressure Transient On February 10, 1987, Unit 1 was in Mode 5, Cold Shutdown with 0% reactor power at 105 F and 350 psig. During trouble shooting of a noise problem on Nuclear Instrumentation Channel N32, the output breaker for Power Supply No. 1 in Cabinet 1PA06J was reclosed and caused an unexpected trip of Power Supply No. 2 which was supplying power to Cabinet 1PA06J.

A damaged dicde was found on Card ITB-453 which allowed Power Supply No. 2 to see the effects of the No. 1 output breaker reclosure. This l

i .. .. .

. . .

.

_ . _ _ _ _ - _ __ - _ - . .

.

. .. .. .

.

... .

,, ,

.

..

unexpected trip de-energized Cabinet IPA 06J for.approximately seven l-

'

seconds. The loss of IPA 06J caused the RCS charging flow control valve to fail.open. This caused a rapid rise in RCS pressure. This rise in pressure caused PORV 1RYO455A to lift and reseat four times within a 39-second time span.

The NRC-Inspector reviewed the RNDT limits for the reactor vessel and found that at the temperature of the event (105 F), the maximum ' allowable pressure is approximately 550 psi. Since the pressure tr'ansient occurred in the 450-500 psi range, adequate margin existed.

No violations or deviations were identified.

l 8. Review of Phillips Getschow (PGCo) N-5 Data Report Program The National Board of Boiler and Pressure Vessel Inspectors (NB) performed an audit of the PGCo N-5 Data Report program which resulted in the identification of 11 findings. The NRC inspector reviewed the findings described in the NB Final Report, dated June 19, 1987, (Attachment A) and the licensee's resolution of those findings. Following is the NRC inspection 'results of those items as they appear in the NB Final Report:

1. For Braidwood, CECO has elected to issue separate N-5 Reports for piping and piping supports due to construction schedules. For piping, Phillips-Getschow Company (PGCo) establishes the boundaries to be included on the Data Report. Piping Support Data Reports are prepared on an analytical subsystem basis. The completed piping Data Reports are provided to Sargent and Lundy (S&L) for the purpose-of establishing the piping support analytical subsystems within the piping Data Reports. S&L and Commonwealth. Edison Company (CECO) complete the subsystem reconciliation and CECO signs the N-5 transmittal form indicating the design is complete.

Since there are no apparent ASME requirements for establishing the subsystem or N-5 boundaries, the CECO system appears to be in compliance.

2. The IRE-1 N-5 Report has been amended to include the subassemblies in question. All-supporting records were included in the package at the time of the audit; however, the subassemblies were inadvertently omitted.

3. The PGCo Procedure, QAP-BR-N5, requires that Code Cases be listed on the N-5 as required by the wording in the specific Code Case.

Code Case N-272 does not contain such wording. The S&L Design Specification L-2739 allows the use of ASME NCA-8430, 1980 Edition, S-81 Addenda, which incorporates the provisions of Code Case N-272.

In addition, PGCo included an attachment to the N-5 packages listing all Code Cases which may apply to the installation.

__--- -

_ _ _ _ _ _ _ _ _ _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ - _ _ _

.. .

1 .

.

1 4. PGCo states that the missing documentation was for Welds 3D, 3E, and 3F, not E, 3F, and 3G. The missing documents could not be located and were therefore restored per PGCo P ocedure QCP-B30. As copies of the construction documentation exist in other files, recreating was not a problem. Adequate documentation exists to support the PGCo position that the missing documents were present and reviewed by the AI at the time of signing the N-5. A " Package Update Status Report" dated January 16, 1986, lists the welds in question along with ANI review signature. Additionally, an indexing of the package performed on August 14, 1986, listed 273 pages, a recount after the NB audit totaled 255 pages (18 pages missing). The reconstructed documents total 18 pages.

5-7. During that period of time when the FFP&D in question was issued, PGCo utilized a pre-typed format. Standard practice was to use the closest format for a particular situation, line out and make changes as required. Some changes on this particular FFP&D were for technical reasons (e.g., change of material, change of welding procedure revision to allow open butt welding of socket to pipe welds),

while other changes were made to reflect the procedure revision in effect at the time of the work.

During the 1981-1982 time frame, the making of changes to documents became an issue at CECO's LaSalle Station, especially the use

'

" white-out." CECO issued a directive to all construction sites at that time requiring the initialing and dating of all changes. In late 1982, PGCo developed a new FFP&D which did not utilize the pre printed format.

In summary, it appears that PGCo document revision control evolved in concert with CECO QA requirements. The staff is unaware of any specific ASME requirements regarding document revision control applicable to the Braidwood Project.

8-9. Per the PGCo N-5 procedure, the " Additional Material" section is ieserved for listing Code material used in the installation but not included as a spool (piping subassembly) on the PGCo as-constructed drawing listed on the N-5. Since all supporting documentation litts the 3/4" drain line ICS55AA-3/4" as being

"fabbed as part of" line 1C502AA-10, this 3/4" line would not be considered " additional material" by PGCo.

The SWF Spool CS-5-7 is listed on the N-5 and supporting documentation  !

shows this spool was edified by PGCo, including the addition of the low point Drain Line 10:55AA-3/4". The fact that the Architect Engineer gave this line another number (due to the size difference) does not change the fact that it is part of Spooi CS-5-7. PGCo has demonstrated that when a spool accounts for two or more A-E line numbers, only one line number is listed on the N-5 along with the spool number. The supporting documentation shows the other line numbers to be "fabbed i as part of" the line listed on the N-5. An identical situation was l demonstrated for spool CS-6-7 listed on this N-5.

_ _ _ _ _ _ _ _ _ _ _ _ _ \

- - _- _

- . - - - - - - - _ - - - - _ _ _ _ _ - _ - - - - _ - _ _ _ _ _ _ _ _ ,

-

p,., ,

.

.

,

While on.the surface, the listing of:some line numbers and not others appears inconsistent, this practice is consistent with the PGCo program and does not result in loss of; traceability.

-10. Penetration 1 PC-16A is taken credit for on Drawing'1C-CS-6 which

'is listed on the N-5. This drawing contains many spools which are-shown-and listed. For. specific information regarding a spool,'you must refer to the individual spool drawings referenced on IC-CS-6 (sub tier drawings). . Similarly, Penetration 1 PC-16A is shown and listed on'1C-CS-6, and for specific information you must ' refer, to

' Drawing 1 PC-16A (sub. tier drawing). The Drawing-1 PC-16A is filed

.

al_ong with the 1C-CS-6 package and is readily available. Additionally,

-Penetration 1 PC-16A is listed on the N-5 in question (1 CS-1). This-practice is considered acceptable.

11. FC0 Nos. 2018, 4006 and 10S-23 were inadvertently removed from the package. It was.the intent of PGCo to remove superseded FC0's from

.the package if no work had been performed on those FC0's. While clearly the superseded FCO's in question should not have been removed from the package.as work had been' performed on them, the NRC staff considers the removal of a5 FC0's to be questionable. The. staff maintains that an FCO, if referenced on another FC0 as " incorporates

.FC0 xxx" or " supersedes FC0 xxx," becomes a quality document regardless of whether that FC0 affected the actual hardware and-should be. maintained'as permanent record.

CECO QA has agreed with the staff position, end PGCo will return all FCO's te the N-5 packages.

To supplement Lie NRC~ review of the NB audit items the NRC inspectors independently reviewed nine completed N-5 packages selected at random.

The following packages were reviewed:

1WE-1 Aux Building Equipment Drain Radwaste Reprocessing 2WE-1 Aux Building Equipment Drain Radwaste Reprocessing 2WE-2 Aux Building Equipment Drain Radwaste Reprocessing 1MS-1 Main Steam - Loop 4 1AF-1 Aux Feedwater 1WM-1 Makeup Demineralized ISI-1 Safety Injection IMS-1 Main Steam - Loop 1 1RC-1 Reactor Coolant - Loop 1 Of toe nine packages reviewed, all were found to be in compliance with ASME Code requirements. Additionally, PGCo has implemented an enhanced N-5 review program for Unit 2, which is expected to improve the quality of the N-5 documentation.

)

!

)

c-__-- - _ _ - _ _ . i

e

..' . ,

'

.

.

9. Open Items Open items are matters which have been discussed with the licensee, which will be reviewed by the inspector and which involve some action on the part of the NRC or licensee or both. An open item disclosed during the inspection is discussed in Paragraph 6.

10. Exit Interview The Region III inspector met with the licensee representatives (denoted in Paragraph 1) at the conclusion of the inspection on July 30, 1987.

The inspector summarized the purpose and findings of the inspection.

The licensee representatives acknowledged this information. The inspector also discussed the likely informational content of the inspection report with regards to documents or processes reviewed during the inspection.

The licensee representatives did not identify any such documents / processes as proprietary.

d

l

. .. .. . .. .. .. .

._ . .

.

.

e

... . ... .

-

p r n a - o.r a

'

Elje National Bourb of Boller anb Urcssure liessel Enspectors D J MCDONALD. Esecutwe Director BOARD OF TRUSTEES 40% cavPPER Avf Nut J G cvAR cotuusJ5. o*co 41279 U 5 A stet.ofDra eo u L s.Now. JR v.oo.....

.....m. , chams, ene. m enen: a w oiAuono s u umutws ,,, u.cn.ne

'"'"*"" "iM*,:,'," ,"'*"*"'

A .im.t.... ,w f , , ,,,,,

• Dc,^',",5",,*"*/;,c""" ***'***"

, ,

. 4 1979

'

-

68 I 19BN ANNevEnsARY

-

" " '$,0 ," u"*, """

June 19, 1987 i

Hr. Burt Davis, Regional Administrator U.S. Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn, IL 60137 SUB .'ECT: National Board Audit Team's Final Report of the Phillips Getschow N-5 Data Report Program for Unit 1 Braidwood Nuclear Power Station, Braidwood, Illinois FINAL REPORT

Dear Mr. Davis:

As requested, the National Board Audit Team undertook an audit of the Phillips Getschow (PG Co.) N-5 Data Report System. A sample of twelve completed N-5 packages selected at random for ASME Construction, Class 1, 2 and 3, were audited over a fifteen (15) day period during November and December, 1986.

This audit resulted in the identification of eleven (11) findings. Tbese findings were verbally transmitted to the NRC on February 18, 1987.

On April 9, 1987, the National Board met the NRC in Glen Ellyn, Illinois to discuss the above mentioned findings and the proposed corrective action.

The Braidwood site was visited on April 10, ?' a by Messrs. C. S.

Lindbeck and R. Siever. The purpose of this visit was to verify the corrective action taken by CECO and PG Co. to close out the identified findings.

As a result of the April 10th visit, the National Board was able to close out all identified findings except findings 1, 3 and 11.

The National Board Audit Team agreed to research the ASME Code requirements regarding the inclusion of component support identification on Piping N-5 Data Reports. The results of this research resulted in the closure

- of finding 1.

The NRC agreed to follow the proposed corrective action to close findings 3 and 11. The NEC notified the National Board on June 15, 1987 that findings 3 and 11 had the agreed upon corrective action taken and these findings could be cleared.

6' 7 M M M R .

- . .

__

_ _ - _ _ _ _ - _ _ .

'

• *

.. .

• '

.,

~

.

This report identifies the findings and corrective actions taken to close

.these findings.

NA-8400 1. FINDING Code Case N-272 It is our opinion that the presant documentation of the hanger installations does not meet the intent of NA-8400, Code Case N-272, or PG Co. QAP/BR-NS. If separate N-5's are'used for piping systems and for component supports, the system N-5 should reference the component support N-5 unique identifying number and vice versa, instead of the. term "Not Applicable" entered on the N-5 under Item 5 on Page 1.

RESPONSE  !

After a rereview and reevaluation, it was determined that PG Co. N-5 reports met ASME Code requirements.

CORRECTIVE ACTION None Required.

This finding is closed.

NA-8'416 2. FINDING One piping subassembly (RE-3-1) manufactured by Southwest Fabricating & Welding Co., Inc. (SWF) and two piping subassemblies (JRE57A-3/4" & 1RE25A-3")

installed by PG Co. are referenced on supple. ental records and are included in the written hydro boundaries; however, the subassemblies in question are not listed in the N-5 for 1RE-1.

RESPONSE N-5 Data Report for 1RE-1 has been amended to include RE-3-1, 1RE57A-3/4" and IRE 25A-3".

CORRECTIVE ACTION An audit was performed of ten (10) additional systems and no deficiencies were identified and we have determined this was an isolated instance.

This finding is closed.

N. .. .

.

,

'. "

Code Case N-272 3 ,, FINDING Code Case'N-272, which is used to justify not attaching all N-2, NPV-1 and NPP-1 data reports to the applicable Form N-5, is not referenced on the N-5 data reports.

RESPONSE CECO has revised the Design Specification L-2739 to allow the use of Para. NCA-84JU Summer 81 Addenda which  !

incorporates the provisions of Code Case N-272.

CORRECTIVE ACTION An attachment sheet shall a attached to the N-5 o data reports listing all Code Cases that may.have been used.

This was verified by the NRC.

.This finding is closed.

NA-4451 4. FINDING NA-4452 NA-4831/32 A review of the N-5 Data Pack for 1RY-1 disc 1: sed that the package is missing construction documentation to support three Class 1 welds, one 45 ell, and one piece of SA-376 Gr TP304 6" nom. pipe. The field welds in question are PG-3E, PG-3F and PG-3G,' located within line IRY01B-6". The reference drawing for this line is IC-RV-5 Rev. B. Each one of the three welds in question is missing the following installation support documentation:

a) Field Fabrication Process & Data Sheet b) Temporary Attachment Control Sheet c) Liquid Penetrant Examination Report d) Weld Radiograph Inspection Report The National Board Audit Team is of the opinion that the Data Report for 1RY-1 is not valid until these three construction packages are located, reviewed, and accepted per PG Co.'s QA program.

RESPONSE Supplementary documentation was reviewed which allowed recreation of the installation documentation for welds 3D, 3E and 3F. Multiple documents exist that allow for j

'

recreation of documents that are damaged or lost.

i

i

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

- .

-

.4 . .

..

..

CORRECTIVE ACTION None required.

This finding is closed.

NA-4440 5. FINDING

.A review of the Field Fabrication Process'and Data Sheets for line 1RY01B-6", Drawing 1C-RY-5, disclosed the lining out of material specifications that have been entered on the FFP & D. The FFP & D for Weld Map No. 1C-RY-5-2, joint number RY-5-FW2A, stated that SA-182 F304 material was to be joined to SA-182 F304 material. One of the SA-182'F304 entries was lined out and a. handwritten entry of SA-376 TP304-6" pipe was entered. This change was made on the FFP & D.without date or initials of the person making the change.

NA-4452 6. FINDING 1he above FFP & D recorded the changing of Welding Procedure Specification revision levels four Jears after the welding had been performed. The FFP & D had an original WPS entry of 1A-MA-88 Rev. 10. This entry was made on 8/22/79. The work that was carried out to this WPS was performed on 7/1/81 through 7/10/81. At some later date, WPS 1A-MA-88 Rev. 10 was lined out'and a revision level of 8 was assigned to the WPS without an entry date for the change. At some even later date, the revision level of the WPS revision 8 was lined out and a revision level of 9 was assigned without an entry for the change.

Based on the information given on the FFP & D, the National Board Audit Team feels-that document revision control in effect from 8/79 through 7/81 may not have been in compliance with the ASME Code.

NA-4452 7. FINDING NA-4530 )

The same FFP & D also recorded the assignment of the Visual Examination Procedure QAP-2 Rev. lA to examine the weld. A review of the installation document package revealed that the visual examination had been performed to QAP-2 Rev. IA on 7/8/81. However, at some later date, QAP-2 Rev. 1A was lined out and a visual procedure of VE-01 Rev. 2 was assigned without an entry date for the change. On 11/18/85 the visual procedure VE-01 Rev. 2 was lined out and a revision level 0 was )

assigned.

i

'

_ _ _ _ __ __b

!" . . . .

.

'

.

i .

RESPONSE (Items 5-7)

A review of additional documentation and field walk-down of the piping systems was performed to verify the changes made to the documentation reflected the field conditions and found acceptable.

The document control program in effect after 1981 requires change control that will not allow changes without the date and initials of the person c;aking the change.

CORRECTIVE ACTION None required.

These findings are closed.

NA-4442 8. FINDING NA-8416 N-5 #1CS-1 referenced PG Co. Drawing IC-CS-5 Rev. F, at location C-7, shows 3/4" globe valve #1CS 074A (listed on N-5, Page 3, Item 5, Line 7) and piping ICS55AA-3/4" between field welds 7E-1 and 7D-1, and 3/4" S.O.L.. fitting to spool CS-5-7. 1CS55AA-3/4" is not listed under Piping Subassemblies on N-5, Pages 4 to 7, nor as additional material on Page 8.

NA-4442 9. FINDING NA-8416 Similarly, PG Co. Drawing 1C-CS-6 Rev. F, at position A-7, shows ICSSSAB-3/4" and 3/4" S.O.L. fitting between 3/4" globe valve ICS074B and spool CS-6-7, with no listing of piping subassembly or additional material.

NA-4442 10. FINDING NA-8416 PG Co. Drawing 1C-CS-6 Rev. F, Note 5, states "For penetration information see Drawing IPC-16A", which is not attached or listed on N-5, Page 2 drawings.

RESPONSE (Items 8-10)

After a review and reevaluation of PG Co.'s procedure it has been determined that this is an acceptable practice.

CORRECTIVE ACTION None required.

These findings are closed.

/

'

F.c. .

.

.

..

NA-4442 11. FINDING NA-4450 ICS-1 N-5 record package for CS-6 portion of this system did not include an FC0 or other documents to document the modifications to this piping subassembly.

In order to assure material identification for modified SWF subassemblies, the FCO's that directed and controlled the modification of the subassembly must be a part of the N-5 record package. Further,

'

the use, existence, and maintenance of FCO's are an integral part of the PG Co. Material, Traceability and Verification Program.

RESPONSE PG Co. has. located the FCO's that verify-the modifica-tions and will return them to the N-5 package.

CORRECTIVE ACTION CEC 0 QA will require PG Co. to return all FCO's to the N-5 packages.

This item was verified by the NRC.

This finding is closed.

CONCLUSION It must be stated that in many instances, the problems identified occurred prior to 1981. Since 1981, the PG Co. QA program has been effective with rare exception.

In addition to the identified findings, the National Board expressed verbally to the NRC many concerns. These concerns were primarily addressed because of the difficulty encountered trying to comprehend the complexity of the PG Co. program.

It is our opinion that the program is poorly written and very confusing and difficult for ai. outside auditor to absenb in the limited time allowed for performing an audit. However, there is no indication that the program, as written, does not meet ASME Code requirements, t

'

The National Board Audit Team would like to express their gratitude to !

both CECO and PG Co. for their cooperation in closing the identified findings. We would also like to compliment both organizations for the professionalism exhibited.

Finally, we would also like to thank the NRC for their able assistance in verifying closure of specific items.

l

-_____- __ ._ _ _ _ _ _ _ _

', ' *

, .

,

,

With the issuance of this report, the National Bocrd considers this audit I as complete and closed.

Again, we wish to thank you for the opportunity to perform this audit. I If we can be of further assistance again, please contact the Director of {

Inspections or Mr. D. J. Mcdonald, Executive Director. )

Re et , I su i d, i

y .

Michael F. Sulifvan Director of Inspections C. S. Lindbeck v

Robert Siever MFS:j1 cc: D. J. Mcdonald C. W. Allison

l

_ ____ ___D