ML20237L183

From kanterella
Revision as of 08:32, 24 January 2021 by StriderTol (talk | contribs) (StriderTol Bot insert)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Partially Deleted Review of Concerns Re NRC Insp Program for Facility.Adequacy of Region IV Insp Effort & Downgrading of Proposed Violations in Insp Repts Addressed
ML20237L183
Person / Time
Site: Comanche Peak  Luminant icon.png
Issue date: 07/08/1986
From: Thomas Scarbrough
NRC ATOMIC SAFETY & LICENSING APPEAL PANEL (ASLAP)
To: Mulley G
NRC OFFICE OF INSPECTOR & AUDITOR (OIA)
Shared Package
ML20237F760 List: ... further results
References
NUDOCS 8708200113
Download: ML20237L183 (36)


Text

  • '

,o ,

4 1

4 July 8, '986 1

I I

  • l l

MEMORANDUM FOR: George A. Mulley, Jr., OIA FROM: 06$ Thomas G. Scarbrough, ASLAP i

{

SUBJECT:

COMANCHE PEAK As requested, I have reviewed the concerns expressed by H.S. Phillips with respect to the NRC inspection program for I the comanche Peak nuclear power plant. Below, I address

(

those concerns which.I consider most significant.

{j

1. Adequacy of Region IV Inspection Effort )

Beginning in August 1984, H.S. Phillips and'NRC contract personnel conducted an evaluation of the effectiveness of the Texas Utilities Generating Company (TUGCO) management at its corporate office in the implementation of the corporate QA program. This i {

evaluation, which was documented in Inspection Report 50-445/84-32, 50-446/84-11 (hereafter 84-32/11), identified several weaknesses in that implementation. In particular, several Level IV violations were identified as a result of the applicant's failure (1) to regularly review the status and adequacy of its QA program and (2) to establish and implement a comprehensive system of planned and periodic audits of safety-related activities. Further, it was found that the TUGCO corporate office had not been comprehensively audited by the NRC since 1974. (The construction Permits 8708200113 870819 Attachment HH

' PDR ADOCK 05000445 O PDR

l 2

for Units 1 and 2 were issued in December 1,974.) According  !

j to Phillips, T.F. Westerman ('a branch chief in Region IV who i l

later became Phillips' immediate supervisor) was involved 4

with the NRC inspection of Comanche Peak during that period. -

3/19/86 Tr. 22.

Westerman was said to have defended his inspection efforts by insisting to Phillips that the NRC audits had been performed. Ibid.

As a result of the limited NRC inspection at the TUGCO rcorporate 5 office in the past, B.S. Phillips was requested by g a division director in Region IV It that time, to review previous inspection reports and comp re them to the guidance of IE Manual Chapter 2512. The pr uct of that review was a memorandum dated January 13, 1986 ith '

o enclosures from I (Phillips' branch chief at that time) to E.B. Johnsona (the current division director) which addressed in detail the NRC inspection effort for Comanche Peak.

According to the memo, the Region IV QA inspections for the plant have been weak.

Further, several technical areas were identified as requiring additional inspection for Unit I while it was suggested that Unit 2 inspections be.

commenced expeditiously.

From a trend review of violations and unresolved items, a few areas were noted in that memo to-need additional inspection effort. i

a. i With respect to the alleged weakness in the Region i j

IV QA inspection effort, I have reviewed the QA inspection j

reportt, the Form 766 data computer printout, and the q t

, .. y l

j 9'

. 3 individual construction OA inspection proced,ures listed'in Chapter 2512.- Below, I discuss the extent to which each construction OA inspection procedure has been satisfied. j

..l Inspection Procedure 35020 Audit of Applicant's Surveillance of Contractor QA/QC Activities Chapter 2512 states that this inspection should be performed 5 months after docketing'and then as needed. The Form 766 data printout lists 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of inspection in report 78-07, 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> in report 78-09, 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> in report 84-29, and-120 h'ours in report 84-32. The initial inspection in report 78-07 included part of the preliminary effort but did not observe the implementation of the program as required by the inspection procedure. Inspection report 78-09 was a brief f.

close-out of an unresolved item. Inspection report 84-29 was the initiation of a thorough' inspection that was co;apleted in report 84-32.

Although the first inspection should have been performed in mid-1975, it was not commenced until early 1978 and even then was nog adequate'. Thus, an inspection completely satisfying this procedure was not conducted until i mid-1984.

i Inspection Procedure 35060 j Licensee Management of QA Activities i Chapter 2512 states that this inspection should be performed 6 months after construction Permit issuance and I q

1 l

I i

- - _ _ _ - - . _ - - - - - - - . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ^ - - - - - - - - - ~ -- - - ^ - ~ ~ ~ - ~ ' ~ ^ ^ ~ ~ ~ ^ ~ ' ^ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~

4 4 ,

then every 18 months, according to Appendix A of that -

i chapter-(Appendix C, which is being superseded by Appendix A, states simply every 18 months). The 766 data printout indicates that this inspection procedure was initiated in

- 1984, approximately 9 years after CP issuance. Inspection reports provided for this investigation do not discuss any .

NRC effort at the TUGCO corporate offices relating to this i l

inspection procedure, except for the few hours of vendor i l

control in inspection report 78-07 for procedure 35020. (As I every inspection report prepared during the construction of i 4

, Comanche Peak was.not reviewed, it is po'ssible that some ,

work in this area could have been documented in t.nother I

report.) In reports 84-74 and 84-32, an extensive effort j i

,. was made to completely perforn procedure 35060.

Inspection Precedure 35061 In-Depth QA Inspection of Performance Chapter 2512 states that this inspection should be per-formed annually. The 766 data printout shows only report.

84-22 for the performance of this inspection. This is somewhat misleading because nearly all of the reviewed '

inspection reports provide some inspection of performance of site work. See, for example, reports 78-07, 78-20, 78-21, 80-07, 84-10, and 84-26, It is clear, however, that the Region IV inspection effort in this area was weak.

q

-L n

5-Inspection Procedure 35065, l

Procurement, Receiving, and Storage l Chapter 2512 states that this inspection should be performed 12 months after Construction Permit issuance and then every 24 months, according to Appendix A of that chapter (Appendix c states annually). The 766 printout list shows inspection report 84-26 as the only report concerning this procedure. However, other reports also have documented NRC inspections of procurement, receiving, and storage of material at the site. See, for example, reports 78-07, 78-20 and 78-23. These efforts were not as extensive as l .

l required by procedure 35065.

Inspection Procedure 35100 Review of OA Manual  ;

s.

, Chapter 2512 states that this inspection should be performed as referenced in the applicable inspection procedures. The 766 data printout indicates only inspection report 84-22 addressed this inspection. However, reports 78-20 and 78-23 involved review of QA manuals. It was said recently that the applicant is revising its OA Manual so the adequacy of those earlier NRC inspections is open to question.

Inspection Procedure 35200 Mid-term Construction Permit QA Inspection Chapter 2512 does not list this inspection. The 766 data printout lists only 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> of inspection in report _j 78-20 and 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> in report 78-23. While detailed l I

!. -. l 6

requirements for the extent of this NRC insp,ection were not <

located, it can be said that a 40-hour effort would not be sufficient to address all aspects'of the construction QA program in detail. It was stated by Phillips that this procedure was superseded by 35060 in order to increase the frequency of NRC inspections of the-applicant's QA program.

6/26/86 Tr. 41. As discussed above, however, inspection procedure 35060 was not fulfilled on schedule either.

i Inspection Procedures l 35740-49 Operational QA Pr gram In inspection reporta 84-31 and 85-8, extensive NRC  !

inspection effort has been made in re iewing the operational QA program for Comanche Peak. These nspections may help resolve concerns for adequate auditing by utility personnel a.nd safekeeping of records. With respect to NRC confidence in past construction activities at the pisnt, these inspections provide only limited support. l Despite Phillips' strong views on the subject, concern ' s for the adequacy of the Region IV inspection effort was not shared by all of the individuals interviewed by OIA. For example, those individuals that essentially believed that weaknesses had ~y existed in the Region IV inspection program i were l(aRegionIVsectionchief),)

'(a Region.IV inspector), T.H. Young (an NRC consultant), and Ja sutton (an applicant consultant). Those individuals that, for the most part, considered the Region IV effort, and the d

4 L - __________________________-_______ _ :___ - ___:. -

7

)

applicant's QA program, to.be adequate were,' Region IV inspectors [ and R.C. Stewart and NRC consultants 4 J.E. Birmingham, W.D. Richins. D.L. Jew, J.R. Dale, . M.K.

Graham, and A.L. Maughan. It should be noted however that

]andR.C.stewartwereinvolvedwiththeRegionIV .

effort at Comanche Peak for several years; thus, their views maybesomewhatbidsed. Also, most of the NRC consultants I I

had bgen working at Comanche Peak for less than a year.

, ]

Based on my review, I agree with the January 13, 1986 i

) memo that the QA isspection effort by Region IV for ComanchePeakhasbeen'wekk. As just shown, a significant

~ number of the QA inspections specified by Chapter 2512 were '

not performed in a timely mlanner. Thus, it is not possible S

to rely on that past Region IV QA inspection effort as .

strong support for the safe construction of the plant.

Other efforts such ac the Construction Assessment Team Inspection 83-18/12, Region IV Special Inspection 84-26, and the Technical' Review Team work (Supplement 11 of NUREG-0797) will pave to be relied upon to determine the adequacy of

, plantfconstruction.

I

  • It is understood that weaknesses in the Region IV inspection effort does not necessarily mean that the applicant's QA program has been inadequate.

However, recent inspection efforts have identified deficiencies in the applicant's implementation of its QA program. For example, the Technical Review Team found that "[t]he pattern of - <

' ' ~

.______.-______-m.____m-__- _-_

~

8

. 1 failures by QA and QC personnel to detect and document deficiencies suggests an ineffective B&R [ Brown & Root]-and TUGCO inspection system." NUREG-0797, Supplement.11, at.

P-35. Thus, it cannot be said that the deficiencies in the applicant's QA' program were simply paperwork mistakes.

b. I ' memorandum, additional emphasis by

\ n the' - l NRC.i'nspectors was suggested for seven technica1' areas in '

the construction of Unit 1 of Comanche Peak based, on a review of the NRC inspection program. It was also noted that the review had revealed a need to commence the required i

/

inspections at U it 2. To assist in dete'rmining the extent (in inspector-hours) of the NRC effort for each of those i

areas (listed'b'elow), I reviewed the.766 data printout for a ,

I Units 1 and 2 of Comanche Peak (CP) and Millstone Unit 3 (as an example of a typical plant).

1. Structural Steel'Outside containment - Inspection Procedure 48063 (work observation)

CP-1 : 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> - CP-2 : 11 bra. Millstone : 62 hrs.

I

~

2. Reactor Vessel - I.P. 50055 (records)

. CP-l': 13 hrs. CP-2 : 4 hrs. Millstone : 8 hrs.

3. Safety-Related Mechanical Components - I.P. 50073 (work observation)

CP-1 : 62 hrs. CP-2 : 52 hrs'. Millstone : 252 hrs.

4. Electrical Components and Systems - I.P. 51055

., (records) l CP-1 32 hrs. . CP-2 : 15 hrs. Millstone : 69 hrs.

I

5. Instrumentation Components and Systems - I.P. 52054 (work observation)

CP-1 : 42 hrs. CP-2 : 0 hr. Millstone : 50 hrs.

l

6. Containment Structural Steel Welding - I.P. 55063. ~. j (work observation) 9 0

0

.a

~

9 9

s CP-1 : 27 hrs. CP-2 : 13 hrs. Millstone : 23 hrs.

7. Containment Structural Steel Welding - I.P. 55065 (records)

CP-1 : 10 hrs. CP-2 : 1 hr. Millstone : 3 hrs.

]

This review of the 766 data printout generally. supports the

~

view that the past Region IV inspection program for Comanche L Peak has been weak. In particular, the Region IV effort in inspection of structural steel outside containment (work '

]

observation), safety-related mechanical components (work 1 l

l observation), and electrical components and systems (records) fo'r CP-Unit I was significantly less than at Millstone. All of the listed areas appear weak'for Comanche i Peak Unit 2.

In the March 19, 1986 interview, Phillips stated that he had'been assigned the task of ensuring that the NRC .

inspection procedures had been satisfied. One part'of his task was to determine whether special inspections (such as  !

CAT 83-18/12) performed over the last couple of years had been sufficient to complete the NRC inspection program.

)

Phillips recently said that Westerman told him not to 6a anything with the instances where the NRC inspection procedures were not satisfied. 6/26/86 Tr. 36. The results  !

1 of that review by Phillips should be used in assessing those l areas where additional inspection effort is necessary.

In a June 26, 1986 interview, Phillips expressed a concern for the adequacy of NRC inspections in the  :

4 electrical area at Comanche Peak Unit 2. The 766 data

- -_---_2 3

i

. /

1 10 printout indicates the following inspector-hours for the inspection procedures of electrical components and cable, and instrumentation.

Electrical Components and. Cable 51051 Procedure Review '27 hours3.125e-4 days <br />0.0075 hours <br />4.464286e-5 weeks <br />1.02735e-5 months <br /> l 51053 Work Observation 7 hrs, i 51054 Work Observation 0 51055 Record Review 15 hrs.

51056 Record Review 0 51061 Procedure Review 6 hrs.

51063 Work Observation 28 hrs. ,

51064 Work Observation 0 '

51065 Record Review 0 51066 Record Review 0 Instrumentation 9

52051 Procedure Review 6 hrs.

52053 j Work' Observation 10% (no hours listed)

, 52054 Work Observation 0 J

52055 Record' Review 0 52056 Record Review 2 hrs.  !

52061- Procedure Review 0 ,

52063 Work Observation 0 l 52064 Work Observation 0 52065 Record Review 0 52066 Record Review 0 52153 Work Observation 0 i

As can be seen, the Region IV inspection effort at Comanche Peak Unit 2 has been minimal in areas of electrical l

components and cable, and instrumentation. Recent NRC '

inspection efforts for Unit 2 could be filling these gaps.

For example, A.L. Maughan, an NRC consultant at the plant, stated in a June 26, 1986 interview that he had been -

reinspecting electrical areas in Unit 2 for about a month and had seen no real problems. -<

l On the basis of the past i

l

i

)

11 1

l l

weakness in the Region IV inspection effort / it is clear I i

t that detailed inspections will be necessary in the I electrical and instrumentation areas.

c. In the memo, three areas (design control, inspection, and purchasing) were identified as exhibiting adverse trends such that additional inspection in these areas may be warranted. In the June 26 interview, Phillips 1 I

l stated that be had not had time to investigate these adverse i trends but suggested that such an investigation would be benef cial for Unit 2. I agree.

2. Downgrading of Proposed Violations in '

Inspection Reports According to Phillips, violations that had been ,

a Proposed by him and other inspectors as a result of their findings during inspections were unjustifiably downgraded to

, i unresolved items by Region IV management. In particular, Phillips asserts that T.F. Westerman and/or E.H. Johnson were responsible for the downgrading of violations in inspection reports 85-07/05, 85-14/11, 85-16/13, and B6-06/04. Below, I discuss each of these reports.  !

a. Report ~ 85-07/05

~

Several violations proposed by Phillips and to be included in inspection report 85-07/05 were downgraded to 1 e

6 I

12

" unresolved" items at the request of Westerman. While the draft report is not available, the final report supplied by Phillips has the proposed violations attached. I have considered each of the proposed violations.

(1) Proposed violation: " Contrary to 10 CFR 50, Appendix B, Criterion III, and TUGCO QA plan, dated July 31, l

1984, (a) Unit 2 reactor pressure vessel installation design criteria-recommended by NSSS vendor, such as centering toldrances, levelness tolerances, and shoe to bracket clearances, were not included in installation specification, l

procedures and drawings; and (b) the criteria were specified in Construction Operation Traveler ME-79-248-5500 but were not treated as design engineering ::iterie as evidenced by.

an undocumented change of shoe.to bracket clearances."

A similar violation was proposed that:

" Contrary to 10 CFR 50, Appendix B, Criter' ion XV and

. TUGCO QA Plan dated May 21, 1981 and Brown and Root Quality Assurance Manual, Section 16, dated March 27, 1985, clearances between the reactor vessel support brackets and support shoes were not within the tolerance stated in Construction Operation Traveler ME-79-248-55 and the condition was not reported on a Nonconformance Report."

Phillips stated that Westerman objected to these  ;

proposed violations without a given reason even though the applicant did-not object. See 3/19/86 Tr. 104.

l

-. . .- - ~

13 The final report described these findings (at p. 14) as follows:

Neither site prepared installation drawings nor specifications (which implemented the WNSD 1 recommended procedures) were available and the '

drawings examined did not show certain specific installation criterion such as centering l tolerances, levelness tolerances and clearance  ;

between support brackets and support shoes. The inspector considers this matter unresolved. . . . I The installation data reflected in aFtachment 3B of the traveler indicated an as-built cicarance of 0.012 to 0.026 inch which exceeds both the original and revised tolerances. This condition was accepted on the traveler based on Westinghouse-concurrence, and there was no documented engineering evaluation onsite justifying the final  ;

tolerances. This matter is considered unresolved pending documentation validating the final installation tolerances.

In my opinion, failures to properly control design .

i changes and to document a nonconformance are violations of the quality assurance requirements of Appendix B to 10 CFR Part 50. For eIample, Criterion III, Design Control, of Appendix B states that "[mleasures shall be established to assure that . . . the design basis . . . for those structures, systems, and components to which this appendix applies are correctly translated into specifications, l t

drawings, procedures, and instructions. . . . Design I changes, including field changes, shall be subject to design control measures commensurate with those applied to the j original design. . . .

Criterion XV, Nonconforming l Materials, parts, or Components, states that "[m]easures I shall be established to control . . . components which do -- d L.____.__m..____ - __-_.._. _ _ _ . _ - _ _ _ __ _ _

1

\

14 not conform to requirements in order to preve,nt their inadvertent use or installation. . . . Nonconforming items shall be reviewed and accepted, rejected, repaired or reworked in accordance with documented procedures."

l Similarly, ANSI N45.2 - 1977, Quality Assurance Program Requirements for Nuclear Facilities, which is endorsed by l Regulatory Guide 1.28 (Rev. <

2, 2/79), states (at pp. 7-8) i that "[m]easures shall require documentation verifying the acceptability of nonconforming items which have the disposition of repair or use as is."

! In Chapter 17.1, Quality Assurance during Design and Construction, of its Final Safety Analysis Report, the applicant states that its QA Plan meets the' intended objectives of the ANSI standards. FSAR (Amend. 25, 8/7/81) at 17.1-14. Sections 3.0 , Design Control (Rev. 2, 6/18/79),

and 15.0, Nonconforming Items (Rev. 2, 2/18/80), of the TUGCO QA Plan contain provisions equivalent to those found in Appendix B and the ANSI standard. Thus, the cited failures should have been classified as violations as opposed to unresolved items. I agree with the Severity Level IV suggested for the violations by the inspectors.

(2) Proposed violation: " Contrary to 10 CFR 50, Appendix B, Criterion XVIII and TUGCO QA Plan Section 18.0, there was no evidence that TUGCO had audited either Unit 2 reactor vessel installation specifications, placement procedures, actual hardware placement, or as-built records." ~'

15 Westerman opposed this violation on the, basis that the applicant was not required to audit that specific area. See 3/19/86 Tr. 107-108.

, The final report (at p. 15) read:

TUGCO did not make available any documentation of an ' audit or surveillance which evaluated specified placement criteria, placement procedures, hardware placement, or as-built records. This item is unresolved pending a more comprehensive review of these activities. .

While a specific audit was not necessary, there should have been QA coverage of this activity. TUGCO should be required to show evidence of QA involvement for the installation of the Unit 2 reactor vessel. It was acc'eptable to keep this item unresolved until that time, but a violation might be required later.

(3) Proposed violation: " Contrary to 10 CFR 50, Appendix B, Criterion VIII and TUGCO QA Plan, Section 8.0, Revision 0, dated July 1, 1978 and Article NA 3766.6 of

~

ASME,Section III, 1974 edition, spool piece 301 (DWG No.

BRP-CS-2-RB-76) had neither been marked with the material specification and grade nor heat number nor heat code of the material."

The inspector made his finding in June 1985, but in August the applicant found the identification number on the spool piece. See 3/19/86 Tr. 108-10. The inspector -

1 l rechecked the spool piece and found the number.

The final inspection report (at p. 17) states that L___.-._._.._____ ___ . - - _ - - - - - - - - - - - - - - - - - -

16 I

The installed spool piece was inspected for . weld quality and to verify that marking and traceability requirements had been met. The item had been marked with the spool piece number (301) and the B&R drawing number which provided traceability to the material certification.

I do not consider this item to be a violation.

Nevertheless, the report should have noted the fact that the I

l identification number was located after the initial drafting of the report. In this way, inspectors would be alert for a 1 i

pattern of such late identification.

1 (4) Proposed violation: " Contrary to 10 CFR 50, 1

Appendix B, Criterion XVII and Articles NB-2130 and NA-3767.4 of ASME Section III, 1974 edition, Certified Material Test Reports were not available for the 22 degree i s

elbow, 10-inch 45 degree nozzle and three thermowell bosses, which,were a part of the loop 3 cold leg piping subassembly."

)

In an interview, Phillips said that the violation was dropped but that he was not sure of the particular circumstances. See 3/19/86 Tr. 114-15. The proposed l violation has a handwritten note indicating that these records might have been located later. The final report '

does not discuss the initial unavailability of the CMTRs.

If the records were found, I see no problem.in deleting the violation but noting the delay in the inspection report.

ANSI N45.2.9 - 1974, Requirements for Collection, Storage, and Maintenance of Quality Assurance Records for Nuclear l

-l i

I I

t

i . .

17' '

Power Plants, which is endorsed by Regulatory Guide 1.88

( Rkif . 2, 10/76), states (at p. 4) that records must be

~

re,trievable "without undue delay" which is open to

~

interpretation.

(5) Also, in Report 85-07/05,- a violation was written as a result of the applicant's failure to provide record evidence that the concrete mixing blades had been inspected

~

quarterly since 1977. In the March 19 interview, Phillips stated that Westerman wanted him to take the verbal assurances of the utility that the blades had been

- inspected. Tr. 137. In the cover letter dated February 3, 1986, signed by E.H. Johnson for this inspection report, it' was indicated that no applicant response to this violation 5-was necessary because (1) the procedure had been revised to provide document,ed inspection of the blades, (2) no abnormal blade wear had been identified, and (3) there had been consistent concrete strength and uniformity tests. While Phillips did not consider this violation to be a hardware problem, he believedethat other procedures should be reviewed to determine if similar problems existed. Tr. 136.

While I do not see this situation as a safety problem,- I consider it to be an example of downplaying mistakes by the _

applicant on the part of T.F. Westerman and E.H. Johnson.

If an applicant mistake is of such significance as to be written as a violation, then it should not be negated by the cover letter'to the inspection report. ,

~

a- ~

1 t

, , l 18-

)

b. . Report 85-14/11 -

With respect to inspection report 85-14/11, Phillips stated that several violations proposed'by himself, NRC.

inspectort and NRC consultant T.H. Young were t

downgraded at the insistence of T.F. Westerman. Those si proposed violations ~(according'to a matrix prepared by I j

Phillips) and my opinion as to the validity of the. proposed

. 1 j

violations follow:

(1) Proposed violation: 'FSAR d'oes not describe TUGC0 Records system.'

FSAR chapters are somewhat general and it is not

\ surprising that there is no detailed discussion of the records system.

As this it'em is unresolved in final report, I do not see a problem with the management action.

(2) PropoRed violation:

"QA Manual does not address ANSI N45.2.9 requirements / commitments.' '

l This item is unresolved in the final' report on the basis that TUGC0 is revising its QA manuals. As TUGCO had committed to ANSI N45.2.9, this situation would not be of i

concern if TUGC0 was following N45.2.9 but, based on'other items in this report, that does not appear to be the case.

This is a violation of Criterion V, Instructions, Procedures, and Drawings, of Appendix .B. .

(3) Proposed violation:

'TUGC0 failed.to have/use procedures to' control shipment'of' original records to Stone

& Webster in NY.* -1 j

I 19 The final report has shortened the discussion but does state that this is a violation. Phillips asserte'd that Westerman wanted this item to be written as unresolved but l l

he accepted it as a violation when NRC consultant T. Young 3/19/86 Tr. 177.

1 insisted. The final report should have mentioned the inspector's concern for lack of inventory of ,

1 shipped records.

(4) Proposed violation: " Original design records shipped in cardboard boxes without making backup copy." )

The final report deleted this violation. According to Phillips, it was Westerman's view that there is no l

requirement to protect records during shipment. 3/19/86 Tr.

169-70. While ANSI N45.2.9-1974 does not specifically l

address shipment of records, it does provide requirements for the protection of records during temporary storage.

These requirements include placement in fire-rated vaults or

_ maintenance of a duplicate copy at another location. Thus, this item was a violation of Criterion XVII, Quality Assurance Records, of Appendix B.

(5) Proposed violation: " Failure to control and account for QA/ design records transferred from site to SWEC."

This proposed violation is similar to one retained so it is understandable that management deleted.

(6) proposed statements: "TUGCO stated' design records shipped without making backup copy because cost too much. -c

. l l

20 This relates to the cause'of violation. Aln.o stated it was company policy to proceed at own risk." j 1

I believe that it was within management's prerogative to delete these statements.

(7) Proposed violation: " Site records of Chicago l i

~

Bridge & Iron (CB&I) work shipped to Houston in cardboard )

i boxes. Originals subjected to little protection without l 1

retaining back-up copy at site."

Final report considers this item to be unresolved. j This is a violation of Criterion XVII of Appendix B, but it is not clear whether there was sufficient evidence to wri J the violation immediately as the shipment occurred months I earlier. 4 (8) Proposed violation: "TUGCO failed to inventory I

CB&I records sent to Houston. Therefore, cannot determine  !

records that must be returned."

_ This is a violation of Criterion XVII but, as stated above, it appeared to be based on second-hand information.

(9) Proposed violation: "TUGCO audited CBI, Houston.

In scope of report, it was stated that it included Criterion l

XVII, QA Records, but did not document the audit of records."

The final report did not discuss this issue. While this is a violation of Criterion XVIII, Audits, of Appendix B, it is minor in comparison to the misunderstanding of QA record control on the part of the utility. --<

21 (10) Proposed violation: " Failure to preclude rain from entering'QA records vault over several years time." 1 The final report did not discuss the rain problem but  ;

2 only indicated that weter leakage from the ventilation duct was an unresolved item. The rain problem was apparently based on statements by utility personnel so there was not a strong foundation for a violation. Nevertheless, the l problem with water from a ventilation duct was sufficient I for a violation. For, in this regard, Criterion XVII, Quality Assurance Records, of Appendix B states that

"(s]ufficient records shall be maintained to furnish evidence of activities affecting quality." More specifically, ANSI N45.2.9-1974 states that "[p]ermanent and temporary record storage facilities shall be so constructed or located as to. protect contents from possible destruction j 1

l by causes such as fire, flooding, . . . and possible

_ deterioration by a combination of extreme variations in temperature and humidity conditions. A satisfactory l

alternative to the establishing of a record storage facility is maintenance of duplicate records stored in a separate l

remote location."

(11) Proposed violation: " Failure to preclude food and coffee pot (fire hazard) from QA records vault."

The final report did not mention this item. This was a violation of criterion XVII of Appendix B but the inspector could have warned the utility this time. ~<

. 22 (12) Proposed violation: " Failure to p,rovide temporary

~

c or permanent storage facility for records 'omingled with in process document's in Paper Flow Group." ,

1 Storage of records in nonfire-rated file cabinets was 1

. deemed unresolved in the final report (at p. 13). Such storage without a backup copy is a violation of Criterion

  • XVII of Appendix B. See Item (10) above. According to Phillips in the March 19 interview, Westerman wanted this item to be considered unresolved because the applicant had agreed to install fire-rated record vaults. Tr. 180.

. Shortly before'the interview, Phillips said he learned that ,

the applicant had decided not to install the fire-rated cabinets. Phillips indicated tha~t he would try to upgrade s, this unresolved item to a violation. Tr. 181-82.

(13) Proposed violation: " Weld rod not identified at main distribution or distribution station."

The final report notes the loss of. identification but

_ simply refers the matter to the utility. This was a violation and should have been stated as such. Criterion VIII, Identification and Control of Materials, Parts, and Components, of Appendix B, states that "[m]easures shall be 1 established for the identification and control of materials, parts, and components, including partially fabricated assemblies. These measures shall assure that identification of the item is maintained by heat. number, part number, serial number,.or other appropriate.means, either on the -- d

1 l

i 23 f

item or on records traceable-to the item, as required throughout fabrication, erection, installation, and use of  !

the item." Section 8.0, Identification and Control of Items (Rev. O, 7/1/78) of the TUGCO QA Plan contains essentially equivalent provisions.

c. Report 85-16/13 According to Phillips, several violations and unresolved items proposed for inspection report 85-16/13 by himself, and NRC consultants J.H. McClesky and T.H. Young l 1

were downgraded or deleted by E.H. Johnson and T.F.

~

Westerman. These proposed violations and unresolved items (according to a matrix prepared by Phillips) and my opinion as to their validity follow:

(1) Action on 10 CFR 50.55(e) Deficiencies Proposed violation: TUGCO failed to develop /

implement a procedure to show or reference objective evidence that deficiencies were corrected.

~

Proposed violation: TUGCO failed to revise implementing procedures before corporate NEO i Procedure CS-1 was implemented, resulting in conflict with five other procedures.

Proposed violation: TUGCO failed to maintain 50.55(e) files (QA records) that were retrievable, i.e., could not produce record in almost a month.

l Proposed violation: TUGCO failed to report-to the

  • NRC the corrective action actually taken and changes to commitment regarding corrective action reported to NRC.

(Two other proposed violations listed by Phillips in the matrix appeared encompassed by those

.above.)

4 u____________._. _ . _ _ _ . _ _ . . _ _ _ - . _ - . - - - -

24 ,

In a handwritten note on a draft of report 85-16/13, E.H. Johnson stated that "most of the apparent violations cited are not violations. Records for the convenience of the NRC are not required." The final report indicated that deficiencies had not been corrected by the reported date nor l were supplemented reports provided to the NRC as to the 1

adjusted completion dates. The final report noted that TUGCO had assembled a task force to evaluate and take action on these matters. The item was stated to be unresolved.

The Commission regulations in 10 CFR 50.55(e)(3)~ state that "[t]he (50.55(e)] report shall include a description of the deficiency, an analysis of the safety implications and the corrective action taken, and sufficient.information to permit analysis and evaluation of the deficiency and of the corrective actio,n. .If sufficient'information is not

]

available for a definitive report to be submitted within 30

_ days, an interim report containing all available information shall be filed, together with a statement as to when a complete report will be filed." Further, Section 15.0, i

Nonconforming Items (Rev. 2, 2/18/80), of the TUGCO QA Plan requires that 50.55(e) matters be documented on a i

( nonconformance or deficiency report. As this information is required by the regulations and addressed in the TUGCO QA Plan, it must be available for inspect. ion. Thus, the proposed violations are valid as a result of the_ applicant's failure to comply with criteria V (Instructions, Procedures,

-4

q 25 '

and Drawings), VI (Document Control), and XVII (QA Records) of 10 CFR Part 50, Appendix B. In addition, failure to report corrective action is a violation of 10 CFR 50.55(e)(3).

The violations should have been included in the final 1 report. Further, the final report should not have taken  ;

credit for the TUGCO task force in deciding to consider this i

! tem unresolved. As. side light, it is not clear to.what extent E.H. Johnson considered 50.55(e) records to not be required. It may be beneficial to have headquarters QA personnel provide guidance in this area to the regions.

(2) Applicant Action on IE Bulletins Proposed unresolved item: TUGCO never responded to all aspects of IEB-14.

Proposed unresolved item: TUGCO IEB files for  !

1982 and 1985 di.d not contain sufficient records or  !

reference to records which show IEB action / corrective action complete.

Proposed violation: TUGCO had replaced NAMCO switches per IEB 79-28 but 2 of 14 that were field inspected were not properly identified on installation traveler.

Proposed violation: TUGCO procedures for handling IEB are deficient in that they do not describe how construction management / personnel handle IEB requiring action, especially hardware repair, replacement and modification.

Proposed violation: No TUGCO construction. focal point was found for tracking IEB actions.

On the replacement of the switches, Johnson said in a handwritten note on the draft report that "we are citing them for a potential hardware problem on-NAMCO switches --

,. c

9 26 find out if there is a hardware problem or not -- otherwise there is no citation."

The final repore discussed the improper identification of the two NAMCO switches on the installation travelers.

Rather than a violation, however, the report considered this  :

)

item unresolved until the applicant had an opportunity to l

determine if other documentation could account for this problem. Regardless of whether other documentation was discovered, a violation was called for as a result of improper or inconsistent documentation of an installed l . }

component. The final report should have considered this a violation of Criterion VIII, Identification and Control of i

Materials,' Parts f and Components, of Appendix B to Part 50. '

, Additional switches should have been inspected to ensure that this was an isolated incident.

As NRC personnel cannot inspect every component or structure in.the plant, it is important that the paperwork be correct to ensure that those not inspected are correct.

If only hardware is considered important, then a very large percentage of the plant hardware must be inspected to have confidence that the plant is properly built. This would include inspection of work in progress where the component or structure would become inaccessible at a later stage.

With respect to the other items related to IE Bulletins, I could not confirm any improper management action. The two proposed unresolved items were combined as mm c

F 27 an open item for IEB 79-14 in the final report. ,

There was not enough information to determine the merits of the other two proposed violarions in terms of whether violation or unresolved was more appropriate.

(3) Electrical Penetration Seals l The certification of BISCO electrical penetration I seals (fire barriers) is under review'as a result of questionable testing. A proposed violatio.n by T. Young was 'k changed to unresolved in the final report. Until all of the l facts are determined, I do not see a problem with this item being listed as unresolved. This item is potentially l serious, however, and;it must be investigated thoroughly by the staff.

l d. Report 86-06/04 .

At an interview on June 26, 1986, Phillips stated that he had identified several violations concerning the

. Brown & Root audit program. While the draft report was not available, Phillips described the important violations which  !

involved (1) the failure to.take corrective action on audit deficiencies by TUGCO from 1980 to 1986, (2) failure to 4 audit the entire QA program during 1980, 1982 and, possibly, 1984, and (3) failure to take corrective action .According i to Phillips, Westerman told him to write these as unresolved items despite Phillips' view that they were violations.

After the exit meeting with the applicant where these were

~

discussed as unresolved items, Ian Barnes-(who aoparently

___._._a_.__---- _---

i i .

2'8 had replaced Westerman as Phillips' immedia.te supervisor) i J

stated that these should be written as violations. At that point, Phillips requested that these remain as unresolved for the present because (1) the applicant had been told that a these were unresolved items at the exit meeting, (2) the I report was being finalized,'and (3) additional problems with the B&R audit program had been discovered by Phillips such that it'might'be better to assign a more detailed violation '

later. Thus, the report will apparently refer to these -

items as unresolved. Phillips stated that he had discussed this report in mid-June with Robert Mart-in'.who had indicated '

that these items were violations.

As the draft report was not available for my review, I have not been able to consider the proposed violations directly. Nevertheless, I believe it safe to assume that these described items were violations. Until more information is provided, the severity of these violations and their significance toward plant safety are not evident.

This episode, however, does add credence to the view that Westerman tends to downgrade violations. Further, although in earlier inspection reports Barnes was said by Phillips to support Westerman's arguments to downgrade violations, Barnes here considered these items to be violations.

Based on my review of these inspection'repor.ts, I believe that there was an effort on the part of T.F.

Westerman and, possibly to a lesser extent, E.H. Johnson to ~

29 l

l limit the number of violations assessed agal'nst TUGCO. As reinforcement of my belief, Phillips stated that, at an exit i meeting for inspection report'86-08/06, the applicant requested that a deviation be written as a violation despite~

arguments by Westerman and Barnes that it was not a violation. 6/26/86 Tr. 26-27. I consider this downgrading of. violations to be contrary to the guidance of IEE Manual Chapter 0215 on inspector objectivity. ,

While I agree that Region IV management is correct in'

.)

attempting to make inspection reports concise, there were several instances in the reports where the severity of the identified problem was downplayed. The instances of management interference,.however, did not appear to h' ave an- J s

adverse effect on the safety of the plant. It is apparent 1y' more of a case of poor management practice.-

3. Accuracy of NRC Form 766, Inspector's Report, Data Following each NRC inspection, Form '766 (Inspector's I

Report) is completed in order to track the status of the NRC J

3 i inspection program by means of computar files. According to

'N 1 N.

H.S. Phillips, the data from these 766 forms for Comanche. s

  • Peak are inaccurate. The extent of this inaccuracy is
  • s .

.c

]

documented in the January 13, 1986 memo fron i

3 to Johnson.

- y, 11 ,

While I have not yreviewed all of th4' instances of i

asserted error in the 766 data,k. have checkef a ra:fficient s (4- w L a number to determine that'the Comasche'Psck;date are. ~

~ Vi l .#f.

{

    • ^-

(s 4 b "? C

_ ( .t M._._-.__.-._---.---.--t- -- '

30 unreliable. For example, inspection report'80-26/26 ,

discusses inspection of the reactor vessel internals.

However, the 766 data for that report indicates that the i

reactor vessel (as opposed to vessel internals) was inspected. Upon. discovering the weakness in Region IV

inspection of the reactor vessel, Phillips requestedg .

to perform an inspection although it was necessarily limited to some extent because, of the completion of the work. 3/19/86 Tr. 56. - '

i Discrepancies are evident in inspection report 85-07/05. In that report, inspections of structural steel outside containment and reactor vessel internals '

installation are not discussed while the 766 forms attached -

to the report indicate that the inspection procedures were 100% complete in these areas. Phillips also indicated that he had found some cases where conducted inspections were not included in the 766 data. 3/19/86 Tr. 129-30.

It appears that Region IV personnel have differing views.as to the proper method of completing the 766 data a forms. For example, Phillips believed that an inspection report should discuss every inspection attributed to that i report by the 766 data. As opposed to that view, however,

. ,, etated in an interview on May 28, 1986 that he believed that inspection modules could be listed in 766 data for any report' merely to close them out. Tr. 19-20. In addition, there appears to be a difference of opinion on

~~

\ b

. j 31 J

whether the percent completion listed in the 766 data is

~

ralated to the inspection procedure itself of, as suggested (5/28/86 Tr. 18), to the plant completion status.

The regions should be provided guidance for the proper completion of this form so that the data vill be consistent j and useful. Until the 766 data are reviewed and corrected, .

I would recommend that the data not be used in any licensing

~

decision. '

4. Freedom of Information Act (FOIA) Request In the Narch 19 interview, Phill'ips state'd that ,

Westerman had told him to destroy drafts of inspection report 85-13/09 because an FOIA request concerning that k report would soon arrive. Tr. 162. From the interviews  !

with other personnel, I have not found any confirmation of I this asserted action by Westerman. Because of the serious n'ature of such a, charge, I would need more information 3 before determining its accuracy. I do believe, however,

, that the potential for problems stemming from allegations that.FOIA requests are being compromised makes it imperative l that all NRC personnel be made aware of the requirements for handling those requests. ,

5. Fort St. Vrain Concerns 1 1

From his discussions with other NRC personnel, Phillips has become concerned that Fort St. Vrain has not.been cited l

. for violations, when necessary. According to Phillips,  ;

1 Region IV management directed the senior resident inspector

  • W w ___ _ - __ - __ _ - _ _ _ _ -

32,

\

at that plant not to write violations there, Supporting this assertion, P.C. Wagner (a past NRC project manager for' Fort St. Vrain at Region IV) stated in an interview that the i

Region IV management was giving preferential treatment to  ;

that plant. 4/9/86 Tr. 4. Wagner did not believe that this action caused any risk to .he public but considered this more of a regulatory problem. Tr. 24.

. I do not have any documentation that demonstrates an effort by Region IV management to downplay concerns at Fort St. Vrain. Nevertheless, I believe that Wagner makes a good case that such an effort exists. As there is no evidence of I

a safety problem, it may be possible to resolve this matter j through discussions between NRC headquarters management and Region IV management.

. CONCLUSIONS Based on my review, I consider the QA inspection effort by Region IV for Comanche Peak over the years to have been inadequate. As discussed, the QA inspection procedures recommended by I&E Manual Chapter 2512 were not satisfied as to their scope of evaluation nor their frequency of performance.

It is therefore clear that it will not be possible to rely on the QA inspection effort by Region IV as evidence of the safe construction of Comanche Peak.

To be sure, weaknesses in the Region IV QA inspection effort does not necessarily mean that the applicant's OA program was inadequate. However, an evaluation of '

~

33 j allegations by the NRC Technical Review Team-(as discussed in Supplement'11 to the Comanche Peak Safety Evaluation i Report) and a recent QA inspection by Region IV (as discussed in inspection report 84-32/11) reveal that the implementation of the applicant's QA program has been deficient in several areas. The deficiencies in QC inspection by the applicant as identified by the NRC Technical Review Team is of particular concern. As a result, it will be necessary to rely, for the most part, on l the detailed technical inspections of various structures, i i systems, and components that have been conducted, and are currently underway, at Comanche Peak. These include l inspections of technical work by Region IV over the years, the Construction Assessment. Team inspection 83-18/12, the i Region IV Special Inspection 84-26, and the Technical Review Team activities.

. With respect to technical aspects of plant 1

construction, the Form 766 data printout of the Region IV inspection effort suggests that several areas may not have received sufficient attention. For example, the Region IV effort for Unit 1 in inspection of structural steel outside containment and safety-related mechanical components appeared weak. The Region IV inspection effort for Unit 2 was even more deficient as illustrated by the small number of inspector-hours devoted to electrical components and cable, and instrumentation. -<

To ensure that all of the weak

34 areas in the Region IV program are identified for additional inspection,'the January 13, 1986 memorandum from, to E.H. Johnson and the results of H.S. Phillips' ongoing review of the inspectio'n program should be considered.

With respect to the downgrading of proposed violations at Comanche Peak, I conclude that there was an effort on the l

part of T.F. Westerman and, possibly to a lesser extent, E.H. Johnson to limit the number of violations assessed d

against the applicant. While the reason for this effort has not been determined precisely, a possible explanation may be

-the fact.that Westerman was involved with the earlier Region -

IV inspection program at Comanche Peak. ,Thus, there might have been concern by Westerman that the identification of 5 .

deficiencies in the applicant's QA program would. reflect e - ,

poorly on him. Other possible explanations are that Westerman and Johnson may not see a need for stringent

, enforcement of QA requirements or may not be fully aware of them. In any event, there is sufficient evidence to support 1

th'e allegation that several violations were downgraded without good cause in inspection reports, such downgrading l

1 of violations without good cause is, in my opinion, contrary to the guidance provided in I&E Manual Chapter 0215 concerning inspector objectivity.

I did not find any evidence that this downgrading of -

~

proposed violations had an adverse effect on plant safety.

However, the questionable certification of BISCO electrical ~ ~ '

I

l 1

... \

)

35

~

l penetration seals is a serious matter and should be resolved as soon as possible. Essentially, I consider the improper-downgrading of those proposed violations that I have '1 reviewed to be a case of poor management in Region IV.

Regarding QA requirements, I recommend'that the regions l.

be provided additional guidance as to the need for stringent j enforcement of those require' ments. Any disagreement between inspectors and regional management should be resolved by I&E i headquarters OA personnel. 'In this way, the enforcement of .

QA requirements may be more consistent throughout all of the  !

regions. l

! l My review has led me to the conclusion that some of the NRC Form 766 data for Comanche Peak overestimate the extent  !

5  !

of completion of the Region IV inspection program. Until '

i the data are reviewed and corrected,'I recommend that they i not be used in any licensing decision. Further, guidance should be provided to the regions as to the proper method of completing this form so that the data will be useful.

As indicated earlier, I do not have sufficient information to reach a conclusion regarding the allegation of improper action in handling an FOIA request.

With respect to the Fort St. Vrain concerns, it does appear that Region IV has been lenient in its enforcement  :

policy for that plant. As there is no evidence of a safety problem, however, it may be possible to resolve this matter

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ . .__ .___.____m

. 36 by discussions betwee'n NRC headquarters and Region IV management, -

jks a final item, I believe that H.S. Phillips~and

,should be commended for their efforts in identifying, and attempting to correct, weaknesses in the Region IV ~

inspection program for Comanche Peak. Their actions obviously have irritated some individuals in Region IV.

l Nevertheless, to ensure that mistakes can be corrected and l prevented in the future, ki'C personnel must be encouraged to raise questions concerning the adequacy of past NRC i

inspection efforts. Along this line, it would have been beneficial in resolving these concerns if H.S. Phillips had -

considered it appropriate to submit a memorandum expressing l bis differing professional opinion.

f O

e l

m

(

._.____________U