ML20205Q731

Forwards Original Edited marked-up Sser 3 Revised on 841106, for First Mgt Review

ML20205Q731
Person / Time
Issue date:	11/06/1984
From:	Matthews P NRC
To:	Oliu W NRC OFFICE OF ADMINISTRATION (ADM)
Shared Package
ML20204J134	List: ML20204J131 ML20205Q707 ML20205Q712 ML20205Q715 ML20205Q720 ML20205Q726 ML20205Q731 ML20205Q738 ML20205Q742 ML20205Q747 ML20205Q755 ML20205Q759 ML20205Q763 ML20205Q769 ML20205Q774 ML20205Q779 ML20205Q785 ML20205Q791 ML20205Q795 ML20205Q801 ML20205Q809 ML20205Q816
References
FOIA-85-59 NUDOCS 8606030036
Download: ML20205Q731 (113)

Text

.'

5 ip

~

sq tw/t o /, m ppm a

~

osu'3-w x GOOV -

Ge's & en c,w e&M d- w '

of sst w-3 +d yw +%\ uhb .

ceA iu(  !

i% W BS T. 4 ,Io < E 4.c_.e c. 6M

\r M

&c \ %,4a n . e, -

nyg

)

~

M 191- +44 3 WWe>

efstw%m/

' ~

FCIA-85-59 L

(OS .

8606030036 860D22 ,

DE 9 PDR G

. SSER WRITEUP DOCUMENT CONTROL / ROUTE SHEET Allegation ' Numbers kOO M s 3d /.1 .

Subject of Allegajion . C ca.T. w.Ja. 7 c h 6ht u TRT Group F2at was 3 *

\ \

Author: uJRyh T

. This sheet will be initialed by each reviewer. It stays with all revisions to the SSER writeup and serves as a routing and review record. It will be filed in'the work package when the writeup is published.

Draft Number Draft 1 2 3 4 5 I -

Auth Yd.,V 2Pd .

Group Leader 4WF if/i4APls _ . . , , . . .~ ~.. , . . , . _ _ . . , _ . . _ .  :

Tech. Editor b1R t)/b '/ '

Wessman/Vietti -

J. Gagliardo T. Ippolito l

Revision Number _ , .-

Final 1 2 3 4 5 .

Author .-

Tech. Editor . . . . ___S E- -

Group Leader --

J. Gaoliardo T. Ippolito i Administrative -

l Writeup integrated into SSER Potential Violations to Region .IV Workpackage File Complete Workpackage Returned to Group Leader Y 19 9

h i

_y--..- . . - - . - . - . - . - - , - - - -

!s '

l

. , i Document N ae: -

CP S KR COATINGS 3 __.

Requestor's 10: r.

DEEQB01 z lf Author's Hatre: _

, William C. Wells _

.! Document Comrr.ents: _

! ,! Draft 3 - 10/19/84 Typed by Ordar. 10/23/84 i 4

I

I

.I i

,le

!t r

e e

's h

=

~

4 l

l 4

49- *,

I

~

, . . .w -; .:a . .. .:- .

= ... . .

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

l

~

i SSER-C0ATINGS 3 DRAFT 3 - 19 Oct. 84 .

WILLIAM C. WELLS '

l l

I

1. A11egaticn Group: Coating 3 raceability

~

2. Allegation llumbers: Parts of Q0-34,l,AQ0-36, and AQ0-62.

3. Characterization: It is alleged that:

- W -

The requ rements of American National Standards Institute (ANSI) Standard i

j N45.2.2 - 1978 were not met for coating material storag AQO-34)

/.s- '

The traceability of coating materials was not always maintaine r. (AQO-36)

.. _ A _ _ _ ,

}

~

Some paint used at Comanche Peak Steam Electric Station (CPSES) in Service Level I areas was contaminated with grease and of (AQ0-62)

' g.,4

4. Assessment of Safety Significance: The implied eefe+ um.arn of these f

allegations is that coating materials used inside of the primary containment

, buildin t CPSES which are not traceable to original qualification tested j*

~-

materials are not qualified. Non-qualified coatings could fail.during or following a pos44Me design basis accident (DBA) and interfere with the

! l i

SSER-C0ATINGS 3 1 COMANCHE PEAK 4 , tA h' , "g*g 8 g ,

e b  % =, .  %

furctioning of engineered safegbard systems. An evaluation of the

,l i

j sisjnificance of this safety concern is reported in (LATER). The Technical Review Team (TRT) evaluation of the Quality Assurance aspects of these -

alli-egations, as characterized above, h di::::: d b:1cm fhN * *

' hjr'0F MAJE iMB. T.RAtTA'BihTf5YSJfhV\

0)Jf' AW

- L- w4 QTTheTRTperformed-an-e/m:tien-cithecoatingsmaterialcontroland -

traceability syster. as ;Jenti ied beivo usud E7 Texa3-Utiiities siectric Conp:ny (T'JEC) at CPSEfr. The purpc:c of-th:2 .evied @ to determine.the- QQ MAN

/ -

adequacy of the system emnley.Riba%G-to-assure ims ia.or.stete that -

coating materials .hich had been applied at CPSES were traceable to batches D

of-coa-ting-msi.c. iais which had passed e. !;k;l qualification tests. An adequate traceability system must involve: (1) mani facture of the material under a manufacturer's Quality Assurance Prohram; (E) certification by the manufacturer that the batch manufactured is essentially identical to the batch i

which was qualification tested; (3) receipt, control, storage, and use of the .

i material by TUEC under a system that assures that the materials are handled and used, as reccmmended by the manufacturer, in a r:anner essentially

!! identical to the manner in which the originally tested batch was used; i

(l j (4) implementation of an inspection and doct; mentation system which provides l ll records to demonstrate all of the above.

t The TRT identified the following procedures which includes instructions for handling, storage, use, and control of' coating materials:

SSER-COATINGS 3 2 COMANCHE PEAK

~ =. .: .a . . , 3 : : P n> -" = ~

.: ,, .Y -

J .-

s: ,

TUGC0 Procedure CP-QP-8.0 and Brown & Root Procedure CP-QP-8.1 provide instructions for receipt inspections at CPSE8'for all incoming items, including coating materials. Brown & Root procedure CP-CPM 8.1 provides -

j instructions for receipt, storage, and issuance of items, including coating, 'l materials. Brown & Root procedures CCP-30, CCP-30A, and CCP-40 which provide overall instructions-for performing coating work at CPSES, include specific instructions for storage and dispensing of coating material-s and provide 1

~

storage temperature limitations. TUGC0 Procedure QI-QP-11.4.-17 provides -

instructions for monthly surveillance by QC of coating material storage.

TUGC0 Procedure OI-QP-11.4-22 provides specific quality control measures which govern the transfer of protective coatings identification numbers when cutting coated stock materials. TUGC0' Procedures QI-QP-11.4-1, QI-QP-11.4-5, and QI-QP-11.4-10, which provide overall instructions for inspecting coating work at CPSES, include provisions for inspecting coating materials just prior to mixing, and during mixing and' application.

The TRT observed that, although most of these procedures had undefgone numerous revisions during the construction phase at CPSES, the overall system for coating material storage, handling, and control had remained substantially

! the same throughout. ,

-i.

~

The normal site routine for handling coating materials which is described by the procedures, and which was observed by the TRT, is as follows: Incoming coating materials are received and receipt inspected at the CPSES site warehouse area by site receiving [Inspectorsinaccordancewithstandard l

i SSER-COATINGS 3 3 COMANCHE PEAK l

~ "N

~

f.'_

. Mu9lUdh.D U? NO: _ . - .- . .

- - - - ~ -- -- - - - -

site receiving routines that apply to all incom g Q -n)1'UN litf items. The

\

i materials are then issued to the coating storage area near the paint shop.

The coating storage area has separate storage for "Q" and "Non-Q" paint -

materials, is temperature controlled, and is equipped with recording ,

thermometers. The coating storage area and storage functions are operated by coatings craft pert inel (not QC), and are subject to monthly surveillance by coatings Q [nspectors. When coating materials are needeg fee-ttsithey are l i xed in the coating storage area, then transp,orted to the a0t $k '

, j use.Mttwgediit pt4am or transported to a mixing area _,near or at-the use-location, A coatingsQN7/nspectorverifiestheacceptabilityofthematerial V

immediately prior to mixing and observes the mixing and application process to verify conformance with requir ments. In those cases where mixing is not

~

i M Mlm performed 3L )T use-lecat{icii, W W *t-l.

at the the i ..i /iQ C. ' who inspects the mixing process attaches a mixing record to the material container to transfer mixing inspection inforination to the inspector at the% "~ ' " '-- A*

A To assess TUEC's implementation of this system, the TRT randomly selected AC approximately-twenty CPSES coating work inspection reports written from 1977 through 1984. The TRT noted that batch numbers.of coatings materials used were recorded on these inspection reports, although in some cases there were other deficiencies in these inspection' reports which rendered traceability indeterminate. For example, Inspection Report PC0127 documents application of coatings to an area of Unit #1 Reactor Buildin ( esumablylinerplate)y on June 27, 1978, and has an attached mixing recor (alsoidentifiedas PC01279 which documents mixing of coating materials for this area on SSER-COATINGS 3 4 COMANCHE PEAK

'~

7 $ ,

.s. a . 1. t. .,r >

R - JP~ _ .

~ --~ ~

-y- e. - - - - -.-- w ,--- ,- -- - ,--u. --. - - - - - - -

~

i .

5 .

.t -

September 8, 1978. A similar situation exists in Inspect'on Report PC03650y

!! which documents surface preparation and primer' application on December 1, 1977, of ' coating inaterials which were mixed on November 22, 1977. ,In both of these ' cases, the TRT observed that it was physical.ly impossible for these ,

materials to have been applied as recorded in the inspection reports due to the time discrepancies between mixing and application. .(Materials cannot be applied before they are mixed, and can only be a ring a limited "potlife" after mixing before they harden [approximately eight hours for most materialsusedatCPSES.) The TRT also observed many other deficiencies in theinspectionreportsp;vica[includinginadequatedescriptionorlocation

~

of areas or items coated, improper changes and corrections, lack of signatures or acceptance initials at interim and final inspection points, and lack of date/ time correlation between various sections of the reports. In many of these cases, the TRT observed, the deficiencies noted were serious enough to I render the inspection reports unacceptable as quality records and inadequate to provide documentation of material traceability. The TRT noted that some i deficiencies in coatings records for the dperiod prior. to November 1981 ,

'l' .

were identified previously in USNRC Region IV Inspection Report 50-445/81-15; 4

50-446/81-15 and TUGC0 QA Audit Report TCP-24, October 2,1981. The TRT also noted that traceability deficiencies for that -t period have been acknowledged by TUEC in correspondence to the TR W ten-ced in % gi Q belewr-SSER-C0ATINGS 3 5 COMANCHE PEAK

-g -

-fr. . ,. ,. m m . ,--- m - ,.. . -

~

~

i l

3G l -

The TRT next selected approximately tM+6y batch numbers of coating materials [

taken from the inspection report cussed-ebeve and from certain nonconformance reportsg i m 4 .' n . The TRT traced these batches back to -

, receivinginspectionreports(RI[))whichdocumenttheirinitialreceiptat I CPSES. The TRT noted during this process tiiat TUEC had an efficient

. computerized tracking system to allow identification and retrieval of both coating work inspection reports and receiving inspection reports by coating material batch numbers. The TRT found that, in general, the receiving records were in order, and the receiving records packages contained all pertinent information, including copies of coating manufacturers' product identity certificationrecordgasrequiredbyANSIN101.4andANSIN101.2. However, ob\ -

the TRT could not locate s.RIR or manufacturer's certification for batch number 0M1708M of Carboline Carbo Zinc 1 which was' identified in CPSES NCR

1. C-81-01724 The TRT also examined the receiving records for the first shipments of coating materials from Carboline, Ameron, and Imperial, the 9

manufacturers of Service Level I steel and concrete coating materials. These RIR and attachments include evidence of' receipt inspections and check 1.ists, manufact0rers' certifications, invoices, evidence of source surveillance by Brown & Root or waiver of same, initial storage location at CPSES, and notation of the governing TOEC purchase, order.

The TRT next reviewed vendor files for Ameron, Carboline, and Imperial. These

, files included evidence of 'nitial evaluation and approval of these vendors by Brown & Root based upo . Q.A. Department Survey Checklists completed by the t' ,

SSER-C0ATINGS 3 6 COMANCHE PEAK

=

g, *] ?.* p , P- ' "T 'P '

N vendors, review and approval of the vendors' Q.A. Program Manuals by Brown &

Root, and facilities surveys by Brown & Root.These files also included records o'f regular au'dits of these manufacturers by Brown & Root fr.om 1977 -

through 1981. Although these audits occasionally included findings of -

deficient conditions, documentation was included which demonstrated corrective actions and verification of acceptability via reaudit. The TRT found that regular audits since 1981 have been performed by TUEC rather than Brown &

'- Root. -

The vendor files also contained copies of the original purchase orders to these vendors together with subsequent changes. The TRT found that'these purchase orders are generally complete and acceptable and include the imposition upon the vendors by TUEC of appropriate regulatory requirements and industry standards (e.g.10 CFR 50 Appendix B, ANSI, N45.2, ANSI N101.4, ANSI

~

N101.2, and 10 CFR 21.) The vendor files also included general correspondence sections which showed evidence that there had been vendor participation in a pre-job meeting)as required by ANSI N101.4, and in one case showed evidence that the manufacturer had transmitted reports of in-house laboratory qualification tests to TUEC. However, the TRT observed that these files did not include complete qualification testing data nor evidence of a review and evaluation of that data by TUEC; this subject is discussed further in7AA.

SS Coatings 2 DB A CW38 c,b cmT~e_*h .

SSER-C0ATINGS 3 7 COMANCHE PEAK

, s. -

l l'

f

! The TRT found that the system and its implementation by TUEC described above included most of the important measures required to control materials and to maintain traceability. However, the TRT observed certain deficiencies in the procedurally described system, and in practices which are not procedurally , -

addressed that could allow traceability to be lostgjs fulives.e

)

Coating storage areas are under the management of craft production personnel (access to stored materials is not controlled independently by personnel) 6 - .

and('Q" and "non-Q" coating materials are identified only by their separate storage areas. The result of this arrangement is that once a container of O m d e dty unmixed material is loaded onto a truck for transport te a "se zee, it is no longer possible to determine whether the material is "Q" or "Non-k The inspection of the containers 'Mch # performed immediately prior to mixing would not be able to detect such a mixup in those cases where there is no W14-evidenTg difference in the container labels, %"Q" materials which had earlier been moved to the "Non-Q" storage area due to a problem with bofAh)4.h% -

. .. certification or prev ous loss of control. (Note that no tagging or label

- A .

change..is required when'this occurs.) In a similar manner, there is no independent control by rsonnel to assure that unacceptable materials

returned to storage from a

sse location are, in fact, returned to "Non-h h rather than "Q" storage, or to dhierdiar prevent the transfer of materials between d,jacent "Q" and "Non-Q" storage areas.

~

l 4

l l

SSER-COATINGS 3 8 COMANCHE PEAK

's -_ -.. mr77~ar, .

- r

~~ -

c

~

The storage area temperature record charts which are attached to the man'thly

{

storageQ7C nspection/eports are not identified as to recorder location (i.e. to'show whe'ther the record documents the temperature of the "Q" or "Non-Q" storage area, or some other location.) ,

l Some paint mixing slips, described above, are not retained.as permanent records but are discarded after the inspector in the coating

! . transcribes the information to his own report. Thus, the original record of

~ ~

the mixing inspections, including the inspector's signature, is lost.

l There is no clearly ssigned responsibility for maintaining inventory and i storage records for certain materials after transfer from the site warehouse

.[ area to the coating storage area, and the monthly /nspectionR'eportsdo j i

not identify what materials are in storage. Thus, the official records do not l

l show, except by inference, what materials were actu' ally in the storage.. area during a given time period, nor is there any record to tie the monthly

.! surveillance reports and storage temperature record to any specific batch of l

'] material. -

i i

There is no procedurally equired control of coatings materials which have wah

.{ been moved from the coating storage area to variou's' ose or mixing areas in the 1 ^

plant. Thus, coatings may in fact be kept in such locations for extended periods of time without temperature control or records, or Cfsurveillance.

4

!! SSER-COATINGS 3 9 COMANCHE PEAK

=

v'E ' .

~*

_ 4 ]M,T ." * [ l

~

The TRT did not find any specific evidence that th ficiencies S n. bed

% resulted in any general loss of material control or traceability (after 1981),exce as specifically indicated herein. However, it is evident based

%A on tm hat a more rigorous system of coatings material control is b -

necessary to provide confidence that materials do not lose traceability through confusion, alteration, loss of status control, or exposure to environ'nental temperatures outside of the specified range, during storage and handling on site. -

r f,RtVIET0F I4(tWitU ALLEGAT10Sv~l Th hvestigat d the sp e m -o whi k e-de G bed' $ the allegatio s ractl. zed above.

R[QUIREMENTS OF ANSI N45.2.2 N6T MET (AQO-34) i f '

bqo-vt -

Tiffs allegation concerns control of the storage and handling of coating

. G A materials. It is alleged that the requirements of ANSI N45.2.2 were not met for the storage and handling of coating materials.

~1n aArann }]CJ$,35IWj '

JheTRTrek;tewedtheCPSESFSAR,ANSIN45.2.2,andcorrespondencefromT the TRT on this subject.

,s-SSER-COATINGS 3 10 COMANCHE PEAK

g. ., .

p in. ivunu wac

-Q he CPSES FSAR, Section IA (B)-16 specifically states that the requirements of NRC Regulatory Guide 1.38,' and thus, AliS1 N45.2.2, will

~ ~

g apply during 'the construction phase.

n The TRT noted that altern. ate methods -

for controlling the storage of coating materials are provided through CPSES -

FSAR commitments to compliance with ANSI N101.4 and related standards, and are detailed in CPSES Specification 2323-AS-31 and pertinent CPSES proceduresg)  :

r aie. cuucd Leiuri. The adequacy of th se alternate procedural requirements and anx O '

their implementation 6-dinn:i chtch;s. in thf a repc.ct..

~ A & MW Q O -3(a n TfACEABILITYNfTA)'WAYSM/INTAINED(AQO-36) g.

It(is aMe F n d iways

~

ntpedr yq ..__... . - - . .

~'

ft -

. A40-3% gf s Because of the general nature of t W allegation the TRT ce d et9 ==

9 irve::t#:;a+4 an ^L the CPSES system for assuring coating materials traceability p andthemethodsusedforstoringandhandlingcoatingmaterials[todetermine whether the system and methods could allow traceability to be lost.-%e_

I data 44s-of-thel lnve>Useliva and--u.e esists are-. prow 4ded-under arctic.. of 44aterial. Control-and-Traceabil+ty-System =,Laboves i

.I The TRT reviewed CPSES nonconformance reports (NCRs) pertaining to coating, and k selected four which bear on the subject of traceability.; Wea-*'--these NrDL pe r ._, ..

, _. , n g-i.u c v Xwnisuucar un cuc snuscu s vi.-

e,. m ac u ivui .. .

-treceshiUty. hree of those NCRs are discussed under the review of AQ0-62, SSER-C0ATINGS 3 11 COMANCHE PEAK l_

3

.~ , - - ,- w --,---

)(d"d(,ig.pl S elow. The fourth, NCR #C-83-02938, October 27, 1983, directly concerns loss l l

of traceability due to the transfer of coating materials from a mixing area to application locations, without maintaining identification of the material -

l . during transport. The TRT's review of this NCR and associated inspection ,

~

lreportsrevealedthatthenonconformingconditioninvolvedsmallquantitiesof

~

/ material that were mixed at one time then divided into different containers to

, be used for touch up and other coating work on a number of small areas. The -

i NCR disposition section described a change to the method for mixing and -

I i distribution to require identification tags to be placed on each container when material was divided after mixing.

] -

I The TRT noted that this NCR concerned a relative isolated incident w I a l involved -twelFvs inspection reports over r

~

(

,three-da eriod October 9-31, /

.9 1983,andthatthecoatingsinquestionwereusedonlyforas#f....t,erof smail arcas. The TRT did not find any other' examples of NCRs concerning loss

+

of treceability of coating materials due to this same cause and because i p:easures were instituted to prevent recurrence, the TRT concluded that this particular incident did not indicate any more general loss of traceability.

The TRT next reviewed TUEC's correspondence on this subject. TUEC letter TXX-4201, June 22, 1984, under the heading " Allegation No. 36... Evaluation of l Validity" states in part: " Traceability does not exist in all cases for  :

coatings applied prior.to November 1981. These coatings, however, are within l ,

the scope of the backfit inspection programs which determines their

, adequacy." The TRT reviewed USNRC Region IV Inspection Report 59-445/81-15; SSER-C0ATINGS 3 12 COMANCHE PEAV,

~. d W

.s- .

,, / 50-446/81-15 and related correspondence, and TUGC0 QA Audit Report TCP-24, which is referenced therein. These documents ' reported deficiencies in records l

'. of coating work performed at CPSES prior to November,1981. The backfit -

d

} in'spection program was established by TUEC for the purpose of determining the ,

adequacy of applied coatings 'for which existing records were deficient. The TRT observed that backfit inspection reports did not record coating material  ;

l i

' I batch numbers or reference any other records where this information was ll l t l recorded. Further discussion and assessment of the CPSES backfit program is ,

1 BA No - -

provided in SSE Coatings 1.

O A ,

, j The TRT determined that the backfit inspection program and theb 'ackfit inspection records do not restore traceability of applied coatings for wh'ich  !

i' , deficient previous records cannot provide traceability. The backfit i

g inspections determined the adhesion and thickness of the applied coatings at ,

I the time those inspections were performed. These t'wo physical attributes do '

o j not demonstrate that the applied coating materials are certified nuclear-grade -

j materials which have been manufactured, stored, mixed, and applied without i

l s alteration or contamination, and do not demonstrate that these materials are j i i k 3 4 i qualified.

s,

- - - - - . . . . .. b ,

_AMLICATION OF C$NTAMINATED P/T5T (AQO-62)' l I -

Q i dMs , allegation $0~$alteration or contamination of coating materials concerns j l

prior to application. Such alteration or contamination renders a coating {

~

material unqualified and untraceable in that it is no longer essentially l

SSER-C0ATINGS 3 13 COMANCHE PEAK l

4 identical to the batch which vras subject-to original qualification testing.

It is alleged that some paint used at CPSES in Service Level I areas was contaminated with grease or oil, prior to application, and was applied anyway. .'

The TRT examined three separate instances of possible contamination of coating ,

materials. All three occurrences were documented in nonconformance reports.

2 4

j NCR C-1729, October 2,1979 concerned  :% nU)&TM3 se nlonconforming A condition

f t

. / -

j N " Carbo-Zinc 11 base-batch 9H5381M was noted for centaining what is '

believed to be grease particles and/or foreign contaminants. Discrepant Qty - -

300 gal./5 containers." The dispositon of the original version' of thi: NCR

( ~-

! was "use as is" (after straining to remove all contaminants This t disposition was based upon a telex from the manufacturer, Carboline, which was attached to the NCR. Revision 1 of this NCF issued on October 17,1979t7 f

7 chang the disposition to " Return to Vendor,'" due to the fact that the straining process had been unsuccessful in remcving the cor.taminants. To '

determine whether any of this contaminated material had been " applied anywah  !

as alleged, particularly in the time frame between the original issue of the ,

NCR and Revision 1, the TP.T examined 44 oating nspection records for j work done during September, October, and November 1979. The TRT found no evidence that the contaminated n:aterial, batch #9H5381M, was ever applied.

-~

4 The TRT noted that the Telex from Carboline, which was provided as i

justification for the strain-and-use-as-is disposition, did not provide any tect.nical explanation by Carboline to indicate why Carboline felt that the 4

s

t. SSER-COATINGS 3 14 COMANCHE PEAK l .,

. , ..+ -

>- = .

~

presence of the particles would have no adverse effect upon the material, nor any indication that Carboline had reviewed ths NCR or was aware that the particles had been identified as " believed to be ' grease'." .

NCR C41-01724, ecember 14, 1981, and revision 1 of the same NCR, also dated December

• 14, 1981 docu. merit indeterminate conditions.in bate OM2708M, OL2531M, and N2791[of Fhenoline 305 Catalyst, Part B. Batch #0H2708M was identified as being " dark wine colcr, rather than the normal amber color, and

~ ^

all three batches had unknown particles jn the material. This nonconformance report was di,spositioned "use as is" based upon connunications from Carboline, which were attached to the NCR. The TRT reviewed the information attached to this NCR and found a Situation similar to that discussed above for NCR

~

1. C-1729, in th the CarLoline statements did not provide adequate justification or assurance that the identified deficiencies would not have adverse co.nsequences. Carboline's statements did n'ot discuss the nature 6f the particles but indicated that these particles probably got into the d y[

material in their plant. Ii1aTelexdatedDecember 31, 1981) Carboline recommended that the batches not be used until they had determined the cause of the color difference. This reconnendation was not included in their letter .

of January 19, 1982, althcugh there was no indication therMW that the cause had been determined.

1 l

9,/ -

f NCR #C-81-01673, datw, Decen.ber 6,1981, documents a condition observed during  !

l mixing cf Carboine Phenoline 305, base batch #1J2789M, curing agent batch ,

i l

SSER-COATINGS 3 15 COMANCHE PEAK n -r -

r. .

,. . .m~.. - - -

~

=-

~ .

~

1. 1F1054M, and thinner batch #1E1861M. The NCR reports that after repeated hand and power agitation the materials would not thoroughly blend, and dark greenpigmentfloatedtothetop6heNRCalsonotesthatthesematerialshad .

already been used to coat nine electrical supports. This NCR was

• dispositioned "use-as-is" based upon a visual examination of the nine ern electrical supports .which revealed no color abnonnalities,.and,a statement by l the engineer who wrote the dispositon that.such " pigment float" sometimes i ~

1 occurs but will not affect the cure or integrity of the coating. The -

~

disposition further requires that an agitated pot shall be used during application if pigment float is noted.

Upon reviewing this NCR and its attachments, the TRT found no clear explanation of the cause of the observed condition, other than the engineer's statement that pigment float somettres occurr,, especially when higher levels of thinner are used, nor any justification f'er the statement that this condition will not affect coating cure or integrity. The TRT noted that the inspection of the electrical supports to which the coating had been applied provided some confidence that the cure and integrity was not affected.

Nonetheless, it is the TRT's olinion i that a more rigorous investigation of

l. possible centamination or alteration, and communication with the manufacturer on this subject, would have been appro'priate prior to.the issuance of a use-as-is disposition, Nuclea ertified coating materials are expected to appear and behave uniformly without variation from batch to batch or container to container; any SSFR-COATINGS 3 _ 16 COMANCHE PEAK

,i . -

5

' . ll -

l * , i ,_ < _

_ ~_

observed differences or irregularities such as those documented in these K should be thoroughly investigated to assure th'at traceability to the original '

-) qualifica' tion tes'ted batch has not been lost through deficiencies 1-n manufacturing, handling, or use. ' '

5. Conclusions and. Staff Positions: .

t 1

ased upon the reviews of individual allegations described above, the TRT concludes that:

'A S Q'4-27 W W g , ' (hddBAN' g "00 This glegation W noncomplianc'd with ANSI N45.2.2 is 4rr&;=t-

[mpliance with that standard is not required by the CPSES FS R J ^# D e 3 t c4dCIdDE .1 M YuHit\  % ~h> 2 ons.;;;t

\- Therefore, this allegation -intit%"t "fer:: cc- c;"creer, safetyQs gq%g g yws significance; ce generic implications, j9.ceW emw-

-b A00 25 % allegation

$0 -%

that the traceability of coatings materials was not

)

- A always maintained,is correct. The primary significance of this allegation is L

that, at present,) applied [coatinos!in areas included in the backfit program j are of indeterminate traceability. -

)

e AQ*3 - bb W allegation that coating materials which were contaminated witn c- A grease were applied at CPSES has not been supported by the TRT's investiga--

tion. In the one instance,-disettssed :bov$ewhere " grease" contamination was observed, the material was returned 'ot the vendor and the TRT found no

~

SSER-C0ATINGS 3 17 COMANCHE PEAK

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

l

.~ -

evidence to in'dicate that any f this contaminated material was applied, i GnTk12aad/Judub Therefore this alfegation 2: ta+ad. is withn'_'t Micr:: cenmm..m ..-

ety sig ificance P W G ,

4 -

However, in the investigation of the subject of contaminated materials in s whichfera'had b . ..

general, the TRT found instances of coatin ""' ;;h had r unknown contaminants or which exhibited other unusual and -

^ v unexplained properties. The TRT found that in these other cases, the [

operation and documentation of the Nonconformance Reporting System at CPSES was inadequate to demonstrate the cause of the irregularities, to provide adequate technical justification for use-as-is dispositions for the deficient materials, or to address and correct whatever fault in the manufacturing, handling, or materials control processes produced the deficient materials.

e$

n' A01t M & h w wud

• 3* h aleM 9

nned u ,on the gena :c myt: ef TUEC's material control and traceability I WVA& WD

}, systeyescribcd-:be/O the' TRT con udes that:

Pr -

l l

The material control system, as it is implemented currently at CPSES, provides I general ontrol and assurance of traceability for most coating materials, but is not adequately rigorous to assure traceability for all coating materials.

The primary area of concern, as discus' sed above, is that materials storage and l issuance routines do not, include the necessary controls to assure segregation of "Q" and "Non-Q" materials and to prevent the use in Service ' Level I areas I of materials designated "Non-h The TRT was not able to identify specific incidents wherein this ' lack of rigorous control had resulted in misuse of l

SSER-C0ATINGS 3 18 COMANCHE PEAK Y 4 .[ f 'a '8-'"

• l 1

~

materials; however, procedural changes.are required to prevent such incidents in the future. The subject of procedures ~and ' required changes is discussed

<L ~thi further% in SSER Coatings 4.

p- h The TRT determined that the material control system was not adequately implemented prior to_ November 1981. USNRC Region IV Inspection Report 50-445/81-15; 50-446/81-15 has identified violations concerning inadequate -

inspection and documentation practices for coating work prior to November 1981, which resulted in loss of material traceability. The TRT has determined that the corrective action performed by TUEC to date (i.e. the Backfit Inspection Program) has not restored material traceability. As' detailed I below, i.t will be necessary for TUEC to investigate this area further in order to quantify the extent of the problems and to permit an asses; ment of its significance.

6. Actions Required-ecause of the deficiencies in costing work inspection records and material control records noted above, and because of TUEC's correspondence to the TRT which stated, " Traceability does not exist in all cases for coatings

~

applied prior to November 1981," TUEC shall conduct an investigation to identify the amount and location of surface areas for which material traceability does not exist, and shall provide this information to the NRC.

l l

SSER-C0ATINGS 3 19 COMANCHE PEAK m -- ==. n,,- . r ., , -< m %... .r------

5~. _ _

l l .

l

' This information will be used by the NRC in its assessment of the Backfit Test Program.. ,

W 9r) CPSES Nonconformance Reports (NCRs) #C-81-01724 and #C-81-01673 provide t

"use-as-is" dispositions for discrepant coating materials with inadequate technical justificat. ion for the disposition. Accordingly,.TUEC shall provide adequate technical justification to demonstrate the acceptability of the -

batches of coating materials listed in these NCRs or, alternately, TUEC shall identify and quantify the areas where these batches were used and place these areas in the coatings exempt log. Additionally, TUEC shall review all other NCRs which concern discrepant or irregular conditions in coating materials.

For any such NCRs which were dispositioned "use-as-is TUEC shall identify the batches and provide adequate technical justification for their acceptance, .

1 or identify and quantify the areas where the batches were used and place these j l

areas on the coatings exempt log.

8. Attachments: None. .

9. Reference Documents: ,

l 1

W/ Brown & Root, Inc., Procedure CP-QP-5.1, Rev. 2, September 12, 1979, l

" Field Procurement Activiti SSER-C0ATINGS 3 20 COMANCHE PEAK 1

e o

=.

• e ,

. 1,

t. .

. .  : 'i ,

.t

...w s

- . . g g. _

'%as., . * --

',2 -f , ; . .' . , .

,-(

, '; ~

s g

-. } . ..

x..

, i

.m. w .

Ch

• r  :

-m - -g-,

1,

.< c. . t, .- . , , ,

A a-p  ? - . *. . ,

4 h "' /. ,' . } V' .

' < s

.'=4* .' 4

_-s, .. .),.

' ,,4 ,- . . [ .. . *9* ,- *t ,

,1 - -- , . ..-, - - - 's

- [ .t

7 .

^*

. y

• * * * 9'

,.. Q . ,' ,J 'j*- ;*h . f, k' . +*..'s -

,~ 'O. .'69 -

e - . *

. ** e

. y .

, y y .. . , . , .. .w .- - *

. fy -' - 9 , , p,

+. - . * .',j . . .- .'g - ,, ) .,,- / ,- ,,,s.

.. .- .ow, . -

. . . b. , p

.' . . * .N * '*

' . ",m .%.t ^ '.*. , .- j 1'J**- I'r

*-.. ' ' " -' ' \ '

q N T'

y' 's, . x. .

. . . . . . . . . . ... ) _ t

• . j .  %

<.,T. . ,. . , .

. ^. .

• 4,

. s

. . .j b. A*' go. ..'

' ~

A - **

• N' [

S ,

'* ',  %  ?. N 't ,, _.

t Q M)h.: , h. .,.*,D , f:,r ' Q. +' x lj,I '; x 2.<: - 4 c. . .) .;Iwy. A, M ; % A ',. , b. .,-'s-M%. v, ; .. ...

e T-4 '. .

, .. _. ._ . . . . . . . . . . . . . . . . . . _. . . l

. . . . . . . . . . . . . . . . . . . .. ... . - .. \

I 1

l 9

. .. . .e.

1. .AAW""~

M &

~

. m n _ _ _ _

,  ; -. . . - - - - _ ,,.:.~-_ . .,_ ,. _

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

e * *

. . )

l M ,TUGC0 Procedure CP-QP-5.0, Rev. 8', June 1, 1984,." Procedure for Field Procurement"

% 3, Brown & Root, Inc., Procedure CP-QAP-8.1, Rev. 8. June 11, 1984, * -

" Receiving Inspection"

@i.TUGC0 Procedure CP-QP-5.0, Rev.1. April 20,1981, " Quality Assurance Review of Site Generated Procurement Documents" . . .

y TUGC0 Procedure CP-QP-8.0, Rev. 4, August 9, 1984, " Receiving Inspections" g )d.TUGC0 Procedure CP-QP-8.1, Rev. 1, January 28, 1980, " Supplier Evaluation" '

y/,TUGC0 Procedure CP-QP-11.4, Rev. 7, November 8,1983, Inspection of Protective Coatings" '

WgTUGC0 Procedure QI-QP-11.4-5, Rev. 29, May 4,1984, " Inspection of Steel Substrate Primer Repair and Seal and Finish Coat Application and Repair" M TUGC0 Procedure QI-QP-11.4-10, Rev'. 19, June 13, 1984, " Inspection of Concrete Substrate Surface Preparation and Coatings Application and Repair" l

) .

l l

.I SSER-COATINGS 3 21- COMANCHE PEAX l I

g:n y -  : z_ . .

- m N m *- .v-

- - . v-~~~

, , . . = .. _~- - : .._._--w-1- .

I t

M b'T'J GC0 Procedure QI-QP-11.4-17 Rev. 6. June 17,1983, " Surveillance of Storage and Handling of Protective Coatin'gse M TUGC0 Procedure QI-QP-11.4-22, Rev. 3 August 7, 1984, "Q.C. Verifi-cation of Protective Coatings Unique Identification Number Transfeg M Brown & Root, Inc. Procedure CCP-30, " Coating Steel Substrates Inside Reactor Buildings & Radiation Areas" (all revisions to date)O -

@ Brown & Root, Inc. Procedure CCP-40, " Protective Coating of Concrete Surfaces"(allrevisionstodateb- .

1. Brown & Root, Inc. Procedure CP-CPM 8.1, Rev. 1. February 24, 1982,

" Receipt, Storage,andIssuanceofItemp" 15 W TUEC correspondence to the NRC identified as TXX-4201 dated June 22, 1984 and TXX-4249 dated August 10, 1984. ,

l k,

@ ANSI N101.4, " Quality Assurance for Protective Coatings Applied to Nuclear Facilitieg ,

i' rl.

, y ANSI N45.2.2, " Packaging, Shipping, Receiving, Storage and Handling of I

ItemsforNuclearPowerPlants(DuringtheConstructionPhase)Q l

l l

I l SSER-COATINGS 3 22' COMANCHE PEAK l l

l. 7 --.n.r,.-,

,--+--i .;_------- -

~ - -- -

Fi;, . ..,;

.. '. .; .~

1 f. . -- .

1

. .'. .t .; * -

, w .;y. .

/ j.r) CPSES Specification 2323-AS-31, Rev. 2,' March 15, 1984, " Protective ,

Coatingy d7,5)n.USNRC Region IV Inspection Report 50-445/81-15; 50-446/81-15 document- -

ing an irtspection conducted at CPSES on October 13-16 and 19-23,1981g)

Q QRC, Pq h L Y 5 f\/l$ d & 50 N W 8~l- % 33.N h k ) ,.9 g lTUGC0 Q.A. Audit Report TCP-24, October 2, 1981, with associated -

4 findings, responses and evaluationsq) -

'rown &., Root, Inc. Procedure ACP-3, Rev. 8, April 18, 1980, " Material a

Receiving, Storage, and Handling

.~ )h . -

7 0

/ -

/

jpsh '

f i l

1 l

I

. 1 E

l

, SSER-COATINGS 3 23 COMANCHE FEAX ...

~

. . -.... . - ., ._.,........,,_....y mm, - .%.....~. _ . _ . . . . . . ,

l

!' Document Name:

' SSER TEST PROGRAM 1 f ,1 uesto'r's ID:

t .

Author's Name:

Chet Posiusny Document Comments:

11/8/84 Final draft, rev 1 W i + L ,' _

AT- b 2, b % s~,J 6 T)T)/0,l/ji3,/)?

e

-./ p .-

Y '

/ {/ (,,,

., ')

n~ ,

.1i n \-

o 'u  ;

l 1

KBA-85-59

Final Draft, Rev.1, 11/7/84 CP1/SSER Test Program SSER 1.

or Allegation Categortes.: Test Program No. I and 2 .

2. Allegation Numbers: AT-1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 13, 17

3. Characterization: It is alleged that the NRC Atomic Safety and Licensing Board (ASLB) can not make the required findings under 10 CFR 50.57(a) and TUEC can not be granted an operating license for Comanche Peak Steam . , ,

Electric Station (CPSES) Units 1 and 2 because of the following: (1) /

TUEC failed to conduct an adequate prefueling hot functional test (HFT) . ,

'J program, in that not all components or modifications were installed which - #

require hot functional testing; (2) TUEC did not intend to check some components and systems until heatup to hot standby or during power ..; M escalation; (3) TUEC and the NRC Region IV staff failed to notice this condition and did not keep the ASLB informed of tie problems encountered; (4) numerous problems were identified during the conduct of the thermal expansion test, as evidenced by Test Deficiency Repo-ts (TDRs) 853 and 855; (5) that the HFT was conducted without considerae. ion of accident conditions; (6) TUEC and the NRC Region IV staff were willing to accept deficient test results; and, (7) that the ASLB cannot rely on the NRC staff to monitor the test program or any reinspections. *

4. Assessment of Safety Significance: The implied significance of these allegations is that if the HFT program was improperly conducted, the adequacy of the plant to operate safely may not be assured.

i The NRC requires that a preoperational testing program on a nuclear I power plant be conducted to demonstrate that plant structures, systems  !

and components meet their safety-related design specifications, as stated in the utility's Final Safety Analysis Report (FSAR), before the plant aces into operation. The Technical Review Team (TRT) conducted indepen-dent reviews on 17 of 25 completed test records pertaining to HFT (which is a preoperational test) and interviewed cognizant TUEC personnel during l

Q(. (mon'?&

l the course of this review. Thereviewincludedfollow-upinspections[

t tsk @.cfqp$r4pfTDRs) that were generated as a res' ult of testing _

deficiencies found prior to and durina HFT. The TRT also reviewed pertinent Startup Administrative Procedures, NRC Inspection Reports, the pre-operational test index with schedule, and a system / subsystem turnover definition and target date index. The TRT reviewed this documentation' l against the FSAR and the applicable NRC requirements and guidance (10 CFR 50 and Regulatory Guide 1.68).

a. The TRT found that the HFT was conducted with several minor com-ponents and equipment not having been installed at the time of the l test and with modifications remaining to be completed after the test. . . . _ . . _ _ . . _ ._

There are system walkdown inspections conducted by Brown & Root l ,

j

• Quality Control, TUEC Startup, TUEC Operations and the NRC to verify

.~

1 ,

installation of components and supports. The TRT reviewed NRC g Construction Appraisal Inspection Report 50-445/83-18 (inspection j l JO //* conducted January 24, 1983 through February 4, 1983), Inspection ,

M}r Report 50-445/83-23 (conducted May 23, 1983 through June 10,1983)

,[ .

and Inspection Report 50-445/84-16 (conducted May 14, 1984 through June 20, 1984). None of these inspections indicated that a g

^

^

' g

,/ undocumented hangers and supports were found missing. Therefore, _

p-i 3 the TRT had no reason to question the adequacy or completeness of , .tg.1 t M e the documentation used to identify missing supports during HFT (or '

1

. e < ,y any other time).

1 i

sJ<

The TRT found that all equipment required by HFT procedures was

( either installed, or docume'n ted as not being installed on a Test-1 Denciscy--Repor.t (TDR) or. Test Procedure- Deviation (TPD). A TOR is j documentation of components and equipment which are found to be deficient or defective at the time of the test and for which sc.e action must be taken to correct the problems; a TPD is.an approved change or deviation from the procedure as originally written. TpRs

and TPCs become a part of the comp 1.eted test record which must be

_ reviewed and approved by the TUEC Joint Test Group (JTG).

i i

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

_ _ - ~ _ , - -_, , __ ,,y_. ,.._-r m_- -_. - - . . .

N .-

Approximately 95 percent of the TDRs issued relative to HFT documented piping and equipment supports and restraints not installed prior to For example, see TDRs 680, 722, 746, 747, 837, the start of the test.

1006, 1032, 124.3, 1244, 1665, 1674, 1724, 1786, 1799, 1851, 2034, 2106, 635, 709, and 732. TPD-l', issued against ICP-PT-34-05, " Steam Generator Narrow Range Level Verification," identified that a ,

substitution was made for steam generator water level detectors.

TPD-2, issued against ICP-PT-22-01,'" Process Sampling," identified that three radiation monitors were not installed at the time of the -

test and were not needed to meet the test objectives.

In every case reviewed by the TRT, missing components and equipment were identified and documented in the completed test record. Any outstanding testing which remained because components and equipment were not installed at the time of the test was tracked by a deferred preoperational testing progr,am schedule implemented by STA-805, Revision 0, " Deferred Preoperational Testing." STA-805 is a CPSES administrative procedure. TUEC's decision to proceed with the HFT }

despite missing equipment appears to have been made to minimize the economic impact of delaying the testing program. Many other power plants have taken a similar approach in'this regard.

All preoperational tests, including HFT, which TUEC committed in the .

FSAR to perform prior to fuel loading and which are can'didates for post-fuel load testing, must be submitted to NRC for review and

. approval prior to issuance of an operating license. Those which NRC approves for post-fuel load testing will then be required by NRC, as a condittor of the operating license, to be satisfactorily completed prior to the reactor being taken critical. The TRT also reviewed a master data base computer printout of work items requiring thermally hot plant conditions to retest. As alleged,there were modifications (about 74), most of which were on hangers, snubbers,and other pipe supports, that required HFT conditions for valid retesting.

9

Thus, the TRT found that while some components and equipment were not installed during the HFT, they were documented and tracked to be included in testing which, if approved by NRC, will be performed after fuel is loaded into the reactor core.

b. In assessing the allegation that TUEC does not intend to check or' monitor some components and systems until " heat up to hot standby" or "during power ascension," the TRT reviewed Integrated Plant Operating Procedure IPO-001A, " Plant Startup From Cold Shutdown to -

Hot Standby." As in the case for HFT, this procedure specifies that the plant be taken up to normal operating pressure and tempera-ture using reactor coolant pumps as the heat source. Whether all HFT items are done before or after fuel load is not safety significant because the fuel is new, i.e., unirradiated. Additionally, some pre-operational tests can be done only after fueling because the reactor core must be installed. Examples of these are: ISU-022A, "RCS .

Boundary Pressure Test and Leakage"; 150-0228, "Incore Moveable Detector System Alignment";. ISU-021A, Pressurizer Spray & Heater Capacity Test"; and ISU-228A, Control Rod Drive Mechanism Operational Test." These tests, and tests on components and equipment not installed during the initial (prefuel load) HFT, were scheduled to be conducted after fuel load but before the reactor is placed into  :

operation (see paragraph 4.a above). The results of this t'e sting .

I

_ must be satisfactory prior to reactor initial criticality. There are no HFT items scheduled to occur "during power ascension" except those that require reactor power. For example, steam and feed water piping does not achieve design temperatures until there is sufficient flow, which only occurs at power. Accordingly, this portion of thermal expansion testing c'annot be completed until that time. l Section 14.2 of tha FSAR and Regulatory Guide 1.68 specify those tests which are to be accomplished during power ascension.

c. It is alleged that neither TUEC nor the NRC Region IV staff noticed that major components or items of equipment were not installed prior to HFT and failed to keep the ASLB . informed of the problems encoun-tered.

.* j 5-The TRT reviewed HFT-related TDRs and the master data base to determine whether TUEC had documented all outstanding work on the master data base for the Lead Startup Engineer to review prior to each test, and that components not installed at the time of testing, l

. -but needed for eventual system operation, were documente'd, as required by CPSES administrative procedures, 'NDRs or TPDs. Fo f' example, as discussed in paragraph 4.a above, Ehere were 20 TDRs

/ identifying the missing hangers and supports associated'with

-- ICP-PT-55-11, " Thermal Expansion." Each was initiated by the -

/ ,f, Startup Group, and evaluated by TUEC engineering for its impact on

~~

the test results. TUEC performed calcuiations an Tinstalled temporary supports and weights during the test so that supports 0j[y - '

which in normal operation would interact with other supports would

not, yield erroneous data. - -

.} A

/ The TRT also determined that the reason for no documentation in

, .NRC Inspection Reports indicating the Region IV staff's knowledge of missing components was because they were documented and tracked in accordance with the TUEC administrative procedures which provide for such possibilities, and b'ecause they were included in planned and documented future testing activities. It is not unusual for an

applicant for an NRC operating license to defer certain prerequisite ~

equipment installations in order to proceed with HFT. However, the .

NRC routine inspection program verifies that a viable system exists

and is being implemented to document and track such missing equipment I

and that the equipment is satisfactorily tested when it is finally installed. This was done by NRC's Region IV staff during various

routine inspections of TUEC administrative procedures and was con-firmed by the TRT during its review, as described in the preceding

! sections.

During the TRT revie.v of the test program, no ir. formation surfaced that was considered relevant to any issue now pending before the j

l ASLB Comanche Peak hearing. The purpose of a prerequisite and preoperation61 testing program is to prove that the design and construction of the plant is such as to provide reasenable assurance  ;

i e

v-- --- -~-n - .,n- .na wya - ---- - - - - . . . ------.-----,-n..--. --.-..--.--g.w-- , - -..n----m-.--e,.v. -

~- - - , --

--mm . - + -

p that the health and safety of the public is protected when the plant ASLB notification, in this case, would only

/

P h[#' fgoesintooperation.

have been required if the testing program revealed a problem which could be considered a safety. issue or which could not be resolved to c.3 f "' r j the satisf.ction of the NRC. Therefore, at the time of the TRT

[te'

' Y review, no ASLB notifications were required concerning the testing

\ % programs, e, W

.e

, g.f~

, / d. It is alleged that 60 percent of the test points of ICP-pT-55-11, -

. I " Thermal Expansion," failed the acceptance criteria, that the i traceability of the measuring devices was lost because they were not logged with the data, and that TUEC engineering had provided I no justification for the "use as is" determination on piping which did not meet expected values.

The TRT staff determined, through discussions with TUEC personnel .

i and by a review of the completed portions of ICP-PT-55-11, that

, about 28 percent of the test points (referred to by TUEC as

" monitoring locations") failed the acceptance criteria for reasons that were not totally unexpected in the course of an HFT. TDRs were issued to document all test failures so that TUEC could provide corrective actions and establish retest requirements. Additionally, about 12 percent of the monitoring locations were not measured -

because of missing equipment at the time of the t'ests; about 7 l

percent were invalidated because equipment was removed during the test; and about 3 percent were invalidated because of modifications to equipment after the test. Therefore, about 50 percent of the monitoring locations still required measurements. Those which must be measured under HFT conditions are included in the testing program '

proposed for after fuel load and were submitted for NRC approval.

The allegation also stated that although temperatures were taken and logged during the test, the specific measuring device used at each monitoring location wa.s not logged, therefore, the calibration.

, of the measuring device could not be traced to the monitored location.

The TRT. staff found that the completed test data packages did contain

-y-1 e the calibration data for the measuring devices used, but as alleged, the devices could not be traced directly to specific monitoring -

locations. While pursuing this matter, the TRT conducted interviews with TUEC personnel who participated in the testing and found that a

~

. test coordinator maintained' a leg which tied the devices to the specific monitoring locations; however, the log was not a part of' the official test package. The TRT pointed out to TUEC that while the direct connection was not required by the test procedure as written, the data must be included as part of the official data -

package. \-

A TRT review of representative TORS, including TOR-853, 854, 855, 1033, 1034, 1035, 1112, and 1113 identifying questionable data or

, deficiencies revealed no cases where TUEC engineering had not provided back-up data and/or calculations supporting a justification for the "use as is" disposition of a TOR.

e. It is alleged that in conducting the HFT, TUEC considered only normal operating conditions and did not consider accident con-ditions, such as loss-of-coolant accident (LOCA) or an earthquake.

During its review of test procedures, the TRT found that TUEC tested' safety systems with consideration for accident conditions to the -

extent possible by simulating certain parameters such as temperature, pressure, flow, etc., thtt might be encountered during an anticipated accident or emergency condition.

Moreover, the NRC does not require testing under actual accident conditions. Each applicant for a permit to can'struct a nuclear power plant must include the principal design criteria for the proposed facility in the application. The principal design criteria in 10 CFR 50, Appendix A, establish the necessary design, fabrication, construction, testing and performance requirements for structures, systems and components important to safety which provide for reason-able assurance that the facility can be operated without undue risk to the health and rafety of the public, including during accident

. ~-

conditions, such as LOCAs and earthquakes. CPSES is designed and constructed with the systems and features needed to mitigate the consequences of an accident, and takes into account lessons learned from past experiences with other plants.

f. It is alleged that TUEC and the NRC Region IV staff were willing '

to accept HFT results which were deficient.

TRT review of the completed HFT and other preoperational test

~

documents indicated that with a.few exceptions as noted below, no deficient test results were accepted by TUEC. Final acceptance of HFT test results does not occur until the Joint Test Group (JTG) has conducted its review of the data and approves the completed test package. The TRT found three minor, questionable items in a sample of 17 out.of 25 JTG-reviewed and approved procedures. These iteins were: -

, (1) Preoperational test procedure 1CP-PT-02-12 " Bus Voltage and Load Survey," intended to demonstrate that during all modes of plant operation, optimum current and voltage will be present

/

/ at all the buses and subsequent equipment. A prerequisite

! condition (paragraph 6.5) of the procedure required certain, transformer taps (TIEB1, 2, 3, & 4) to be set at -5% and others-(TIEC3 & 4) to be set at -2.5%. During the test, the voltage acceptance criteria (paragraph 2.0) could not be met and the t \

j '

- procedure was changed, in accordance with administrative require-

ments, to specify -2.5% in lieu of -5% and +2.5% in lieu of h[s i 9 -2.5%, respectively. However, engineering review and evaluation

,' of the tap changes after the test was completed resulted in the ,

requirement to return to the originally specified transformer  ;

taps. As a result, the' completed test record contained data that was taken with incorrect tap settings. In addition,

}

I voltages recorded in paragraphs 7.8.2.1 and 7.8.3.1 did not I

meet the acceptance criteria (paragraph 2.0) of the procedure.  !

e

'W 6

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

. .g.

(2) Procedure ICP-PT-34-05, " Steam Generator Narrow Range Level Verification," intended to demonstrate at hot, no load con-ditions, that the narrow range level channels for each steam generator indicate properly at the upper and lower instrument

. taps, and compare pro;ierly with each other for actual changes in steam generator water level. Level detectors 1-LT-517, 5'18, and 529 were not availa'ble at the time of testing and thus temporary equipment was substituted. The test was performed

with the temporary equipment. After the test, the specified -

detectors (from a different manufacturer) were installed. The ,

doint Test Group (JTG) approved the completed test package con-

.taining data taken with temporary detectors. The only retest specified after installation of the specified detectors was a cold calibration which does not meet the objectives of the test.

(3) Procedure 1CP-PT-55-05, " Pressurizer Level Control," i.. tended to demonstrate the control aspects of the system in conjunction

~

with the chemical and volume control system. In addition, -

l there was a note on Page 12 of.the procedure that stated, "This '

test is provided to verify the capability of the pressurizer level control system to monitor pressurizer level over the range ,

of installed instrumentation and to observe that all alarm and ,

control functions are operational." A prerequisite condition -

(paragraph 6.13) required the plant to be in hot standby.

During conduct of pressurizer level indication testing, in accordance with the procedure (paragraph 7.1), the System Test Engineer noted that a level detector (1-LT-461) was registering marginal readings. He documented this and recommended a cali-bration check of the detector. After the' test was completed,

, this was done, and it was determined that the detector was out l of calibration, and attempts to calibrate it were unsuccessful.

The corrective action was to replace the detector and perform a .

cold calibration. The JTG-approved test record contained level data that were taken with a detector thaf. was sub.sequently i

proved to be out of calibration, thereby invalidating the test i data.

- . _ __. m y . - . ,, -. .- ,_ _ _ _

,g. .#-- _ -

l When the TRT effort began, the NRC Region IV staff had not yet begun their inspections of HFT qorpleted test packages. This NRC inspection effort has as an objective to assure that all test data are either within previously established acceptance criteria, or that deviations are proper *.y dispositioned. $ince the Region IV review had not yet befin, the implication that the Region IV staff was willing to. accept tiefic1ent results was not appropriate.

g. b is alleged that the ASLB cannot rely on the NRC staff to monitor '

t.he complptf or, of any aeditienal t6st and reinspections cue to the lack of pancor by both TUEC and the NRC Region I8) staff, which calls ,

into quesY.fon their credibility and/or compctenca.

Ouring its review, the TRT tound TUEC personr.21 and Regicn It staff candid and forthcoming. Since the Regicn IV, staff had not yet be' gun their igspectiori of HFT-completed test pactages, it was premature to -

allche that the ASLB could not rely of: the NRC staff to monitop the

, completion of any additional tests ard irspecticas. There were to findings tnat indicated a lack of candor on t9 part of 1UFL or the NRi' Regten IV staff curing the corduct pf the testing program or brot ght into question the credibility or com$etente of either. '

l ,

5. Conclusion and Staff Positions; The TRT found no information during review' gf of these allegations which would preclude the ASLB from mating a decision

\ pursuant to 10 CFR 50.57(a). ThereWerenosignificantdeviat3sorvio- ,

g. , [ lations of NRC requirements identified du' ring the course of HFT to support p.'

,A

' . 01 the contention that it was deficient. Although the HFT was incomplete,

\

,,b TUEC's plan to complete it after fuel loading.and prior to initial criti-

[. cality appeared technically soun'd and without any safety implications as it was then constituted. However, it will be necessary for TUEC to obtain NRC approval to devfate from,certain limiting conditions of the CPSES Technical Specifications since these were written for an operating power l plant. For example, Section 3.7.9 requires snubbers to be "0PERABLE" 1 l

before heatup above 200*F. There are snubbers which must be tested with.

e l

_11_

the plant at normal operating temperature before they can be considered "0PERABLE", which is what HFT accomplishes. Since the reactor core will not have been irradiated, the TRT does not consider such a deviation to be of safety significance. -

The HFT portion of the preoperational test program appeared to be accom-plishing its intended purpose, that is, te identify problems such that they can be corrected prior to reactor operation. There was no evidence .

found that either TUEC or the NRC Region IV staff was willing to accept *

' deficient test results and it appeared that the overall objectives of the CPSES Unit'l preoperAtional test program would be met, thus providing reasonable assurance that the plant is properly constructed and that its operation will not pose a threat to public health and safety. '/hile some of the, allegations had valia bases, none were considered to have overall safety significance nor gensric implications.

6. Actions Required:

i

a. Section IA(B) of the FSAR commits TUEC to complete all preoperational testing, including JTG review and a%v@ oval, prior to fuel load with exception of those tests which cannot be conducted until the core is installed. This subject is discussed in Section 4.a of this report.

Now that TUEC is obtaining NRC approval to con act some cf this - -

testing after fuel load, TUEC shall commit to having the JTG. or a similarly qualified aroup, review and approv$hl all post-fueling phaoperationaltestresultspriortodeclarin'gthesystemoperable l W)F41 q in accordance with the technical specifications.

v b. Section 4.f of this report refers to three predperational tests cen-ducted during HFT that the TRT determined were not completed to the extent required by the objectives stated in the test procedures.

• According, TUEC shall review all complete preoperational test data ,

/ packages to ensure there are no other in.stan.ces.,where._t_e..st objectiv.es

..g were not met, or prerequisite conditions were not satisfied. The

.. ~ . _ _. . ... . . . . _ , , , _

e e

three items identified by the TRT staff shall be included, along

! with appropriate justification, in the test deferral packages pre-sented to the NRC.

c. The TRT de-armined, as indicated in 4.d of this report, that the data ,

in ICP-PT-55-11 " Thermal Expansion," did not include information

needed to trace the measuring devices to the monitored locations,' but the information was available in a log maintained by TUEC. TUEC shall

incorporate the information contained in the log into the official -

i i ICP-PT-55-11 data package so that the traceability is maintained, and shall'also establish administrative controls to assure appropriate test and measuring equipment traceability during future testing.

i d. The TRT pointed out in Section 5 of this report that in order to conduct preoperational tests at the necessary temperatures and ,

pressures after fuel load, certain limiting conditions of the .

, proposed technical specifications cannot be met, e.g., all snubbers I

will not be operable since some will not have been tested. .

I Accordingly, TUEC shall evaluate the required plant conditions fg i the deferred preoperational tests against limiting conditions in l

l proposed technical specifications and obtain NRC approval where 1 i deviations from the technical specifications are necessary.' .

a s F ,ag a g g .

, 7, ,.< . , . . 7.

.g[

2 T'

) -

[' N j 45 -

- . , r 1 f a t* s-1 .. a. , . .. ., , 3 e j .' 1 , ' . '., ( ' ,

4 ' -

", .1 s a

g, (, - .

g., . .

i

. s n., 1

.+ o, e

i i

i l .

.e. - Jg. .__i aa m - 4-.*s- w -g pwe A -5 F a -- yew--s'4-=vF-G S

i

, , 9- '

...+.m . 3 + , ,, ..

ep . j.

>. s 4

. . .. ,g,

., . . . .s . . . .

,j . -

• D

. , g. .. t' * ,. -7., ,

1  ; , , 1 .

e

f*.:.

.,W-4.. -.

l ' . <

.' .f':.

J 2 .- g :Qy g.;. :( . ., .

L-w ww.

.. .. _. g ., p ..e.

1

... 4+ yn

9. s .s l . .

b }3;^ _' y , ' .Y ..

N- s f.[.

y .[ .

' A

ffW q.9 - j ,4 - o9- 1: ..g .g p . ... .

,s. g.

.~g - ,

~ , ( . . ,. .. .

(_,

'?)

n '1. .

s ., '

**L,, !. r . *&l .

,, ' \. " . ,

4.( .-,'

f N.' ^

> .; -ef-$'y.y ,', f,,;iQ..

. . r. (- . ' ' v.. .- .

. . , .2

[.

. l ...

~.4

~._:.3 y _

._ n-%.;.,;g g j

,,e.;3 f , , . . .. , . g .

- . . . . .- , . ....,.f. . . , , ,. -4 & c Ayw )

- ..., ! ; s ..- . ..

+ - -

1: .. . . n ,. . .

,,,Q_ l

, g- ...K

.f'i

• u

,.. , 2.,

- Q.. ^'.

l e -gq

, ' , , . .. '. c . . . . . . Jo Wi 4 J4

[')

, 2.I '#

~

.c e . ( .

7 , ',, . . ' . , , .. k .

.g ... , .< ,v .. - .. , . p. . . . ,y .

c .

j~ ~ .

j h. '

.. . , . 4 i

..N

% 1r g \vg . .,' #

r , .. g.p .\; , .

_ j, .. . s 4 . . . . . f. ,

'l N. 3 hg'$ ,

. . - . .i , - e ,

3 y

. 4; e

. . . . ~  : .. ,.

j ..- ..4,., .

. .f 4 .

(

'I

. j;gp. 'I.',*,

f. ,- ,4, , -

,, is. '}

.?, Q ,

., , K.. ., x . . ,,,>g. e . ' ,' e ~~.s, .y a 3' .p 1

,n

g. ,::.Mtp2% L.';W.vc% y ' % u.% g y _.

~. g .}n.; x k %;f D; 2 . ~ ~ s . %.;

• 4,.s ja$. ;,4 ].- '; .J e

m. T..

p ~.y g [ n c w ..p ;,Y.%.~1 .

" v n. w 4kW.

y,;.;.y:.,.'-up r W W

, ~

. 4.. ~

y

.gy.g e >f{y w.Apg., 43,g.;. g g, l .

v..g ~

c A. . .. me,. . .2 , , e f. O

..o 4 l

M . f ;, .  ?

f,<f

\

ewee g g ==

t . l .-

$5h ,f .y  ? .. *k.t*':E&

,, _ ,kW.19hf'N . '*

j d ,)(44.Y?@.' . ) L " .: l '- i M ...

,.[:l g Q :.h f %.$ } ..k

. ;: L  %

5-f ggy-

- q.5 p A g .q7a+ '. a

% -

,. .):<: - y . pk;g

..,.L_. +y

.g it , ,'

^

~, p n V  %*.,

c. .

n'%.'..*',L .

% ", Q, es. s .

- v.

. g%.; ' _

a. , :, . . , , , , - - . , .- p -

i , ,- - . c3 .  :

. y

' gv

.- ..,.e+ $.e . -(w 3 w  ? # 3 ,- w .

. 4.m I

=. ' .' 4 i, -

M h

. j , . .. w ,.

g. .

+.y s

.., e..

y [., .g h .

e.,,f

( g .h.

2

_ . , . (4 s .,

' w. .o n...

.,y ., .

A. 4) i

+ 9 ,..

c.  !, - -

f sgM  % '.41.'

3 . . .- .

s.. .

P' g i V.. ;i K t

~~- '^

l J'--[. g.c s;. .

. A,% 7 .

n w . ' ..

• . yj

,,' J$ j~ . .

' pp.

?-- - ..

l . . .. . ~.

R .-

j. .,

.m s:. q>

. g  : .'- ,

,%4 s..

. L' 3'

.M b ' - * - '

ft(;gl '

.? , ,:2a,. ' : , . [- __ _

v.> .. ,,.. : Q. [ 3. , _ ; t,..

's ' A.g . v '. . f'.> .v -

i q_

.- 3 . ' , . i.I f"  ;;W.44 - -.

j ,. , . c , ,p .

3 s.

, y* # .33 + -- _..j....c,.a,.

1 .-

. 4, e

l

<. %s. ... :6 e y ka* -

j e i .

s g_ M

' ^

l . . . , .. 9:3 '

~- ' f , . k. . . e ,

' - -* . y. ? . -

. , j ,. [,

w. .hv[.- ,; [-[,$' ) ' % .

7 V. .

y.,

}

1 .. . .a .

.... . . . . .y ,, .

p n. v-x 7,. . : .x ., y

,. y . o . '

o

.-,-t4, a . 7 7.

4. .

s. . ..  :

[.';. - . . . _ -

If

~

.: .T r -

1.:/M, > h- .cv?,< ., : . . . . .

.q

m - ,.

.< Q.",,-. .

s ..

. s .. . -

3 . .' . .,.

nn

. .. . y g* , . r - .. .; .. --' ...

f , .n .J ,.% ,

. :2tx. 7 .% ~ '.* - '

n , , '. >

~ . ,

I .

L1 e._ < N .N N ;1 V ._ ~ ? . . - - _.

.w t; .

~

g -

[)l ., y .

-.s t:y : f. _

, _ - i. .

- '; (%. ,

o. . .

,M, ,,s.

• q- (3 # ff . -  ?. - ,,.", '

~

f f g ?:,, '. ,

', .u q'.~_3 l' , , .

M .k,',

~,

~ ** '

s ,-] ,

,. j ., . - .. .; .- . .,

_l... s -

o . ,.

, , .. ,1. t , . ..

' V. y. , 7 . , . .

. ; .j. 3 l .

., -t . , _ ..

f.f.;. .: # f.i , t ,

. ps .'. -, . t . 4 7 -l . . ' (! 3, v

, '. ..f3 f - ) '.~ ff.M ., ~.'S _ . F. , ' ..

N --

N I 1

AL(;- ll Nl.-

k, (->

m Attach-ents: Nene. fd l

, p.e v

/ f. L 6 g*

• r t 0 ,. u M., .

, j o. l

• ('

r -

(\ ..

, 1.

'.. ) p . .,v)' ..x.. c )

I IY I

p. ,

}' 'N

15-

9. Reference Documents: .

1, ICP-PT-02-12 " Bus Voltage and Load Survey" (Rev. 0/2-7-S3) '

2. ICP-PT-02-13 " Power Transformer Load Test" (Rev. 0/.2-1/83)

3. ICP-PT-02-14 , "480V Switchgear Transformer Load Test" (Rev. C'/2-7-83) ,

4. ICP-PT-22-01 " Process Sampling System" (Rev. 0/2-1/83)

5. ICP-AT-22-02 " Secondary Sampling" (Rev. 0/9-27-82)

6. 1CP-PT-22-03 " Post Accident Sampling System" (Rev. 1/4-8-83)

7. ICP-AT-28-01 "Feedwater System" (Rev. 0/1-21-83) ~

8. ICP-PT-34-02 " Steam Generator Safety and Relief Valves" (Rev. 1/3-11-83)

9. ICP-PT-34-05 " Steam Generator Narrow Range Level Verification" (Rev. 0/3-4-83)

10. 1GP-PT-37-03 " Auxiliary Feedwater Turbine Driven Pump" (Rev. 0/2-18-83) ,

11. ICP-PT-45-06 " Containment and Pump Room Coolers" (Rev. 0/3-25-83)

12. 1CP-PT-55-02 " Hot Functional Test Sequence" (Rev. 2/2-18-83)

, 13. ICP-PT-55-03 " Pressurizer Relief Valves" (Rev. 0/1-7-83)

14. ICP-PT-55-05 " Pressurizer Level Control" (Rev. 0/2-25-83)

15. ICP-PT-55-06 " Spray and Heaters" (Rev. 0/1-21-83)

16. ICP-PT-55-09 " Reactor Coolant Pumps" (Rev. 1/4-8-83) ,

17. ICP-PT-55-11 " Thermal Expansion" (not completed) (Rev. 0/12-21-84)

18. CP-SAP-21,Rev.2 "ConductofTesting"(3-7-84) -

19. CP-SAP-12,Rev.2 "DeviationstoTestInstructions/ Procedures"(3-18-83)

20. CP-SAP-16,Rev.8 " Test Deficiency and Nonconformance Reporting" (12-7-83)

21. STA-805, Rev.0 " Deferred Preoperational Testing" (4-23-84)

22. IPO-001A,Rev.1 " Plant Startup from Cold

• Shutdown to Hot Standby"

23. CPSES Final Safety Analysis Report (FSAR) Amendinent 59,7-14-84)

24. Regu,latory Guide 1.68, Revision 2, August 1978 " Initial Test i Programs for Water-Cooled Nuclear Power Plants"

25. NRC Construction Appraisal Inspection Report 50-445/83-18 dated April 11, 1983.

26. NRC Inspection Report 50-445/83-08 dated June 30, 1983 ,

27. NRC Inspection Report 50-445/83-23 dated July 27, 1983 1

I

9

23. NRC Inspection Report 50-445/83-22 dated May 25, 1983

^

29. NRC Inspection Report 50-445/84-16 conducted May 14-June 20, 1984

30. NRC Inspection Report 50-445/83-26 dated June 15, 1983

31. AT/P.T Test.Index with Schedule, Unit I and Common dated July 2, ISt4. .

32. System / Subsystem Turnover Definitions and Target Dates, dated .

February 7, 1984.

33. Citizens Association for Sound Energy (CASE) proposed Contention No. 26 of October.13, 1983. ~

10. This statement prepared by:

Ward F. Smith Date TRT Reviewer Reviewed by: -

Richard R. Keimig Date

. Group Leader Approved by:

Vincent S. Noonan Date Project Director .

8 1

5 I

)

j l

e

6. s

• Doce.ent Narne:

• • SSER TEST PROGRAM 4 R uestor's ID: ,

I Author's Name:

Chet Posiusny '

Occument Comments:

11/8/84 Final draft, rev.1

~^

"***= % .

g. 7g p -re.snuact N,

I t4G2 -

shculd GC M%

-b@d d'ho $@[

o ASS 6w1b(f P C 'h ' a.+i o /)

sexts gwws\q  ?

w

_ 3.sz -p. 2. )~'~3 _ ,

'f  !

f ,.

jzJw'

- ~

Ly , ,. /'

t CDBA-85-59 n ,+

1 a j

i.

l Final Draft, Rev. I 11/7/84

. SSER Test Programs /CP3 l SSER l

l 1. Allegation Cate.ory: Test Program No. 4 '

l

2. Allegation Number: AT-7

3. Characterization: It was alleged that the leaks encountered during the .

containment integrated leak rate test (CILRT) were numerous and of such magnitude that they will have to be' corrected and the test repeated before fuel loading.

4. Assessment of Safety Significance: The implied significance of this allegation is that the containment butiding might not be capable of meeting its intended safety function of acting as'the final barrier .

against the release of significant arrounts of radioactive fission products to the environment in the event of an accident.

A condition for an operating Itcense for a water-cooled power reactor, such as Comanche Peak Unit 1, is that the primary reactor containment building meets the leakage test requirements set forth in 10 CFR 50,

• Appendix J, " Primary Reactor Containment Leaktge Testing for Water-cooled, .

Reactors."

l Appendix J of 10 CFR 50 requires preoperational testing of the overall leak tightness of the containment building (CILRT or, Type A test) and establishes acceptance criteria for the test. The testing is conducte'd to assure that total leakage through all designated penetrations and building flaws, if any, does not exceed the value specified in Appendix J or the Technical Specifications (which are currently under review by the NDC as part of the lice'ising process). .

Coth 10 CFR 50, Appendix J, and the Comanche Peak Steam Electric Station rinal Safety Analysis Report (CPSES/FSAR), Amendment 12, dated October S, 1930 specify the use of the American National Standard (A'lSI) MS,4-1972.

l I

. -2

" Leakage Rate Testing of Containment Structures of Nuclear Reactors,"

March 16, 1972, to carry out the test. A later revision of the ANSI standard (ANSI /ANS 56.8, " Containment , System Leakage Testing Requirements)"

prescribes essentially the same test procedure for the CILRT as ANSI N45.4-1972 but prescribes another method for calculation of the leakage rate. ANSI /ANS 56.8 has not been endorsed by NRC and is not prescribed in 10 CFR 50, Appendix J.

The TRT reviewed the as performed CILRT procedure, 1-CP-PT-75-02, ,

" Structural Integrity Test and Integrated Leak Rate Test," Revision 0 and the resultant test data to determine compliance with Appendix J and the proposed Technical Specifications. The TRT determined that, as alleged, numerc,us leaks were detected during the first two of three attempts to measure the containment building leakage rate. On each of the first two attempts, when it was determined that the leakage rate wou!d exceed the maximum allowable rate, the test was terminated, the containment pressure reduced to a safe level for entry into the building, and the suspected leald corrected. Priortothethirdattempt,testpersonnelidentifLed three containment electrical penetrations (E-49, E-62, and E-68) for which the individual leakage rates. were excessive, but for which a method to stop the leakage was not then apparent. These three penetrations were isolated prior to the third test and documented on test deficiency reports for later disposition. Theresul'tsofthethirdCIIRTattemptwereco_n.,

sidered satisfactory by TUEC. The CILRT was observed by two NRC inspec-tors (reference NRC Region IV Inspection Report 50-445/83-04) to ascer-tain whether the test was conducted in accordance with the approved pro cedure. The NRC inspector also independently calculated the leakage rat ,O (ch using the method defined in ANS N45.4-1972 and Draft 3 of ANSI /ANS 5' yc-56.8-1981 to determine the validity of TUEC's test results. 6-9 Subsequent to the third test, the three isolated penetrations were 6 ,

individually leak tested to establish their specific leakage rates prior t,o repair. The cause of the leakago was identified as improper assembly of the penetration seals. The penetrations were reassembled and individually leak tested again with satisfactory results, in addition to these three electrical penetrations, four other penetrations that needed 9

-3 ,

l .

_to be onen to conduct the CILRT were individually leak tested. The i measured leakage rates from the repaired electrical penetrations and the measured leakage rates from the four penetrations used to conduct the test were added to the measured leakage rate from the CILRT. This

addition was insignificant and did not alter the least significant digit  ;

in the previous total leakage rate. The total resultant leakage rate was  !

] less than the allowed maximum for the containment building under the '

proposed Technical Specifications and 10 CFR 50, Appendix J.

T

• During the third test, test personnel recorded data and calculated contain-

i 1

l ment building leakage rates as prescribed by ANSI N45.4. These leakage rates remained consistently lower than the maximum allowed in 10 CFR 50, Appendix J and the proposed CPSES Technical Specifications. However, the i calculation of the containment leakage rate included in the summary report l submitted to the NRC, as required by 10 CFR 50, Appendix J, (" Comanche'  ;

Peak Steam Electric Reactor Containment Building Unit One Preoperational -

4 Integrated Leak Rate Test," 1983, Docket Number 50-445. Texas Utilities ~

j Generating Company and addendum, July 1983) was performed using the method j -

prescribed by ANSI / ANS 56.8. This value was consistent with the value calculated by using the method in ASNI N45.4 and confirmed that the con-tainment building leakage was less than that allowed by the CpSES Technical 1 Specifications and 10 CFR 50, Appendix J. -

l

5. Conclusions a_nd Staff Positions: The TRT determined that while numerous ' [

1eaks were identified as alleged, these leakage paths were documented and i the leakage was stopped prior to the successful completion of the CILRT,

• with the exception of three electrical penetrations and the four

• per.etrations which were needed to conduct the test. The leakage rates-I from these penetrations were later added to the total leakage rate. The preoperational leakage rate was calculated and found to be lower than the

{ marinum allowed by NRC regulations, a determination verified through independent calculations by NRC inspectors. Therefore, the containment 1

i J

I

• i #

e-w,---w+ --ev_

. -4 building was proved to be capable of meeting its intended safety function.

g The method for calculating the leakage rate was as prescribed by ANSI /ANS LC 56.8-1981. While this method differs from that prescribed in ANSI N45.4-

'

• 1972, because of the stable and censistent data obtained during the con-

. duct of the test, the leakage ra.te which resulted from the use of the cal-culation method in ANSI /ANS 56.8-1981, would be essentially equivalent to

/ the results which would be cbtained using the method in ANSI N45.4-19,72.

Q ,... -"'*~~~

r/

However, it is the TRT's position that TUEC should have obtained $

, \. .

authorization to utilize a calculational method not endorsed by the NRC to report the results of the CILRT. Further, the TRT considers that conducting the CILRT with three electrical penetrations isolated, while perhaps technically insignificant with respect to the test results, is contrary to the purpose of the pre-operational CILRT and should not have i been done without specific approval of the NRC staff.

- - - - ~

.._, __.. _ . _ _ _ . . . . _ . .~

1. This allegation has reither safety significance nor generic implications 1

The two problems described above (ave b'e'ili')

,'..l..c,,forpurposesofthisreview.

forwarded to the NRC Office of Nuclear Reactor Regulation (NRR) for reso-

, ., 9 y pi

• 1ution.

i

6. Actions Recuired: prior to fuel loading, TUEC shall identify and justify QI to tha NRC. staff-why the preoperational CILRT (Typ e A test) was conducted,

[8

' with penetrations isolated and why TUEC used ANSI /ANS 56.8 rather than

,, A53fI/N45l4"to calculat'e the leakage rate. TUEC shall also identify and justify any other deviations or differences between the CILRT as performed, y ,] and the applicable requirements or commitments.

, hot v7 sj , . -Q (N y  :  ;/

7

.,$ .  ;. +

c -

' 4.- ; ; .ssy -,

,c!

y ,, [ny, . .

t

. y.. .  ; - ,. n .. - n T J; is (V .

~

^m 5

t.']

S.. ., .,;.

. . . y.

.a .,

, .. . v ,

v ..,.

'e

'y, r ,J

. . .I

,~

t ,

t.'. . '

,1 # +

h.

g .: 3 . n ~ . a ,. + c , .m - ~ .x - -

n. e i[a1 1 - .---.,-

- - - -'-" F ' " ' ' " " ' " * " " ' " " ' ' ' -~ '~

~

/

-5

8. Attachments: Nc. e.

9. Reference Documents: ,

1. Comanche Peak Steam Electric Station Final Safety Analysis Report, Amendment 12.

2. 10 CFR 50, Appendix J.

3. ANSI N45.4-1972.

4. ANSI /ANS 56.8 - 1981.

5. Comanche Peak Steam Electric Station Preoperational Test Procedure 1-CP-PT-75-02 Structural Integrity Test and Integrated Leak Test,.

Revision "0." Test performed beginning Janua'ry 12, 1983 and completed, I

February 8, 1983.

"I '

l 6. NRC Region IV Inspection Report No. 50-445/83-04.

7. Comanche Peak Steam Electric Station Reactor Containment Building Unit One Preoperational Integrated Leak Rate Test, 1983, Docket Number 50-445, Texas Utilities Ge'nerating Company.

8. Comanche Peak Steam Electric Station Reactor Containment Building

Unit One Preoperational Integrated Leak Rate Test Addendum, July ',.

l t . 1983, Docket Number 50-445, Texas Utilities Generating Company.

,9. Citizens Association for Sound Energy (CASE) proposed Contention l l

No. 26, October 13, 1983. 1

\

\

1 j

l i

I

b .

t ..

e

-6

10. This statement prepared by:

Arthur Mackley Date -

TRT Reviewer e

Reviewed by:

Richard R. Keimig Date Group Leader Approved by:

Vincent S. Noonan Date Project Director e

9 e

G e

b 4

a 9

e O

6 O 9

_. - - , . . ..i - , .r. _, y._, . ,.., , - . . -.--- , . . , , ".gf y . - , . - -

--w-.

- Document Name:

* j SSER ALLEGATION AT-14 p O Requestor's ID:

• PAT Author's Name:

Chet Posiusny Document Comments:

11/8/84 Redraft 1," space

~

'i' g 7g nuct

/

T Ml a, b) c) Y '/

J'/.

./

/ / \,

w

, l ,

i0

.

• j[}#

,1, '

i l 4' Jp ' W . ,J A'/ ,. m . -

/

_7

, 4 , *

/ .,>

m

/

.' v' i

\'

1l M A-85-58

• / ,.<

, Final Draft, Rev.1, 11/7/84

, CP2; SSER Allegation AT-14 SSER

. l

1. A11ecation Cateoory: Test Progr,am k g'98f No. 5

= .

2. Alleoation Number: AT-14a, b, c, d m

3. ' Characterization: It is alleged that prerequisite testing was performed w y bcraft personnel not qualified in accordance with ANSI N45.2.6, "Quali- 1 fication of Inspection, txamination, and Testing Personnel for Nuclear Power Plants;" that System Test Engineers (STEs) were signing for tests that were conducted by craft personnel when in the majority of cases the STEs were not present during testing; and that test documentation was made to look as if the tests were performed by STEs, when in fact they were performed by craft personnel and the STEs only reviewed the data.

4. Assessment of Safety Significance: The implied safety significance of this allegation is that if prerequisite testing activities were performed by craft personnel not trained and qualified in accordance with industry standards endorsed by NRC, errors could be made which could invalidate the prerequisite test results. However, this could only have caused problems

.during conduct of the subsequent pre-operational tests which followed.

The CPSES Final Safety Analysis Report (FSAR), Section 14.2, describes the initial test program. It is implemented by TUEC through a series of administrative procedures contained in the "Startup Administrative Proce-dure Manual." The initial test program is divided into three successive phases: (1) prerequisite testing; (2) preoperational testing, and (3) initial startup testing which occurs after the license is issued for fuel load.

Tne allegations adaress the prelicense categories of " prerequisite" testing and "precperational" testing. " Prerequisite" tests are performed prior to "preoperational" tests to verify the complete installation, cleanliness, and initial operability of individual plant components (this test cycle is

3 also referred to as initial checkout), using a series of generic instruc-tions contained in the "Startup Prerequisite Test Instruction Manual."

These tests involve checks of electrical resistance, transformer polarity, relay and circuit breaker operability, motor rotation and initial opera-tion, initial r mp operability, systems cleanliness, and piping support adjustments. Prerequisite testing involves work normally within the expertise of journeyman craft personnel, and TUEC utilizes them in this 4 f-capacity under the direct supervision of an STE. The STE must certify the initial conditions of a prerequisite test are satisfied. He is responsible .

for the proper conduct of the test, and must verify and certify that the results are complete and satisfactory.

"Preoperational" tests follow the " prerequisite" tests and are conducted prior to fuel loading to demonstrate the capability of components, systems or structures to meet safety related performance requirements in the FSAR,

~

which is accepted by the NRC. These tests are performed by STEs who are qualified to ANSI N18.1-1971, " Selection and Training of Nuclear Power Plant Personnel." In order that. component problems can be identified and i

corrected prior to the start of preoperational tests, the prerequisite 7 tests described in the previous paragraph are performed as an initial v '

check of individual components to better assure their operability during ,

conduct of the preoperational tests that follow.

f, / .- -

.g '

In assessing the allegation that TUEC used craft personnel to perform pre-qf f.,

j

/ " qui e sti rtg--w management and support craf ut.2 E present, the TRT interviewed startup anagement personnel, and reviewed 35 prere-

't O , quisite test recordsT TUEC' representatives stated that they permitted STEs to utilize craft personnel to perform those portions of prerequisite tests that the STEs considered within craft personnel expertise, based on know-ledge of the craftsman's capabilities as determined by direct observation by the responsible ~STE. TUEC's position was that these employees are not

" test personnel," but rather are " craft personnel," and as.such, they were not required to be qualified ,NSI N45.2.6, as stated by that standard.

The NRC has accepted that position as consistent with TUEC's FSAR commit-rents to ANSI N45.2.5 and NRC Regulatory Guide 1.53, " Qualification of Nuclear Power Plant Inspection, Examination, and Testing Persennel."

l

TRT reviews of craft personnel records revealed that they were receiving appropriate indoctrination training. For example, the Electrical Test Group (ETG) craftsmen were indoctrinated in 10 pertinent startup administrative procedures and 14 generic prerequisite test instructions.

The TRT held discussions with selected STEs, ETG craftsman, and ETG manage-ment, and observed ETG craftsmen at work. No apparent training or quali-fication deficiencies were found for the type of work they were performing.

In some cases the ETGs' knowledge of the components and test equipment '

directly contributed to the success of the test.

It was alleged that documentation of prerequisite testing misleads the reader to believe an STE conducted the test-when-in-fact it wa..s perfor ed  %.

by craft__support _personu1MRT's review of the 35 test record inter- N

/ views with startup test personnel confirmed that craft personnel did,p in

~ N act, conduct

~ ___ some of_ the test.ing where appropriate.__.'The craft ' signatures

~

i on the data she~ets supported this finding; however, the STE was still responsible for the test and he verified satisfactory completion and signed the data sheets. This is not prohibited by TUEC's administrative proce-dures, excep_t__tkat t c aft personnel' performed veri.fication and signoff of initial conditions on some data she_ets,..which is a. function that must be done by the_STE per section 4.10.9 of CP-SAP-21, " Conduct of Testinsi."

Further investigation revealed a memorandum issued by the Lead Startup -

Engineer on March 31, 1983, countermanding this requirement in CP-SAP-21.

The subject of the memorandum (STM-83084) was "ETG Personnel Schedule Change," but it also indicated that craft support personnel (ETG) may verify prerequisites for Prerequisite Test Instructions XCP-EE-1 and y XCP-EE-14. Publishing such a memorandum in lieu of executing a properly approved change to CP-IAP-21 is in violation of CP-SAP-1, "Startup Administrative Procedures Manual," section 4.4.3.1, which requires a permanent or interim change to be approved and issued to all manual holders in accordance wit, CP-SAP-1. It appears that as a result of the memorandum, 24 of the 35 ttsts reviewed by the TRT had prerequisites

- ~ _ _ _

improperly

~--~~

verified by craft support personnel. Fifteen were XCP-EE-14, but nine were XCP-EE-24, " Fixed Battery Pack Operated Emergency Lighting Units," which were not even authorized by the memorandum.

; 3 f.h. < - "

d

,5. j

'y.\. l ... p tp

\

5. Conclusion and staff oositions: Asalleged,TUECutilizedcraftpersonn who were not qualified to ANSI N45.2.6 standards to support prerequisite testing activities. However, this practice is permitted by ANSI N45.2.6, as augmented by NRC Regulatory Guide 1.58 (Regulatory Position 7), which permits personn:1 who do not meet the ANSI Standards to engage in data-

. v) . taking and plant and equipment operation provided they are supervised by a

,-- )

0" e '.)\ T qualified individual and that they have sufficient trainin'g to ensure an /

- ~

C

);a/;J -

W sacceptable level of performance. Based on its review the TRT finds that j ; J e'd f . the craft personnel used to support prerequisite test.. activities were t- i

~

/. / sufficiently trai_ned, by virtu _e gi_theirj . -~ -ourneyman rating, tb perform

~-

theM 'ac~tivitiHIAdditionally,In thoulh they-were WoTunder the con-stant supervision of an STE, which is not required by ANSI N45.2.6 and Regulatory Guide 1.58, the TRT considered that because the prerequisite

/ test results they recorded were reviewed and certified by the STE responsi-ble for the test, and were approved by a test engi,neer with higher qualification than the STE, adequate supervision was exercised. .

g W.r! Q Allowing the STE to sign for data recorded by the craft personnel (and, i I'/

/ thereby, having the documentation appear as if the testing was performed f by the STE), is consistent with the procedure which directs the conduct 7 of testing activities. This procedure was widely disseminated onsite and was contained in TUEC's system of manuals and procedures. Therefore, the TRT concluded that there was no intent to deceive anyone on the part of .

TUEC and that the practice as implemented was sat.isfactory.

3 , .,> 3 -

-~~--- _ _ . . . -

1g This allegation has neith_et Safety significance nor generic implications.

I ,[ j /'W.which

~

6 Action Required:

TUEC shall rescinc memorandu.m STM-83084 of March 31, 1983j was issued in conflict with CP-SAP-21, and take action to ensure F

J 3 t h that there are no other memoranda issued which conflict with approved pro-J *' s,

(\ ,

,. cedures. TUEC shall'also conduct a review of all other prerequisite test c# u  ; :ords to determine those that had prerequisites signed by craft persec-d nel, and assess the impact of those improper verifications on subsequent hd(}/j,,d testing.

f .- cE. u, a$ s h w. l. QL x A A du h n .. +El s.-f a

4 b ..d

.cw A( ~M %

1 -L n.. s & p se~) nn-. .J~wanu . . . .-

- - e. + ,E b W *' ;,..

. . ~

. , 2 .

f. .,*Q

,._ ,. ghn ;

, w , n . v. i ' . -

- u ? . ' ..A .' *

-_ r<

. . , . - ....s.7. v ,4. - . -.2'..,.r ..: v . , #, ' %. .- ~e

,4. s . i 1" , 'AM..

y, . ' s i M. . l;t v&:Y 4 .,

,'M ' y?.

t > . f'

.s,- .L- T & [,* 7 '

.l l , r  %-

R :

4 e.t;.

,I,'  ? : .~r k'eg>m.c'-' ~g[ .

c

.': h. -

9 . . . < . . . - ...;.

. .g

.. e -.y . ',.~......

4 P, .x - ;.. z .

. . ., g.jp7 . , . . , . j. .-

9 - ,; ~.- ., W o.3 , - ~..

'.,s .p

.. ; ? g. -  ;

a g:y ' . .- .- ..

-[- ~ , ..,, [ l..

j RQ i ~ '. _ . . . . i .f..

w

.[

. g

p i

.'L Q}'

. ;4[; ;, j[ }' ps .; '. ' ! _ ' .] [ ,: *_: ._

^

/. -; . . p & - * ^ .O ? . ' T

.c l

,,.  ; d 1.5'f [ ' ' l. I ' ' - ..

] '- .

. ) '. ,' .-

.] I ~ E ,.,j ' N

,'lu- a I'<l ; 'a d., 'y .' f . .( '. j_; L ,'?' ^.;;.*' ' W 4 *,.,' ; . .l ,. i . * . . '; -  ; j ' ,, , - ~

MfC_,

. r, a ... d,' f .. , 7.- y'? , l . e., ' . ] 4.. lO V -

J ~t R$~l. { N n; j'f*%l'?; i4- % : ~. .- ,?'1 .va. ?. ,[. 4 _

c<.: 9<. ;:.- ! , . . z . .. ., - .. .

. . . ._ l .. F., ., Y

~; ,,' 3. y ; ;. ; .. , ' ,

I1. _ (, ..c : ; , .e. :..

.y .; ; , . . . ..; -

g ..:' _

M.,4-,,#'.'

..s . 4

• m- ' .-

f, .. .g,.

p .' . . .1

.. ,. # 4

- ,, Y */a.. , ,, , .

t 3  ;. . , ,_. g-, . . S- -

. V ;. ..* ,,

' e; ' .4 n *.

-,f .

.. '*. :R '.'

p y...-.

. - g' j>.,> '.. o ;;;nm3 ,.r'- ., m( _' .4;,

,p.

7 .g  : . p y . ., , '.;q,.e, . :, .u . ...^ ., . .x t % v j.

7. ~

.p..

, y y.y c. ,e

%..t. .

..~..

~,...

. . . . ,. ..- e ..

aa , s

._;T q' a 9;<:.s;;34n -

% & k.Q ~{} M. ,n.

EQ~. v7. si,-.i

.; . 7;. .

4  :

a - 9 < _.7.;.. f, + . .  :. +=

. . , c.,y 4: l- ' . . ..... -

'g.W,e.h. g, .?;: (. ;;: FK L ^ 2 ? " - , ~:L?' Wi? h M. ..

r

.,/.l. *-* i f I e D ,' rJ- P' s ,.,p !

8. Attachments: None. M(-. 'dp .q

,f. .

\ - -l

c. . t -

9. Reference Documents:

1. ANSI N45.2.6-1978 " Qualifications of Inspection, Examination, and Testing Personnel for Nuclear Power Plants."

2. CPSES Final Safety Analysis Report (FSAR) - Amendment 50, July 14, 1984.

3. XCP-EE-1 "Megger Hi/ Pot Testing" (Rev. 7 - June 4, 1982).

4. CP-EE-14 " Molded Case Circuit Breaker and Thermal Overload Relay / Heater Testing" (Rev. 9 - March 23, 1983).

5. CP-EE-24 " Fixed Battery Pack Operated Emergency' Lighting Units" (Rev. J - June 17,1983).

6. CP-SAP-1 "S .artc; A: mini:tra .ive Procedures Manual" (::ev. S-January 19, 1953).

7. CP-SAP-21 " Concoct of Testing" (Rev. 2 - March 7, 1954).

.a ,

'8.

Office Memorandum STM-83084 of March 31, 1983,

Subject:

ETG Personnel Schedule ~ Change. -

9. CP-SAP-11 " Review, Approval and Retention of Test Results".(Rev. 5 -

January 4,- 1983). ,

10. CP-SAP-19 " Training / Qualification Requirements for Startup Personnel" (Rev. 5 - May 6, 1983).

11. CP-SAP-10 "Startup Program Qua11ty Assurance Plan Implementation" (Rev.'2 - July 9, 1982).

12. CP-SAP-16 " Test Deficiency and Nonconformance Reporting" (Rev. 8

- December 7, 1983).

13. Regulatory Guide 1.58 " Qualification of Nuclear Power Plant .

Inspection, Examination, and Testing Personnel" (Rev. 1 - September 1980)

14. "Startup Prerequisite Test Instruction Manual" (a compilation of generic prerequisites).

15. Affidavit,of GAP, Witness H. .

.16. GAP handwritten notes by Billie Garde from conversations with Witness H.

17. Regulatory Guide 1.8, " Personnel Selection and Training."

18. ANSI N18.1-1971', " Selection and Training of Nuclear Power Plant Personnel." ,

e

4 s 6 qo , .

- ~. .

10. This statement prepared by:

Ward F. Smith Date TRT. Reviewer Reviewed by:

Richard R. Keimig Date Group Leader Approved by:

Vincent S. Noonan Date Project Director i

1

~

i i

1 r

l I

i

L' tT S +

Document Name: .

] )Q'gj SSER TEST PROGRAM 7 R uestor's ID: I d Author's Name:

Chet Posiusny Document Comments:  !

11/8/84 Final draft. rer.1

- l 7.!

46

/

/

/

[ . .

- /

F01A-85-59 I

g l

s

~

,. Final Draft, Rev. 1, 11/7/84 cp3/SSER AT-16 SSER

1. Allegation Category: Test Program No. 7

2. Allegation Number: AT-16

3. Characterization: It was alleged that TUEC upper-level management had a tendency to interpret their commitments to the Final Safety Analysis .

Report (FSAR) and to NRC Regulatory Guides (RGs) liberally rather than conservatively.

4. Assessment of Safety Significance: The implied safety significance of this allegation is that such tendencies could lead to plant construction at a standard'below that required by the NRC, which in turn could adversely affect public health and safety.

The prima tat 1sgati ppeared to b that the TUEC Start-up Group did not require craft personnel who support testing activities g oerfonn&ta11ation checkouts to be qualified to ANSI N45.2.6-1978.

The TRT review of that allegation, presented 'in Test Program Category 5 of this SSER, concluded that while qualifying those personnel to ANSI N45.2.6 may have reflected a conservative management attitude, TUEC did not commit to that level of qualification in the FSAR. ANSI'N45.2.6-1978, Section 1.2 leaves the imposition of its requirements to the discretica of the employer for personnel who perform work which is well within their normal craft expertise, e.g., calibration and installation checkouts.

Additionally, that work was performed under varying degrees of supervision by qualified System Test Engineers (STEs) who were held fully responsible for the correct performance of that work and for the review of data recorded.

In order to determine if there were any other bases for this allegation in the testing activities area, the TRT reviewed FSAR Chapter 14, which describes how the testing program is to be carried out, and the RGs to

1 which TUEC committed. These were compared with TUEC's Startup Administra-tive Procedures and Startup Quality Assurance Plan which prescribe, in detail, the conduct of the testing program. In addition, the TRT reviewed procedures related to the test program in Test Program Categories 1, 3, 4, 5, 6 and 7 of this SSER. With the exception of some minor deficiencies identified in Test Program Categories 1, 4 and 5, the TRT did not find' any substantive evidence that the Startup Group interpreted FSAR commit-

~

ments or RGs in a nonconservative manner.

The TRT found, however, that some of the decisions made by startup manage-ment may have appeared to be less than conservative. Through discussions with startup management personnel,.the TRT perceived this to be due to the heavy workload and schedule pressures inherent in a testing program of such magnitude. These burdens apparently resulted in decisions by startup management, in the interest of expediency, to delay some aspects of a particular test to a later date when the workload and impact on schedule would be lessened. The TRT found several examples of this with respect to' preoperational testing.

One such' example was the TUEC decision to conduct the Containment Integr-ated Leak Rate Test (CILRT) with three electrical penetrations isolated.

While it is technically reasonable to do that (as long as certain controls are maintained), it is preferred that this test be conducted with the

~

containment building as close as possible to the configuration it would be in during normal plant operation, i.e., with no penetrations isolated. An allegation concerning how the CILRT was conducted is discussed in detail in Test Program Category 4 of this SSER.

Another such example concerned pre-operational tests which were originally scheduled to be performed prior to fuel load, but for which TUEC was seeking NRC approval.to defer until after fuel load. The Hot Functional Test, in particular, is discussed in detail in Test Program Category No. 1 of this SSER. These decisions were apparently made because of schedule considerations and, while not the most conservative course of action, nevertheless, were acceptable from the point of plant safety.

-, * - -- ,- , , - - . , . , , - ., .-..n . , - , -

5. Conclusion and Staff Positiers: The TRT found no substantive reason to

• believe that TUEC startup management has a tendency to liberally interpret FSAR commitments and NRC RGs. As discussed above, startup management has' madedecisionswhich,theallegercould'havecol/s'trNIdaslessthancon-u servative. The TRT found the programs that TUEC had. implemented for con-

• ducting and inspecting preop'erational testing, as verified b,y NRC staff routine inspection of testing activities, were sufficiently comprehensive to reveal all safety significant or generic problems. Accordingly, this allegation has neither safety significance nor generic implications.

6. Actions Reouired: None.

l

8. Attachments: None.

9. Reference documents:

l

. 1. N05-103-1, " Guidance for Assessment and Distribution of Industry

. Operating Experience Reports",(Rev. 2, February 15,1984).

2. CPSES " Final Safety Analysis Report" (FSAR), Amendment 59, July 14, 1984.

( 3. Memorandum STM-83084 dated March 31, 1983 to all startup engineers, subject: ETG Personnel Schedule Change."

4. ICP-PT-75-02, !' Structural Integrity Test and Integrated Leak Test,"

(Revision 0). .

5. CP-SAP-1, " Conduct of Testing," (Rev. 2, March 7, 1984).

6. Affidavit of GAP witness ti

7. GAP handwritten notes by Billie Garde

s,. .

l. .

t

• b

10. This statement prepared'by: .

Ward F. Smith . Date .

p Reviewed by:

o *

-[f Richard R. Keimig Date  !

Group Leader r

h Approved by: '

Vincent 5. Noonan Date i Project Director a

- I i

j e.

I

, t 4 e i

w 4 ,

b 9

l I

i r

i

. _ . _ _ . . _ . - . _ _ _ . . . .. - - .- . . _ . _ . _ _ . , _ _ _ . m .. , _ _ _ _ . . . . . . . . . . _ . . . . ,_.__ _,.. .....__., .. . -._ ..__ ,_, -

s. * , ,I

\

Document Ha~:e:

)

SSER AT-15 Requestor's ID: vq ['p d Author's Name..

Chet Posiusny Dccument Ccmments- -

11/S/84 RedraftI,[caca '

Is i

/

/

/

~

,' = l

/

' e t

4 J ' '

s s '&

l ' ~'

, x I

g/

Q- s \ c, -

/ lb  % d\ - .

) / .L. -

L J.<~ \#

E'

1. 6

\

V

\ 'l et

> ,J .

n .a ,5 '

J G

( .

+

/

/ '.

l /

I '

h \ e' l

,/

}

/

7%-o'b-De s

,e Final Draft, Rev. 1, 11/7/84 SSER AT-15 SSER

1. Allecation Catecory: Test Progr.am No. 6

2. Alleoation Number: AT-15

3. Characterization: .It is alleged that the preoperational test program is flawed because (1) there is a dual numbering system and many system test engineers (STEs) for electrical / mechanical plant systems may test one system, or one STE may test a part of many systems, a condition causing overlap, confusion, and the possibility of omissions, (2) STEs are not provided with a " computer printout" which informs them of all tests renuired on a system, (3) calculations for instantaneous trip settings on approximately 100 breakers had not been performed correctly, (4) portions of prerequisite tests are being used to meet Final Safety Analysis Report (FSAR) commitments, (5) a system can pass through both the prerequisite test and the preoperational test without ever undergoing an energi:ed functional test; and, (6) STEs are.not provided with current design information and, therefore, must spend too much time researching and validating drawings.

4. Assessment of Safety Sicnificance: The implied safety significance of these allegations is that if safety-related systems described in the FSAR are not fully tested, there is no extra assurance that the systems will meet their intended safety functions.

.The TRT reviewed the " Prerequisite Testing" method u, sed by TUEC to ensure that the systems were ready for preoperational testing. Through a series of' generic tests, such as XCP-EE-1, "Megger Hi/ Pot Testing," XCP-EE-3,

" Control Circuit Functional Testing," or XCP-ME-1, " Initial Pump Operation," TUEC verifie: that construction was co.7pleted as reouired for the precperational test ;rogram. The prerequisite tests ccnsisted of such

~

items as initial instrument calibration, system piping flushes and cleaning,' wiring cor.tinuity and separation cnecks, hydrostatic pressure

Test Program tio. 6 tests, and initial functional tests of components. These tests facilitate the safe and orderly progression to the preoperational test program, as outlined in Regulatory Guide 1.68, and as committed to in the CPSES FSAR.

There are additional comments on the conduct of prerequisite testing in Test Progran Cz egory 5 of this SSER.

The TRT discussed TUEC's process for STE plant system assignments during an interview with startup management personnel. Each STE was assigned to at least one system, although the more complex systems had two or three -

STEs assigned, with one STE designated as the leader. These assignments were documented on a " System Assignments" sheet, which the STEs used to keep track of who was responsible for each system.

The TRT found no evidence that indicated confusion, gaps, or overlaps between systems and subsystems because of STE assi,gnments. As documented in CP-SAP-2 "Startuo Program Organization and Responsibilities" and CP-SAP-21, " Conduct of Testing," the STE was responsible for en.suring that assigned systems were properly tested and that the tests were coordinated with adjacent systems tests. The STEs were responsible for cooperating with other STEs when the. scope of testing overlapped into subsystems under their responsibility. The " subsystem definition packages" provided to all STEs facilitated this cooperation. The same applied to cases where startup work authorizations (SWAs) applied to more than one system. The master data base, a multi-functional computerized tracking system, showed outstanding work or deficiencies on all subsystems by code. The STE had the resources and the training to keep the coordination between STEs and plant systems under control.

The alleger also implied that tn'e numbering system used to identify sub-systems was a " dual" system, that is, the same component or terminal could aopear on two adjacent systems if it were on a boundary, thereby causing confusion. The IRT determined that it was not unusual for tnis to ha;;ec,  ;

but if a work item was generated on such a component and it appeared on  :

1

ne master data base under a system assigned to an STE wno did not nave '

responsioility for tnat component, it would be necessary for that STE to  ;

i i

- e Test Program No. 6

(

cooperate with the STE responsible for that component to get the work item  ;

closed. This degree of cooperation was common among STEs and did not appear to be a source of problems. The TRT found no evidence that indicated confusion, gaps (missed tests), or overlaps between systems and ,

., ,subsystems because of the " dual numbering system."

As alleged, STEs were not provided with a computer printout that informed i

them of what tests were required on a system. The STE was responsible for '

determining what testing was required based on design specifications, -

drawings, the FSAR, and other applicable documents. When the STE, the group leader, and the Joint Test Group (JTG) agreed on the testing require- t j ments, a computerized test index was published to track the number, name, and status cf all tests. The index was made available to all STEs and.was

, reviewed by the TRT.

i i

1

In its investigation of the allegation related to incorrect instantaneous ,

breaker trip settings, the TRT found that prior to September 1980, a total j of 74 molded-case circuit breaker instantaneous trip settings were adjusted to the specific values required by.the design drawings. None of these

breakers were safety related. Early in prerequisite testing, a few "nuis-

_ ance" trips were encountered, i.e., some breakers instantaneously tripped

at iccked rotor current values, which was not proper. Startup test engineering correspondence and discussions with Gibbs & Hill, Inc. (G&H),

ne design architect-engineer, resulted in a revision to XCP-EE-14,  ;

i "Volded Case Circuf: Breaker and Thermal Overload Relay / Heater Testing,"

j which is the generic ;rocedure for determining, testing and reccrding .

these settings. The revisien, which changed the method of determining the 1'

trip settings, included consfderation of the motor starting KVA, horse-power, voltage, and full-load current which were not- accounted for in the

• previous revision, and therefore, this revision was more accurate for a particular meter. The IP.T reviewed the speciffe data associated with the P

i trip-setting calculat'cci alleged tc be erroneous, which were foca d by tre e'ectrical test gr:u: cr or about March 15. 1934. Tne data of concern involved 21 creakers tnat were initially set in accordarce with GiH cesign  !

drawings. While cec:a-ing ne instantar.eous trto settings from GiH with 1

m __ _ - _ _ __ _ _ . __ . _ , _ _ _ _ _ _ _ _ , -. ._ - _

Test Program No. 6 settings calculated in accordance with the revised version of XCP-EE-14, the TRT found that, on the average, the calculated values varied by plus or minus 10 percent. During interviews with startup engineering personnel the TRT learned that TUEC considered this variance to be insignificant when it was idetified because (1) the XCP-EE-14 ' calculation method was placed in effect subsequent to using G&H values; (2) the eculpment was not safety related; (3) the variance resulting from calculating with the newer XCP-EE-14 method was minor; and, (4) no nuisance instantaneous trips were occurring on breakers which remained at the previous settings. Tha equipment involved included the turbine building roof exhaust fan, cir-culating water traveling screen, polymer mixer, and other similar compo-nents that are not safety related.

The allegation also implied that because some prerequisite tests were not repeated as part of the preoperational test, porti,ons of prerequisite tests were being used to prove FSAR commitments. FSAR Section 14.2.1 states that prerequisite testing is one of the three major phases of the testing activity at CPSES (the other two are preoperational and initial startup testing). This section also states that prerequisite testing will be con-ducted in order to verify the integrity, proper installation, cleanliness, and initial functional operability of components.

The' alleger stated that a " system" can pass through both prerequisite and-preoperational test programs without ever undergoing an energized func-tional test. The TRT found that there were cases where scme circuits, such as those for lights indicating valve and breaker positions, may only have had continuity checks during prerequisite testing without having been specifically checked during preoperational testing. However, the TRT learned in interviews with start'up personnel that even when such circuits were not specifically tested, the circuits would be verified during the preoperational testing of the components they served. Eor example, when a given preccerati crai tes. recuires a remote-operatec valve to ce ccenet, the operator would expect to see a change in the indicating light. If nis'did not hapoen, the operator would indicate the abnormality to the

i Test Program No. 6 STE so that corrective action could be taken. Most of the time when this happended, it was a burned out bulb, which the operator replaced on the spot. If- not, the STE documented it for resolution. .The TRT considered this approach to be reasonable. .

/ It is alleged that system drawing packages were being delivered to the STEs with design change authorizations (DCAs) several years old that were not updated on the design drawings; that packages were being received by

-j ,'.- STEs with DCAs issued against other packages; that print changes were -

p'SC ' ,f . arriving with no DCAs in the packages; and, that there was no procedure to i

~-

ensure that the STE had the proper documentation to conduct a valid test.

The TRT interviewed three STEs who were responsible for major fluid and electrical systems at CPSES. At each interview the STEs commented that the substance of the allegation relative to oup;Laj;ad_dehd_rawings was correcMn the__- gait, but that some improvements were since made. That portion of the allegation dealing with the lack of procedures could not be substantiated because those interviewed insisted that there were always procedures to ensure the STE had proper documentation. CP-SAP-21, " Con-duct of Testing," Section 4.9 required the STE to use current information when he prepared to conduct a test. Thus, the responsibility was placed on the STE to ensure he was working with current documentation. To accom-plish this he was required to go to the document control center and update the documents. This was very time-consuming and burdensome. The STEs who.

were interviewed told the TRT that after much discussion with TUEC manage-ment the documents given to STEs had improved greatly and, at the present time, only the "less important" systems, such as vents and floor drains, are causing the STEs a problem. Even though the STEs are now being issued current drawings in a controlled manner, CCAs are not being promptly transmitted to them econ _ issue. Therefore, the STEs have more current

drawings at present, but they still must go to the document control center for DCA updates.

1 S. Cenclusion and Staff Positions: The TRT found no deviations or violations of NRC recuirements regarding the alleger's concerns over the prerequisite i ~

ud preoperational test programs at CPSES, even though some of :ne aliega-

1. cns were substantiated.

Test Program No. 6 s in-With regard to the specifig characterizedpparagraph 3 of this section of the SSER, the following conclusions were reached:

(1) The " dual numbering" system and STE assignment methods did not cause confusion,. gaps or overlaps in testing. The STEs appeared to be in full control of the systems for which they were being held responsible.

2.

1 (2) The STEs had sufficient information to determine what tests were -

required for their assigned systems and were sufficiently trained to make that determination. In addition, TUEC pointed out that there were backup reviews performed by startup personnel of greater experience and training, as well as the ultimate JTG reviews.

(3) Calculations, where required, were performed properly for the instantaneous trip settings on molded case circuit breakers, and the disparities found by the technician were acceptable.

(4) Portions of prerequisite tests were not being used to satisfy FSAR commitments other than addressed in the FSAR, but, rather ace- db:E W A'f s- A 'W t assur{$4 j the success of subsequent preoperational tests which in ?b i 'pf m satisfy FSAR commitments.

, 1 (5) Though all parts of all systems may not be incrementally and specifi-cally subjected to an energized functional test, the preoperational test program subjects all systems committed in the FSAR to an ener- '

gized operating condition as a minimum such that failures would be I detected. This is an acceptable approach.

U (6) STEs are now provided with current design information by virtue of their access to'the document contro,1 center for updates, however the day-to-day changes are not delivered to the STE in nis place cf work.

4 Ne problems apcear to have surfaced as a result of the STEs having to pursue cesign information updates on their own initiative, however i

k the TRT ::nsiders the intent cf 10.CFR 50, Acpendix E, Criterice !!

i, i i was to have this information celivered to the STE's workola:e.

s

Test Program No. 6 The TRT concluded that these allbgations have neither safety significance i nor generic implications.

6. Actions Recuired: TUEC shall establish measures to provide greater assur-

nce'that STEs and other responsible personnel are provided with current controlled design documents and change notices.

l f

8. Attachments: ?!cne.

9. Reference Documents:

1. XCP-EE-1, "Megger Hi/ Pot Testing" (Rev. 7 - June 4,1982).

2. XCP-EE-14, " Molded Case Circuit Ereaker and Thermal Overicad Relay / Heater Testing" (Rev. 9 - March 23, 1983). Also Rev. O, Rev. 1, and Rev. 2. f e

3. XCP-ME-1, " Initial Pump Operations" (Rev. 6 - March 1, 1983).

4. XCP-EE-8, " Control Circuit Functional Testing," (Rev. 6 -

February 22, 1983).

5. System Assignments, April 25, 1984.  ;

2 ,

i

6. CPSES Final Safety Analysis Report (FSAR) - Amendment 50, i July 14, 1984

7. Regulatory Guide 1.68, Revisicn 2, August 1978, " Initial Test Programs for Water-Ccoled ?!uclear Power Plants."

8. CP-SAP-2, "Startup Pecgram Organization and Responsibilities" (Rev. 5 - November 1, 1983).

-y:-:'., "::.. ;  : ie s t n " ( c e. .. -

- .* ; .. y. , , 9 : .- )

a Test Frogram No. 6 ,

10. CP-SAP-19, " Train (ng/ Qualification Requirements for Startup Personnel" (Rev. 5 - May G, 1983). '

11. AT/PT Test Index with Schedule, Unit 1 and Common (July 2,1984).

12. Affidavit of GAP witness H I

13. Handwritten notes taken by GAP (Billie Garde) from conversations  ;

with witness R, and Witness A.

! 14. Completed XCP-EE-14 test data sheets (quantity 74), [

{ 15. Owg. 2323-El-0010, 0013, 0014 & 0015 " Normal 480V MCCS One Line Diagrams  !

I i

! i

10. This statement prepared by: ^

Ward F. Smith Date t TRT Reviewer l -

t Reviewed by:

Richard R. Keimig Date .

', Group Leader i

1 Approved by: l j Vincent S. Noonan ~ Date  :

Pr'oject Director q

)

i j .

- . I r

.,, < . , , ,, .,,.y --.y, , . - , , , v.o,p._.- g , .u 9 3. wg,,y, , _ . . ,w.,,-,. ,, -. v- ..e ~ , , .

---

e.--

, ,o m-- %--,,%-,_,cm,.,,,,. ,f,_,.

7. z __ s u _ _ . __ _._ _ _- - . . - . _ . - _ . - - -

1. ' -- D\c.,dfa.<VdW (unpecded-h(Ok%4 b

Draft 2

1. AilegationGrouo: Protective Coating 1- Backfit Test Program '

.i

2. All'egation Number: Parts of AQ0-18,19, 20, AQO-21, 37, 46, 47, 54, 55,

56. .

3. Characterization: It is alleged that:

~ -

- Visual defects were not identified during backfit inspections. (AQG-1d) -

- Backfit inspection procedures are vague. (AQ0-19)

- Adhesien_,teating of the protective ccatings was not performed properly.

.(AQ0-20)

~

- Adhesion test data were not corrected for calibr6tien error. (AQO-21)

- Area stated to have satisfactory documentation, in fact, had primer coatings exceeding the allcwed thickness. (AQO-37a)

- - The maps fer the Backfit Program were incorrect. (AQO-37b)

- Dccumentaticn'for the Backfit Program was forged and falsified. (AQ0-37c)

- CC Inspectors completed Inspection Reports (IR) without perfcrming the inspections. (AGO-37d)

- Unacceptable substrate conditions were observed thrcugh Tocke Gauge tests.

at " g)

H seg

}

1

-. . - - . . . . - - . . . - . ... .. .. - . . ~. - . . . ... .. : = .. - . .. -

l 3, l , D.

ll

~

i -

Adhesion test dollies used during the Backfit Program were observed to have rust adhering to the paint (underside) at the completion of. the test.

(AQ0-46) s: .

25 adhesion tests were performed in violation of written instructions. .i;i.

(AQ0-47) ._

During the Backfit Program, only the first unsatisfactory reading was recorded, adversely affecting the trend analysis. (AQO-54)

Areas identified during the Backfit Program requiring coatings removal did'

~

not have the coatings removed. (AQO-55)

Original documentation for the Backfit Program was destroyed by QC manage-ment...IAQ0-56) -

4. Assessment ~of Safety Significance: '

In order to assess the individual allegations characterized above, the TRT

, reviewed the background and scope of the Backfit Test Program and j independently evaluated the program test as presented belcw. The TRT i assessment of the individual allegations follows in Section 4.d below.

a. Backfit Test Program Background and Scoce

1)

Background:

In 1981, Region IV of the NRC conducted an inspection of protective coatings at Comanche Peak. As a result of this inspecticn (Inspection Report 81-15), a Notice of Violation was issued regarding the failure of TUEC to follow quality assurance program procedures for the inspection of protective coatings. Specifically, from late Spetember 1979 l through October 1981, documentation for protective coatings inspections was not maintained or was inccmplete. In response to this Notice of Violation, 1 l

l l

! l l l 1

1 -

~_

' ^ ~

.._ _ ~___ _ . _ _ _ _ _ . - -- - - - - - -

t l --

~ . .

I

~

t TUEC, in their letter of November 19, 1981 to NRC instituted a Backfit Test

! Program (BTP) as described below and the cited discrepancies were documented as.

j nonconforming conditions. TUEC proposed a complete review of existing records ,

and reinspection, using destructive testing, of coated areas for which ,

documentation was missing or discrepant.

.j The re-inspections were to be based on a statistically sound sampling plan using both dry film thickness tests (Tooke Gauge tests) and adhesion tests (E1cometer tests) to evaluate the condition of the applied coatings.

Discrepant areas, if any, were to be clearly identified and corrected in '

accordance with approved procedures. -

On January 19, 1982, the NRC responded to TUEC's November 19, 1981 letter The NRC had no questions at that time and informed TUEC that they would review the corrective'ac.tions during a future inspection. ,

P In January 1984, NRC Region IV contracted with'Brookhaven National Laboratory (BNL)toprovidetechnicalassistanceinperformingoisitereviewsofandin commencing technical evaluation of allegations of deficiencies related to the

, protective coatings program at Comanche Peak. As part of BNL's review work at Comanche Peak, the NRC requested that BNL perform some independent testing of the protective coatings, which is discussed later in this report.

On Apr'il 25, 1984, BNL sent to Region IV an Interim Report on protective coatings. This report a) requested information frcm TUEC regarding the BTP, b) presented BNL's independent test results, and c) reported BNL's interim findings on TUEC's protective coatings. procedures and documentation.

On June 13, 1984, BNL sent to Region IV a draft Status Report on Protective Coatings. Allegations. This report provided the status of SNL's investigation up to that time of sixty allegations regarding protecti've coatings at Comanche Peek plant.

e i

^

i; o

'! On July 9, 1984, che NRC Comanche Peak Protective Coatings Technical Review

Team (TRT) assumed responsibility for completing the investigation of. .

protective ccatings allegations, The TRT included personnel who participated in the BNL ' review effort.

.4

2) Scooe: The BTP was to cover liner plate, concrete and miscellaneous steel already coated for which documentation was missing or discrepant. The Tooke Gauge tests were to measure the dry film thickness (DFT) of the primer and topcoat. The Elcometer adhesion test results were to show that the protective coatings had adequate adhesion to the substrate. TUGC0 selected a 200 psi pull

~ '

criterion which is the same as that presented in ANSI Standard N5.12-1974, -

" Protective Coatings (paints) for the Nuclear Industry, June 20, 1974, Section 6, Physical Properties Tests." It was TUGCO's intent that DFT's and adhesion tests were to-be used in lieu of the missing and discrepant documentation.

_ _ , ~

. It is the TRT's engineering judgment that if DBA-qualified paint was applied and its traceability was maintained, then the BTP, reperly adninistered, could provideuseful,butindirectinformationonthequalilyanddesignbasis accident (CBA) survivability of the coating work having missing or discrepant documentation. The DFT's could demonstrate that the protective coatings were applied in the same thickr : ranges (for primer and topcoat) as the DBA qualification tested samples. Adhesion tests provide data which indirectly demonstrate the adequacy of surface preparation, in that the primary p'urpose of surface preparation is to provide good adhesion of the coatings. The adhesion tests also provide seme assurance that application and curing of the coatings was adequate to produce satisfactory coating film integrity and internal strength. The adhesion criterion of 200 psi is appropriate because it is the criterion that applies to physical properties testing of DSA qualification test samples in accordance with ANSI N5.12.

e 4

t 1

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

l' l - --

[ -

} -

~

j 4. b. Evaluation of the Backfit Procram Test (BTP) Results (1) General .

'I ,

. In examining the results of the BTP inspections and tests, the technical 41

. t review team (TRT) paid particular attention to the adhesion test results because they provided the most direct' indication of coating adherence tc containment surfaces under accident conditions, which is the primary safety concern related to coatings.

Adhesion testing is done with a device called an Elcometer, which measuref '

the ferce required to pull a protective-coating off the coated surface.

The Elecmeter model used by Texas Utilities Generating Company (TUGCO) inspectors for the BTP had a total range of 0 to 1,000 psi and could be read in.,the. field with a precision of approximately 50 psi. The Elcometer '

, readings tended to read high after repeated use; therefore, they were periodically calibrated by dead-weight tes' ting in-the onsite Brown and

~

Root instrument shop.

By private communications from L. Bielfeldt of TUEC to D. Lurie and L. R.

Abramson of NRC dated March 29, April 17 and April 23, 1984, partial results of the adhesion tests and dry film thickness tests on the containment steel liner, the concrete surfaces and the surfaces of miscella.neous steel components in Unit I were reported. The failure rates (Elecceter readings below 200 psi) e ue very low. For the containment liner, only 2 out of 405 doll .% fei' :d. For the concrete surfacss, there were no failures in 1,691 ro d 4 'cr the miscellaneous steel, there were 20 failures in 1,517 readings. Un the basis of these low failure

rates, TUEC by memorandum to distribution from R. G. Tolson on February 10, 1984 discontinued all routine destructive testing (adhesion tests and Tooke gauge coating thickness tests). -

LE _ ' _ .. . i. .- ... . .,. ___ . - __- _ _ _ . _ _ . , - .- m - - - 1

~

o 1

~-

l '

[, In a letter to the NRC dated February 15, 1984, TUEC referred to a verbal j notification on January 16, 1984 of Mr. 'R. G. Taylor of the .NRC .of .a j . " deficiency regarding an error in the tolerances used in the calibration

[ of the' adhesiot1 tester." At the July 11, 1984 site meeting, TUEC briefed ,

the TRT on the overall scope of the Coating Backfit Program. R. Tolson .j

~

(TV'EC) informed the team of the discrepancy in calibrating Elcome'ters used for the coating adhesion test tilat wa's discovered after most of the BTP adhesion tests were ccepleted. This discrepancy would allow in-plant test results to be in error by as much as 200' psi in the non-conservative direction. Thus, any Elcometer reading less than 400 psi represented a '

potentially failed area. ~ '

After learning of this deficiency, the NRC by attachment to the NRC meeting notice memorandum dated July 27, 1984, requested that TUEC provide the TR,T,with. corrected adhesion test data and analyses of failure rate for the containment liner, the concrete, and miscellaneous steel. The corrections to the Elcometer readings could be made from calibration data

~

available in the instrument shop for each Elcometer for each date' on which it was checked. -

Part of this information was provided by TUEC by a letter to NRC Region IV dated July 16, 1984, prepared in response'to a previous request in NRC's letter of May 23, 1984. This package contained the calibration data for each Elcometer and presumably all of the inspection reports (PCRs) for the backfit progre.m. TUEC's transmittal letter stated that the package contained results for 869 adhesive tests for the liner, 2,128 tests for the concrete and 4,714 tests for the miscellaneous steel. The PCRs

recorded the original uncorrected adhesion data and the dry film thickness i

data.

1 For the containment liner coatings, a list of PCRs with adhesion test readings belcw 200 psi af'ter correction and an analysis of failure rate were transmitted to the NRC by TUEC letter dated August 14, 1984 including 1

d L ,_ _ . . _ _ _ _ _ . . . . . - _ . . _ _ _ . . _ = _ _ . _ _ _ _ - - ---

., interoffice memorandum frem R. C. Levine to R. G. Tolson dated August 10,

.l 1984. For the concrete surfaces in containment, another TUEC interoffice a

memorandum from R. G. Tolson to file dated September 10, 1984 provided a list of PCRs with adhesion test readings below 200 psi after correction , ,,

i and an estimate of failure rate. Ae (2) Liner Plate Test Results

-(a) For the coatings on the containment liner, TUEC found that 51 out of

~

869 adh'esion test readings, or 5.9%, were below 200 psi after '

correction for calibration error. The failed area, calculated by -

summing the areas corresponding to each failed reading was 5,148 ft2 ,

or 3.5% of the total liner surface, estimated by TUEC as 145,088 ft 2,,

~

The TRT believes the liner surface estimate is too high. Considering the liner surface as a cylinder (diameter 135 feet, height 192 feet) topped by a hemisphere (diameter 135 feet),'the surface area's is 2

110,000 ft . The TUEC personnel interviewed by the TRT were not able to explain the high value of 145,000 ft2 used for the coated liner areas. ,

Also, TUEC estimated that 95% of the total liner surface was backfit tested. Based upon its review of the liner surface backfit test map, TRT finds the 96% esticate to be reasonable, so that the liner surface tested for adhesion would be 106,000 2ft . Using this area, 4.8% instead of 3.5% of of the tested area of the liner failed the -

~'~

adhesion test.

l The TRT independently assessed' the liner failure rate from the j original adhesion data after correcting the readings according to the Elecmeter calibratic'n data from the instrument shop. After correction, the TRT found 36 adhesion test readings below 200 psi out a

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

n ._,- 2 _. , _ . ._.

l' ....~:.- .. - . . . . . . . . . . . . - . . - . . - . . . - . - . . . . -

~

i l of a total of 834 adhesion test readings, for a test failure rat'e of i 4.3%. The failed area, calculated *by sa :ning the areas corresponding.

to each failed reading, was 3,092 ft 2, l

The package of PCRs delivered by TUEC contained only 278 PCRs, with 3 .4

' adhesion readings recorded on each PCR for a total of 834 ad'hesion test readings. The TUEC transmittal letter of August 14, 1984 for the package and the enclosed memorandum of August 10, 1984, referred to 869 adhesion test readings. This corresponds to about 290 PCRs assuming that the usual 3 readings were reported on each PCR. '

However, a total of 339 areas on the containment liner map provided - - .

by TUEC are labeled with different PCR numbers. Apparently not all ef.the liner PCRs were included in the package delivered to the TP.T nor. in the group discussed in the TUEC August 10, 1984 memorandum.

Siv,eyALTUEC personnel interviewed by the TRT were not able to account for the discrepancy and' agreed that t'he 339 areas on the liner map represent the total number of PCRs.

On the assumptien that the average liner area per PCR was the same for the missing PCRs as for those delivered to the TRT, the adhesion-tested area is 106,000 ft2 multiplied by 278/339, or 87,000 ft2. Using this area, the TRT finds that 3.6% of,the tested area of the liner failed the adhesion test. The TRT assumes'the same failure rate for the liner area represented by the missing PCRs.

When a similar correction is applied to the area failure rate calculated by TUEC the TUEC failure rate becomes 5.6% of the tested area. To account for the discrepancy between the number of failures

[

b found by the TRT (36) and by TUEC (51), the methods of correcting the original adhesion data for calibration error were examined. The TRT  !

determined that TUEC was more conservative (used larger corrections)

{

in correcting for calibration error so that some adhesion. test readings which were 'below 200 psi after the TUEC correction were i

! l l

i

! j

....,_..,.a .

.: m.-.  : - ... ... ..r....-,4 . _ . . . , . . ,_ _ _ . _ _1

- y - - - - - - - - - - , . _ _ _ __

i l.

! above 200 psi after the TRT corrections. This accounted for mos't of

{ the discrepancy. .

. ~

b. Dry Film Thickness (DFT) Testing  ;

'?

. A second parameter of importance in determining the quality of liner coatings for the BTP is the dry film thickness (DFT) of the primer coat and of the total system. These thicknes'ses are measured with a '

Tooke gauge. With this device, a sharp V-shaped cut through the coatings to the substrate is made, after which the edges of tne cut

• are examined with the optics of the instrument to determine ~ the -

minimum, maximum and average thicknesses of the primer and of the total coating system.

The_prf 'inal JUGC0 backfit inspection procedure for protective -

coatings on steel, QI-QP-11.4-23 issued on November 19, 1981 specified the following acceptable thickness. ranges:

Single Average of 5 Coating Reading, Mils Readings, Mils Carboline CZ-11 1.5 - 5.5 2.0 - 4.5 Primer ,

Ameron D6 1.5 - 5.5 2.0 - 5.0 Total System -

Primer & Phenoline 305 topcoat 7.0 -11.5 7.0 -11.0 A majority of the liner PCRs reported at least one DFT value outside

. the original specifications given above. Subsequent revisions of the procedures broadened the acceptable thickness ranges. The latest version, Revision 13 dated April 18, 1984 lists the following ranges:

O I,

I D

.mi= w.. .m ..m, --- -.#_ m_ _ m_.

- - - -- . ..- - - .-._ . : .2 . . . - . . . . - - . - . . . . . . .. - .

j . .

. .'i .

~

. Single Average of 5 j Coating Reading, Mils . Readings, Mils Carboline CZ-11 1.5 - 7.0 1.5 - 7.0

?

j Primers - Ameron 06- 1.5 - 7.0 1.5 - 7.0 .  ?.

l!

Carboline 191 1.5 - 7.0 1.5 - 7.0 i

~

Total System - From full " hiding" From full " hiding" Primer & Phenoline 305 by topcoat to less by topcoat to less -

Topcoat than 15.0 than 13.0 TRT technical concerns involved in enlarging the acceptable range of

• thicknesses as related to DBA qualifications of coatings are .

discussed in SSER-Coatings 2.

Even after the DFT specifications were broadened, many of the liner PCR,s A ntained one or more DFTs outside the allowable range, most often on the low side for total' system (primer plus topcoat). As a consequence, according to interviews with TUEC QC perscnnel, nearly

~

all of the liner coatings in place at the start of the backfit program wer.e reworked. TRT review of records of the disposition of NCR's document rework for many liner areas. The repairs were perfomed in accordance with the approved TUGC0 repair procedure (QI-QP-11.4-23) and re-inspected.

A few small unrepaired liner areas totaling about 110 ft 2were placed in the Coatings Exempt Log (Items.8 to 18). The protective coatings exempt log is reviewed in coatings SSER 6.

In sum, after completion of repairs, under 1% of the containment liner coatings do not meet the DFT specifications of TUGC0 procedure QI-QP-11.4-23, Revision 13.

l l

i I

i _. . . . . .___ _ -_ _ ,_. _ .. .___,_..m. ,..- _ . _ . - . . .-.._u.-. . ._-_j

. . - . - - . - . - ..-.: L . _ . . . - . . . - . . . . . .. - - . -.- - .

-l

!l --

'i -

(3) Concrete Test Results 1 (a) Adhesion Testing

} -

. l.

As stated in Section 1 above, TUEC provided the results of 2,128 At

. adhesion tests on concrete coatings (634 PCRs recording 1 to 5 adhesion tests on each) in their letter dated July 16, 1984. TUEC stated that the 2,128 tests represented approximately 50% of the concrete surfaces in containment. The concrete surfaces not subjected to backfit inspection were (1) areas not coated at the time '

of backfit inspection; (2) areas inaccessible to test equipment; and" '

(3) areas not inspected due to termiriation of the test program on February 10, 1984.

A,s y s mentioned in Section 1 above, the initial TUEC reports on the adhesion tests on concrete indicated no failures in 1,691 tests.

When the Elcometer readings recorded on the'PCRs were corrected for

~

calibration error, TUEC by memorandum from ft. G. Tolson to File dated September 10, 1984, reported 65 adhesion test readings out of 2,128 tests with values below 200 psi, corresponding to a failure rate of 3.1%. TUEC did not provide a calculation of total concrete area with coatings failing the adhesion test nor a calculation of 1!he percentage of concrete area whose coatings failed the adhesion test.

The concrete surfaces, because of their more complicated gecmetry, are less amenable to accurate determination of the tested surface area than the containment liner surfaces. Since the test procedure QI-QP-11.4-24, called for approximately one adhesion test and one set 2

of DFT readings per 100 ft of concrete surface, the failure rate in terms of area can be approximated by the test failure rate.

f

}

Because of time constraints, the TRT did not conduct a complete independent analysis'of the massive amount of adhesion data for the i concrete coatings as was done for the liner coatings. Instead, the TRT elected to restrict its audit of'the adhesion data to the ,

i

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

.__ _. _ _ l. _ L _ _ __ _ _ . -

l-4 1-1 I coatings on the interior surfaces of the concrete compartments f surrounding Steam Generators #1 and #4. These surfaces were selected in part because of their proximity to the Unit I containment sump

~

l screens. ,

. f

-Areas of the interior walls of Steam Generator compartments #1 and #4 which had been backfit-inspected were delineated and labeled with PCR numbers on TUGC0 drawings of these surfaces (Drawings PCRM-018A1 and PCRM-01881 for compartment #1; drawings PCRM-019Al and PCRM-01981 for compartment #4). From the data recorded on each of these PCRs (20 in '

~ '

compartment #1 and 11 in compartment #4), the date of testing, --

Elcometer readings, Elcometer used, and the area tested were tabulated by the TRT. For the concrete adhesion tests, the number of adhesion tests per PCR varied from 2 to 5, with approximately one t a bper 100 ft 2of sampled area. The TRT corrected the original

, adhesion test results for calibration error from the calibration data provided by the instrument shop. -

After correction, the TRT found 8 adhesion' tests reading 200 psi or~

lower out of a total of 116 tests for a test fail re rate of'6.8%.

Three of the 8 failures were within 10 psi of 200 psi. If these are not counted as failures, the test failure rate would be 4.3%, in better agreement with TUEC's estimate of 3.1%.

The total failed area, calculated by summing the areas corresponding 2

to the failed tests, is 745 ft . The TRT assumed in calculating this area and the total tested area that each recorded set of DFT tests corresponded to 100 ft2 . The total tested area represented by the 31 PCRs is approximately 11,000 ft2 . The TRT therefore finds that 6.8%

of the area of protective coatings on the interior concrete surfaces of Steam Generator ccmpartments #1 and #4 failed the adhesion test.

If the three borderline failures were neglected, the failure rate by area would be 4.0%. Based on TUEC's estimate of 285,000 2ft for

total concrete surface area, the TRT 6.8% failure rate corresponds to

?

O

I *- ,

l -

2 a failed concrete coated area of 19,400 ft . Whereas the TUEC 3.1%

failure rate corresponds to 8,800 ft 2, ,

The TUEC letter of July 16, 1984 stated that approximately 50% of the. ,,

concrete area was backfit inspected. From the drawings of Steam

, .f Generator compartments #1 and #4, the TRT estimated that the total -

area of the internal surfaces was approximately 20,200 ft2 . The backfit-tested area was approximately 11,000 ft2 , or approximately 55% of the total area. This is consistent with the TUEC estimate that 50% of the total concrete area in containment was backfitted.

b. DFT Testing of Concrete Coatings The original TUGC0 backfit inspection procedure for protective coatinss on-concrete, QI-QP-11.4-24, Revision 0, issued on ,

February 5, 1982 specified the acceptable coating thickness range for Reactic 1201 topcoat on concrete as a minimum of 3 mils and a maximum of 12 mils. As required by procedure, five scratches were made with the Tooke DFT tester spaced randomly over each 100 ft2 of.the sampled concrete area. A single reading is selected as representative of coating thickness of each scratch. The " minimum" recorded on the PCR is the lowest of the five readings. The " maximum" is the highest.

The average recorded is the average of the five. The permissible

" maximum" thickness limit was expanded to 16 mils in Revision 3 of the procedure on June 29, 1982.

By private communication dated April 17, 1984 from L. Bielfeldt of

• TUEC to D. Lurie and L. R. Abramson of NRC, TUGC0 reported that 101

. recorded DFT readings on concrete coatings failed to meet the thickness specifications given above, out of a total of 4,623 recorded DFT readings, for a test failure rat'e of 2.2%. Most of the failures were for low topcoat thickness, i

-i i

.- . .. . _ .. . ___.. _ _..__ . . _ _ _ . -___ ._. - - _ _ . . . . - ~_

,. . I

' ~ - l

. l

, J

]; In a manner similar to that for the adhesion tests on concrete, the TRT restricted its audit of the backfit DFT tests to the coatings on the interior concrete surfaces of Steam Generator ccmpartments #1 and

f4. The TRT examined the DFT data recorded on the 31 PCRs for these , ,,

surfaces, finding 10 recorded DFT readings which failed cut of a .)

total of 297 recorded DFT readings, for a failure rate of 3.3%._ Only one of the failed tests exceeded the allowable thickness; the remain'ing failures had thicknesses below the allowable 3 mils minimum.

It is not clear why the DFT failure rate in the Steam Generator ~ ~

compartments #1 and #4 is greater than the overall DFT failure rate for concrete reported by TUGCO. In any case, according to statements by TUEC QC perscnnel and TRT inspection of the disposition of NCRs f_or.r.any. concrete areas, nearly all of the concrete areas with failed

, DFTs were repaired according to TUGC0 procedure QF-QP-11.4-24 until

. the DFTs were satisfactory. -

The principal exceptions were the coatings on the concrete surfaces of the reactor cavity (3,135 ft 2) and the coatings en the interior of the elevator enclosure (2,700 ft2 ). These areas were placed in the Coatings Exempt Log. The bulk of the concrete coatings .in containment either had satisfactory DFTs on first backfit in'spection or were repaired until they passed the DFT test.

4. Miscellaneous Steel Test Results j a. Adhesion Testing As stated in Section 1 above, TUEC provided the results of 4,714  ;

l adhesion tests on miscellanecus steel coating's (2,189 PCRs recording '

I to 3 adhesion tests on each) in the package transmitted by TUEC

letter dated July 16, 1984. TUEC stated that the 4,714 adhesion i

i l l

l 7

~

- ..x-.-..--.L-..' . . . . . - . .= - . - . - ------ - ---

=

1' j-

~

j - --

i tests represented approximately 22% of the coated miscellaneous steel

.i J surfaces in containment. The miscellaneous steel category includes items such as pipe supports, cable tray supports and conduit

'i supports.. The surfaces not subjected to the backfit inspection were  ;,

.j (1) areas not coated at the time of backfit inspection, (2) areas .

inaccessible to tert equipment, and (3) areas not inspected due to _

temination of the test program on February 10, 1984.

The TUEC letter of July 16, 1984 inificated 26 failures out of the i 4,714 adhesion test readings, for a test failure rate of 0.55%. '

i; TUGC0 did not provide an analysis of the adhesion test data on * ~

I' miscellaneous steel after correction for calibration error of the Eleometers used.

j IJLa manner similar to that for the adhesion test data for concrete '

, coatings, the TRT did not conduct a complete independent analysis of

, the massive amount of adhesion data for the miscellaneous steel category. Thedatarecordedonarandomlylelectedgroupof'42PCRs were analyzed in detail. The surface areas of 22 of these items were

,: recorded as less than 10 ft 2, on which only one adhesion test was l2 usually made. The TRT observed that 15 adhesion. test readings were

J telow 200 psi after correction for calibration error out of a total of 78 adhesien test readings, for a test failure rate of

,. approximately 19%. Because of the large variation and uncertainty in the area represented by each adhesion test for miscellaneous steel

~

items, the TRT did not attempt to determine a failure rate in terms of area frc, the available data. -

i

,j The TRT failure rate is so much larger than the rate reported by

, , TUGC0 (0.55%) because of the effect of correcting for calibration

] 1 error. Only 1 of the 15 failed readings observed by the TRT was less 4

than 200 psi before correction for calibration errer, resultir.g in an 5

i i

i A

...---..,n.._. . .. ,-- --.-- ..- . - - ~ . - - - - - - -- ~ - - - . ~ ~ - ~ ~ ~ ~ ' ~- ~

_ . . .__ __ .. .____- . _ _ . _ . . . _ _ . . _ _ _ .x. _ . _ . ._. . . _ . _

uncorrected failure rate in fair agreement with TUGCO's uncorrected y rate. .

1 -

} The 42 PCRs chosen by the TRT for detailed audit represent a small' ,

fraction of the total of 2,189 PCRs on miscellaneous steel.

It is . 't

therefore quite possible that the selected group was not

, representative of the total population of miscellaneous steel items.

However, the TRT's uncorrected failure rate did approximately agree with TUEC's uncorrected rate for the total population. In the July 16, 1984 package of backfit data provided by TUEC, the total

• area of miscellaneous steel was given as 180,0802ft . On the basis - -

of the corrected 19% failure rate obtained by the TRT, the failed area of miscellaneous steel could be approximately 35,000 ft2 ,

assuming the TRT sample is a representative sample.

, The TRT therefore believes that TUGC0 should apply the calibration corrections to all of.the adhesion test data to make a more reliable ~

estimate of the area of coatings that failea the test (see' Actions Requiredsectionbelow). The failed areas should be repaired or

.; placed in the Coatings Exempt Log.

i

b. DFT Testing

• An analysis of the DFT testing results on miscellaneous steel by TUEC is reported as an attachment to a letter dated April 17, 1984 from a TUEC statistician to an NPC statistician.

3 The allowable thicknesses for coatings on steel were given above in l Section 2. The TUEC analysis indicated that 129 DFTs were outside the acceptable range out of a total of 1,517 readings, for a test 3

failure rate of 8.5%. -

i i

+

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

_. - _ . . . . . . _ _ _ . . . . . . - , _ - _ _ _ - 1

j - --

t l The TRT did not attempt to conduct a complete independent analys'is of

! the massive amount of DFT data for the miscellaneous steel category.

l '

The set of 42 PCRs examined for coating adherence was selected to

!, audit the DFT data as well. The TRT observed that 39 DFTs were t.

outside the allowable range out of a total of 252 readings, for a .4

• t

. failure rate of 15.5%. Nine of the failed readings exceeded the j allowable DFT; the remaining 30 were too low.

(c) "NL Adhesion Test Results As discussed above in Section 4a, BNL and its consultant under - -

l contract to Region IV of the NRC performed some adhesion tests on randomly selected areas of the containment liner, the concrete surfaces and the miscellaneous steel surfaces. Only ten adhesion t_es,ty.xere made on each of the three surface types, although it was recognized that such a small number of sample's would have limited statisitical significance. -

As reported in BNL's Interim Report to NRC Region IV, dated April 25,

, 1984, for the liner plate, 4 out of the 10 adhesion tests failed the -

200 psi acceptance value with corrected readings of 156, 186, 186 and 186 psi. For the miscellaneous steel coatings, no failures were observed in the 10 tests. For the concrete surfaces, one of the 10 tests failed. However, the failure was in the concrete substrate, not in the protective coating.

Comparing these results with those reported by TUEC and with the TRT audit of TUEC's results, the BNL results are consistent with the corrected TUEC data for the concrete surfaces and for the miscellaneous steel coatings, considering the small size of the BNL i

sample. Hcwever, the 40% failure rate on the containment liner coatings observed by'BNL is much higher than the failure rate of approximately 5% based on corrected TUEC adhesion test data.

p m.,a

( -

i .

. l i l

_j . - --

i

! By letter dated July 20, 1984, TUEC stated that three of the fou'r

!} failed liner tests sampled an area *under the equipment. hatch .

1 . (Elevation 812 feet, Ozimith 225') which had not been painted until

.i

.i after.the.backfit program had been terminated. The TUEC letter . .-..

'j stated that further adhesion testing by TUEC showed that the failed .4

. t I

area was extremely isolated and was repaired.

I \

The TRT examined inspection records, including maps, pertaining to the additional adhesion testing and' repair in the lirer area under theequipmenthatch(PCR1-0031601 and Traveler UI-006720) and '

confirmed that the area failing the adhesion test was limited to a - -

few square feet. In interviews with TUEC QA personnel, the TRT learned ,that the equipment hatch area was one of three small liner areas in Unit I that were coated after the backfit program was t,erpinated. At this time, the " coin" test (see Coatings SSER, l'

Allegation 44 for discussion of'" coin" test problems) only was used to determine when the primer coat was sufficiently cured to permit l- topcoating. Also,duringtheBNLadhesionlesting,BNLandTUECQA

, personnel detected solvent odor on the test dollies after they were

'l pulled off the coated surface. This suggested a primer curing

. problem and may account for the high failure rate in the equipment hatch area. .

Only one of the 5 adhesion tests of the liner at the 945 feet elevation failed. The TRT confirmed from inspection records and map (PCRI-0031602) of the additional TUEC adhesion testing that the

failed area ccmprised only a few square feet and that adhesion tests f within a foot or two on all sides gave readings above 200 psi.

1

] ,

The TRT concludes that the high adhesion test failure rate observed by BNL for the containment liner represents a very limited area which was probably topcoated before the primer coat had cured. A more s

l reliable curing procedure to assure proper curing of the primer coat j was followed for the large majority of liner coatings. Therefore, l the liner failure rate of approximately 5% based on the corrected

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

_ . . _.. .. .~ _. c __ .__.. . - .

f_. . u.. ; _ . _ . - _ _ . . . . . _ _ _ _ _ . . _ . . . _ . _ . _ . _ _ . - - . _ -

[ . .

f

~

+

TUEC data on 869 adhesion tests is more representative of the l condition of the liner coatings as a whole. .

J i (d) Assessment of Individual Allegations .

1 '?

' The TRT investigated specific allegations related directly or indirectly to the Backfit Test Program.

Visual Defects (AQ0-18)

It is alleged that QC inspectors are not allowed to identify visual - -

defects such as cracking or blistering during backfit inspections.

}

TUEC stated in its June 22, 1984 response to the NRC that "It was not int,qnded that-visual inspections of coated surfaces be performed as part of the backfit inspection program." This is incorrect, because NCR's C-81-01567, Rev. 1, 11/22/81, C-81-01373, Rev. 2, 11/3/81, and

~

NCR C-81-01613, Rev.1, 2/5/82 all require 01sual inspection's as part of the NCR disposition. Instruction QI-QP-11.4-23, Rev. 2, dated December 17, 1981 requires a visual inspection, Rev. 3, deletes this requirement. Similarly, instruction QI-QA-11.4-24, Rev. O, dated February 5, 1982 has a visual inspection which is deleted in Rev. 1.

TUEC's response goes on to state that visual inspections are part of the finish coat final acceptance inspections in procedures QI-QP-11.4-5 and QI-QP-11.4-5, Rev. 5, November 18, 1984 thru and' including Rev. 27, 11/18/83 and procedure QI-QP-11.4-10, Rev. 2, 11/11/81 thru and including Rev. 18, 1/16/84. Both of these

procedures have steps included in the Inspection Report (IR) for visua1 inspections.

The TRT considers th'at though visual defects were not recorded as part of the backfit inspections they were recorded using irs and i

-: x:. r.: y- . ---

- - - .= - - - - -- --

7 . -

- - . - . - - . . w . -_:_ ._i a_ . . . . _ .-. .. - .... . .- ..:-.._

i -

j other procedures. There are some concerns about procedures j QI-QP-11.4-5 and QI-QP-11.4-10 and' visual inspections which will be j ,

discussed in the TRT's procedures review, see TRT Protective i Coatings,. Group 4.  ;

l

.it Backfit Program Vague (A00-19)

It is alleged that Instruction Number QI-QP-11.4-23 and QI-QP-11.4-24 are veqy vague regarding the way the backfit inspections are to be conducted. '

The TRT reviewed both QI-QP-11.4-23 (through Rev. 12, dated October 24,1983) and QI-QP-11.4-24 (through Rev. 6, dated July 14, 1983) and it is the TRT's engineering judgment that these procedures q.ou,1d he. interpreted properly by a well trained QC Inspector. In

, addition, it is our judgment that a QC Inspector that was not well trained in these procedures could have difficulty properly ~

implementing them. The TRT has additional concerns about TUEC

~

coating pro.cedures and inspector training which are discussed in Coatings SSER Groups 4 and 7.

Adhesion Testing-Imoroper (A00-20)

It is alleged that adhesion testing of the protective coatings is not performed properly. It is alleged that QC Inspectors are instructed not to cut around the adhesion test collies when conducting adhesion tests, and that this is contrary to the instructions which are l provided by the manufacturer of the adhesion tester. (The "i

manufacturer's instructions are referenced by CPSES Specificatien AS-31.)

TUEC in its June 22,'1984 response to the i;RC stated that they did not follow the manufacturers instructions that came with the 1

,I e.

m.___,. .

,-,_ . ._.m__.._ ~- ~._ _ - . ...:-----

, .. .; -_ _ _ _ . 2______m._ . . _ __ _m. m . . _ _ m _ _ . . . . ..._. . .._ _._ 1 l-I -

instrument. The TRT has determined that the allegation as j characterized above was substantially correct; QC Inspectors did not scribe around dollies although the manufacturer's instructions do

instruct the user to score around the dollies prior to performing the,  ; ,

j ,

tests. TUEC has provided results of on-site testing to support the' .;-

position that scribing or not scribing does not affect test results.

The TRT has reviewed these tes't results as well as input frem other qualified individuals in the industry and considers that the failure to scribe around dollies did not affect test results and was not a technically improper procedure. It was noted that TUEC has since

, mcdified site practice to institute scoring around dollies in "' ~

compliance with the manufacturer's instructions.

Adhesion Tester Calibration Correction (A00-21)

, It is alleged that Brown & Root is doing the calibration on these adhesion testers, and they are not using a corrected value curve (which should have been supplied with each unit).

As alleged, TUEC did not correct adhesion test readings for Elcometer calibration error until after the backfit inspection program was terminated. The effect of making these corrections on adhesion test

- failure rates is discussed in detail in Section 4 b(1) above.

Backfit Records (A00-37)

It is alleged that: -

(a) for the backfit program, areas that were stated to have i

j satisfactory primer documer.tation ended up having 10 mils of

)

primer, which exceeded the allowed maximum; I

-1 1

1 . ._. . . _ _ _ . . . . . . _ . . . _ . - .. ... ..m._,.

s .

I 1

.[

i 1' .

I[ . - --

l (b) none of the maps showing areas of adequate primer documentation oi are correct; . .

d '

!j (c) documentation for the backfit program was forged and falsified;

~

E

-(d) a QC inspector on the night shift wrote up acceptable irs for the containment dome area without ever performing those

,' inspections. Also, that several QC inspectors would " buy off" anything.

Plus on several occasions at least one QC inspector conducted his -

coatings inspections from several floors below where the paint was being applied.  ;

P (aj,, Satisfactory primer exceeded the allowed maxieum. TRT discussed j, --

this issue with TUEC management, Quality Engineering and Quality  !

l- Control Inspectors. It was indicated to TRT by all parties contacted that it was true that in son cases during backfit inspec.tions performed on liner steel. spot DEF readings for primer coated surfaces with acceptable documentation were found to be outside of the acceptable DFT range of the primer and g docu onted as unsatisfactory. TUEC in its June 22,.1984 letter ii to NRC also stated that unsatisfactory readings detected during 4

the backfit program were tracked by irs until the conditions as 1 corrected. Examination of backfit IR records by TRT indicated l that unsatisfactory conditions were noted and a new IR generated

after the item is corrected. 'The TRT has determined that an  !

jj alternate disposition is to prepare a design change 1: authorization (DCA) that widens the range for the DFT readings.

lf Coatings SSER 2 discusses TRT concerns with these DCAs and their d effect on DBA Qualification of coatings.- l

l

~

,1 1

{ J

!i.  ;

.j '

I .T._ ~ ___.i _.ZZ._ Z.T._.

~ 3. _ -

i ~' i -

. . Z .Z_ Z ' . . ' .

Li_, . . ,

i .

4 i

.i ,

. - 23 ..

1 ,

a 1 Allegations 54 and E5, which are closely related to this r

j allegation provide additional assesscent below in.this $5ER. [

(b) Backfit liner maos incorrect. TRT initially examined the Ifner ,  ;,

plate maps used to keep track of backfit testing and rework. .

These maps indicated areas that had primer applied, inaccessable areas and uncoated liner plate. These areas .were clearly .

t indicated by bold dark border lines with backfit inspection 4

report (PCR) numbers and azimuths and elevation identified in

{

the center. There were no apparent discrepancies noted by l

reviewing the maps and associated PCRs. The only discrepancy, ~ -

l that could occur is incorrect azimuths given by the QC  !

I' inspe~ctors. TRT discussed this issue with the QC inspectors who j were knowledgeable of the mapping system. The QC inspector

• _.4tated- that there were no azimuth markings on the liner for QC

• l

, to obtain precisa ceasurements for location, QC inspectors used approximate locations. This would cause a gap or overlap  ;

. , i between adjacent inspected areas when laid out on maps even j though there was a full coverage of inspection performed on the liner by QC.

l' TRT then reviewed the final liner plate map cutlining all liner

{~ areas that were backfit tested. As stated in 4.b.(2) above, TRT

• i reviewed the data in 869 liner plate PCRs and in ccmparing their ,

indicated locations wit the final map, did not find any sericus 1

J discrepancies with this map, which showed that 96". of the liner  :

plate area had been backfit tested. -

I (

i (c) Backfit test progran documentation was forced or falsified. '

.. ,This allegation was investigated by the NRC Office of Investigation (0I) and reported in QI Re~ port 4-83-011 dated July 24, 1984. The investigation-established that there was no violation of procedures.- 01 forwarded the report to NRC Region ii 1

. . . . _ _ _ , . . . _ _ - - _ ; 7_.r - - .

._..r_1_._,___ -.r_ _._n e ._.m ._. = y__ . _.m m_,___ _

j .= _ L_._ __-._ _ ---

~,

1 .  !

q ,

1 t IV and recommended a review of the technical adequacy of the

)j coating records. -

TRT reviewed the inspection records contained as exhibits in the l- OI report. Mostly TRT observed that there were xeroxed copies ,

.j of paint batch mix sheets attached to separate coating  :

application inspection checklists for different items that  !

appeared to be coated at the same time and in the same

) environment. In each exhibit there was scme common identifier such as containment liner location or item descriptien that -

ll appeared to connect the two documents. Also, use of xeroxed -- -

i j paint mix sheets was not a violation of procedure. TRT did not  ;

! , observe any technical inconsistencies in the exhibit inspection {

reports. , l

, .~ - - -

j -- TRT then folicwed up by reviewing three ' additional folders 4 .

4 l containing approximately 250 inspection, reports which were '

randomly selected from the site QA record vaults. Of this

! group, TRT fcund two inspection reports where the coating mix

date and application data were inconsistent. These two reports

! applied to the containment liner and referred to NCR C-81-01567

, which dispositicns such items to be backfit inspected.

] Based on the above review, TRT could draw no conclusion as to falsification or forging of records. Also, the two instances of l technical inconsistencies that were found were backfit inspected and applied to the centainment liner for which 96% of the area r was backfit inspected. Consequently, the TRT believes the coated area associated with technically inconsistent records is i

very small.  :

q .

l

_O-'%- ______2- -_t-__.u._r * * " A v4n' W t 's _' ' Pee.*L%-___ h ** 1e T..* a Ont . a.__ r_s--- 9 s- m

. _ ..-_.___,_..._.____.m.._._._

'I..

,i.

'i r

~

l

^

4i (d) A OC Inspector wrote us acceptable irs without ohysically '  !

i!

l inspecting the work. Also several QC inspectors would " buy  !

I off" anything. -

i n

\ .

l J

The TRT Grcup Leader contacted the alleger (A-31) by telephone

-l on September 10, 1984 to obtain further details to aid in -

investigating this allegation. The alleger re-iterated the same '

two inspector names he provided to 0I in October 1983 and included in OI Report 84-006 dated March 7, 1984. However, he -

was not able to identify specific plant locations involved or --

specific IR numbers or dates.

TRT then reviewed approximately two dozen irs prepared by one of  !

j the. named inspectors and about 6-12 irs prepared by the other* .

~

named inspector. TRT could not determin'e frcm these records J that the inspections were not actually performed. TRT did not i j attempt further investigation of this~111egation and considers it to be indeterminate. .

Rust Seen Through Tooke Gauge Tests (BNL-46) l

l. (a) It is alleged that during tooke gauge tests, ti was observed that  !

rust was seen on steel substrate, and grease, grime, filth, and other

] contaminants on concrete substrated.

{ (b) It is further alleged that Elcometer adhes. ion dollies, after being . ,

pulled off of a coated surface, had rust adhering to the underside of the dolly. It is also alleged that the QC Inspector Lead was aware -

5 of this condition and failed t'o take any correct,1ve action.

Review of part (a) o'f this allegation by TRT indicated that it is not reasonable for anyone to observe rust, grease, grime, filth and other

!! contaminants under five to six mils of primer and finish ccat through l a Tooke DFT Gauge. An experiment conducted by TRT and a TUEC l 1 ,

1 l

l

_.-... .e . . 2 . L..- -- - - - - ---

l_. , .- -

1 i '

representative with a primer coat of Dimecote 6 on steel substra'te l illustrated that Dimecote 6 could have possibly been mistaken,as a

{- '

light shade of rust. Dimecote 6 has a light reddish gray colcr. The

] QC inspector, possibiy not having a great deal of experience with ,

Dimecote 6, could have mistaken the reddish pigmentation for rust. ,

TRT found no irs /NCRs that document this allegation for evidence.

Note: SSER Training assesses allegation AQO-33 raised by another alleger and which relates to coating failure due to rust on the seal -

table A-frame steel. However, the A-frame rust was not determined by- _.

a Tooke DFT Gauge.

Rev.iew of part (b) of this allegation consisted of the TRT igery,iewing.six QC inspectors certified for Backfit Program inspections. The inspectors each stated that! they had seen or heard reports of adhesion tests which revealed surface rusting. However, no specific locations were indicated. The inspectors stated their belief that in each instance the affected area was identified and repaired in accordance with procedural requirements. TRT made no attempt to verify such repairs since specific locations were not

'known. The significance of this allegation, to the extent that it is substantiated is best judged by the backfit adhesion test results presented in Section 4.b above.

Alleaation #47

, It is alleged that for an installation hanger for the steam generators, in violation of a written instruction, QC inspectors were instructed to perform approximately 25 elcometer adhesion tests.

Whdn a protective coating'on steel fails an adhesion test, TUEC procedure  !

QI-QP-11.4-5 requires additional adhesion tests in the vicinity of the i

i 1

5

--=s________=_:_-=_-__,-_--- -- -=" - --~ 2-- -

~ ~

,. : - .=.>--..w..... = . . . . - - . :i - - ---

} .

3l 4

~

~I failed test to delineate the extent of unsatisfactory coating which must j be repaired. Contrary to the allegation, the written instruction does not.

place an upper ifmit on the number of adhesion tests required to determine s

.; the extent of unsatisfactory coating. , ,

In response to this allegation, TUEC described a case where QC inspectors were requested by construction personnel to perform a total of 32 adhesion tests on the insulation support ring of Steam Generator No. 4 after

' failures we.re observed in the first two sets of three adhesion tests. In '

this case, there were two failures.in the'last 26 adhesion tests and the

~

entire ring was stripped and recoated. It would have been a violation of~ -

Instruction QI-QP-11.4-5 ff areas which failed the adhesion test had not been reworked.

In the_hsence- of evidence that requests for additional pull tests were

. used to avoid reworking areas of unsatisfactory coatings, the TRT finds ,

this allegation to be unsubstantiated. -

A11ecation #54 -

It is alleged that during the Backfit Program, only the first i unsatisfactory reading was recorded, even if the following readings were either higher or lower, meaning further out of the acceptable range. it i's further alleged that the trend analysis was adversely affected by no -

including the actual readings.

, The alleger's main concern apptn.re to te that if the first unsatisfactory

reading of coating thickness was only slightly out of specification and it I was the only reading recorded for the sampled area, then if the range of acceptable, thicknesses was later widened by a DCA to include the recorded i

reading, the sampled area,would be dispositioned a's satisfactory even though other unrecorded thickness readings might have been outside the i

i I

i

.._ .- . . _ ._ ._.m . . . -. _. . - . _

I .

I ,

i*

.3 i

} widened thickness specifications. Consequently, such an area would neither be repaired nor entered into the Coatings Exempt Log. .

The sp'ecification changes widening the acceptable range of coating ,

thicknesses on steel and concrete are discussed in Section 4.b above, .'.'

assessment of Allegation #55 below, and in Coatings SSER-2-CBA Qualification Testing. The TRT was not able to ascertain independently '

the extent to which inspectors recorded only the first unsatisfactory thickness re.ading of coatings. However, the alleger stated that the problem was not widespread and inspectors generally did not follow the verbal instruction of a particular QC supervisor to record only the first~ -

unsatisfactory reading unless the supervisor was watching.

Other factors tending to mitigate the practical effect of the first-read-ing-only practice are the following. ,

(1) Most of the out-of-specification thidkness readings were too low.

For coatings on steel, adhesion is more strongly affected (decreased) by primer or topcoat which is too thick. For coatings on concrete, adhesion is not significantly affected by primer thickness, and excessive topcoat thickness has a lesser effect than on steel.

1

(2) Approximately 96% of the containment liner and 50% of the concrete surfaces in Unit I were backfit-inspected for coating thickness.

Except for a few areas of limited si:e which were placed in the Coatings Exempt Log, all nonconforming areas of the liner and concrete were reworked until the coating thicknesses were in the

allowable range. Only 22% of the miscellaneous steel coatings were backfit-inspected. The miscellaneous steel items with nonconforming

primer or topcoat thickr,tesses were entered into the Coatings Exempt  ;

! Log by NCR-C-83-0130,5 instead of being repair ~ed.

l1 I

t i tN

,.-.....n-...- .g . . ~ , . . . .

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

~

~ . .. . . .....-. . L . ..: : ....:. -  : :- ....--..=-

'l,. .

I.

t l - --

(3) According to TUEC QA personnel interviewed by the TRT, a large ~

l' majority of the 22% of the backfit testedmiscellaneous steel i.tems

[ . with low topcoat thicknesses were in fact reworked with more topcoat, s.

partly for cosmetic reasons. Similarly, partly for appearance's 1 l; sake, additional topcoat was applied to most of the coatings which ,.,

l' ' had not been backfit-inspected (4% of the containment liner,' 50% of l theconcretesurfacesand78%ofthemiscellaneoussteelsurfaces).

Because of the three mitigating factors and the alleger's statement that inspectors f;ollowed the first-reading-only practice only while under the supervisor's scrutiny, the TRT concludes that the practice will affect the- -

quality of only a small fraction of the protective coatings.

Insofar as possible wrongdoing relative to the verbal instructions given by the QC~ s,u,pervisor as indicated above, the TRT Director forwarded to NRC

. Region IV OI by letter dated August 24, 1984 the t'ranscribed NRC interview with the alleger for OI review. .

Allegation #55 .

It is alleged that areas identified during the Backfit Program as being outside of the acceptable range for applied coatings were not reworked as 4

required.

As discussed in Section 4.b.above, the ranges of acceptable coating thicknesses were widened by DCA's (e.g. 12, 145). The widened ranges were incorporated into the later revisions of procedures QI-QP-11.4-23 and .

01-WP-11.4-24. After.these revisions, some coatings which met the expanded specifications were dispositioned, "use as is" whereas repair would have been required by the earlier specifications.

In the'TRT respense to Al' legation #54, it was stated that NCR-C-83-01305 permitted miscellaneous steel items with nonconforming primer on total l l

ll-l l l- -. . . . - - . - . . . - . . . . . - - - . . - . . -

7. ; _

~

. -. - .. ; . - . . : . ... J-- --.. - - ~ - - -

1 - --

~

coating thickness to be placed in the Coatings Exempt Log (CEL) rather i

1 than being repaired. TUEC procedures and NCR's required the repair of j .

nonconforming coating thicknesses on containment liner and concrete surfac'es. .

i .s

. t In its review of inspection rehorts and NCR's related to protective coatings. the TRT found no instances where unsatisfactory ccatings wire dispositioned "use as is" without being entered into the CEL. However, according to interviews with TUEC QA personnel, the entry of miscellaneous-steel items into the CEL (the first four entries in the CEL) was based on a TUEC assumption that 5% of each category of miscellaneous steel was of

• 3 --

indeterminate quality. As discussed in Section 4.b above and in Coatings SSER-6 Coating Exempt Log, this assumption may be low. One of the actions required regarding protective coatingt is that TUEC correct existing adhesion,. test-data for Elcometer calibraticn errcr to provide a more

, accurate estimate of the percentage of miscellanecus steel coatings which are of unsatisfactory quality. "

In a sense, the uncertainty regarding the miscellanecus steel CEL entries

may be considered partial substantiation of the allegation. The TRT fcund j no other evidence confirming this allegation.

l DocumentationOestroyed(BNL-56) i It is alleged that original decementation related to the Backfit Program was destroyed by QA management.

j This allegation involves possible wrongdoing which is outside the TRT

] assigned scope at:d therefore the TRT did not directly investigate this allegation,. Further, since the alleger could not identify which docurrents were destroyed, TRT was unable to make any crossch'eck fron related

] documents to substantiate the allegation or make any technical assessment j of the impact of the alleged destroyed documents on the coatings quality.

I

?

l ,

. . _ _ ,. , ... . _ . . . . - . , - , . . . - - - . .. .- . .._-~-:--- .__ u_._ _%

. .. ...... - - - -.-- -.- .-.-.: - = - - - : -- - - - -

=l

. \

I l

l - --

The allager indirectly relates this allegation to forging and r

j falsificatien of BTP documents (alleged by another person) which is assessed in Allegation #37(c) above.

}

By letter dated August 24, 1984, the TRT director forwarded the

.;5

transcribed NF.C 1.iterview with this alleger to NRC Region IV OI for their

! review.

l 5, Conclusions and Staff Position /

. a. Specific Allegations Concerning BTP ~ '

TRT conclusions concerning specific allegations related directly or indirectly to the coatings Eackfit Test Program (BTP) are as fo11cws:

_n

. .(1) Allegations 18,20,37(a).46(b),47weresubstantiated;however,for t the reasons stated in the corresponding assessments presented in the

~

i j section above. TRT does not consider them to be a significant tec'hnical

], concern in regard to the BTP. *

• 1-

!! Allegation 21 was also substantiated and as discussed in detail in Section l

4(b) above. TRT finds that correcting adhesion test results for e1cometer i" calibration errer has only a small effect on liner and concrete test

$ failure rates, but could significantly change the test failure rate for j, miscellaneous steel. For the liner plate, the failure rate only changed  ;

from TUEC's uncorrected 2.3%(20/869) to TRT's corrected rate of 4.3%

i a

(36/834)ortoTUEC'scorrectedrateof 5.9%(51/869). For the concrete. -

the change was from a zero% uncorrected failure rate to TRT's corrected i

6.8%(8/116) rate or to TUEC's corrected rate of 3.1%(65/2128). For

i miscellaneous steel, however, the test failures rate could change from

'j less than 1%(26/4714) uncorrected to as much as IM corrected based on a limited sam'ple analysis by TRT.

i .

-- -_._i._. .

1. -- _e

j- ..

1 .

}.

i . - --

! (2) Allegations 37(c), 54 and 55 were partially substantiated . On i

allegation 37(c) TRT could not draw a c6nclusion as to forging or .

] falsification of documents but did find two instances of technically l inconsistent inspection documents which could involve only a small area of , ,

liner coating. Allegations 54 and 55 are not considered to have any .4

. e significant effect since the amount of area not reworked due to coating thickness out of tolerance is believed to be small. However, TUEC's estimated 5% miscellaneous steel coating area allowance due to indeterminate quality may be low when correction of adhesion test results is made as discussed above. '

(3) Allegations 19, 37*b) and 46(a) were not substantiated and therefore is no concern.

(4) A,ll,e.ga.tjon 37(d) was found to be indeterminate.

(5) A legation 56 was not investigated by TRT since it involves wrongdoing and there was insufficient information available to assess its

~

i potential extent, and affect on coating quality.. This allegation has been l referred to NRC Region IV OI for their review.

Even though TRT found that most of the above allegations were of small

-technical significance with respect to the BTP overall pass / fail ' rate, many of them, as indicated in the individual assessments, relate directly to TRT generic concerns about other aspects of the TUEC coating QA program and are addressed in other Coatings SSER sections.

5.b Evaluation of Backfit Program Test Results TRT evaluated the BTP inspection and test field data directly to determine independently, within practical limits, the amount of coated areas tested and the extent to which tested coatings passed their DFT and adhesion test l acceptance criteria.

i

l.

l.

1 l , . .

It is the TRT's engineering judgment that if DBA-qualified coating.wa's j applied and its traceability was maintained, then the BTP, properly.

j . administered, could provide useful, but indirect information on the l quality and DBA survivability of the coating work. Hdwever, TRT found

serious shortccmings in DBA qualiffcation and traceability of coatings as " .. j i

~

dis' cussed in Coatings SSER-2 and and 3 respectively. Further, TRT does not consider that the BTP scope or results provide information that resolve these shortcomings. -

, 1 Nevertheless, even with these shortccmings, the TRT concludes that the -

thousands of tests conducted under the BTP provide a useful overall - -

measure of two important coating quality parameters;'namely, adhesion

. strength and coating thickness.

The rey,1,Ls of. the.BTP adhesion tests, corrected for Elecmeter calibration -

errer, are summarized in the attached table. On t'he basis of these results, reasonably accurate estimates of the adhesion test failure rate

, and the corresponding failed areas can make for the coatings en the

, containment liner and concrete surfaces. For the miscellaneous steel j surfaces the adhesion test failure rate could be only crudely approximatt:d 1 because corrected adhesion test data are lacking except for the small i sample audited by the TRT. TRT's corrected data failure rate is about 19%

.- in contrast to TUEC's uncorrected data failure rate of less than 1%.

A 19% miscellaneous steel failure rate corresponds to a failed coating area of 34,200 ft2. As shown in the table, this failed containment liner f and the concrete gives a total fo approximately 57,500 ft 2 f7 .catings j which failed adhesion tests.

~

I i Besides the adhesion tests, the oth'er BTP test was coating thickness j measuremen't. For the liner and concrete, coatings with thicknesses the I allowable range were rewo'rked until the thicknesses were acceptable

(although the allowable range was expended several times) in nearly all f .

l m 3

y.p. .; w ..,-.-,,,, .y.q .-+y.m.- z. i -

_ - - > - . - - - - .---,---s- .-

- - f- - --- . - - -- ..-- J

r .....:.. -- =. .. .... .....;,c....

. .. . . . - .. . . . . . - . - . ~ . _ - . . -. -

2 .  !

J f

ADHESION TEST DATA

'}

i ,

l:

., ' I TRT ESTIMATE TUEC ESTIMATE

' ' [

Total Area  % Area Failure - Failed  % Area Failure Est.Failgd a

Item ft2 Audited Rate, % g, Area, ft2 Tested Rate, % Area, ft l

'. j

. Containment 110,000 79% 3.6% 3960 96% 5.6% -

6160 l l Lir.er -

3

.t.

i

.i Concrete 285,000 3.9% 6.8% 19400 50% 3.1% 8835 i

l Miscellaneous 180,000 0.36% 19% 34200 22% Not Not ,

[

Steel Available Available -

i. e Total 575,000 57560 i  !

% i 1 - ^. i i

] .

i *

,1 ,

1 '

i i

h I

't 3

I l 4 -

B

. -> 1

. e 0 i

( --

l .

I=

I cases. In the low percentage of cases where repair was not feasible,'the 3

discrepant areas were placed in the CEL. .

I For miscellaneous steel with unsatisfactory coating thicknesses, the item ,

could be dispositioned "use-as-is" and placed in the CEL by NCR's such as

)

C-83-03103. Rather than attempt to estimate the area of each discrepant item, TUEC arbitrarily entered abcut 5% of each type of miscellaneous 2

steel area totaling 8150 ft in the CEL. The observed coating thickness

' failure rates appear to be well below 5%', so that the CEL entries are considered conservative.

In summary, based on TRT review of BTP data,' TRT estimates that 90% of all coated surfaces meet adhesion test requirements (assuming 19% failure rate for miscellaneous steel) TRT estimates that greater than 90% of coated surface thri<kness was acceptable (allowing for repeated relaxation in '

. thickness tolerances). However, the extent to which the BTP demonstrated that FASAR coating quality requirements wdre met is dependent on

_- ~

resolution of other TRT generic cuocerns 'related to DBA Qualification and Traceability of coatings as discussed in Coating SSER's -2 and -3.

, For an integrated assessment of the estimated total coating area inside .

containment which failed the BTP and/or had indeterminate quality for other reasons, including vendor equipment with ijnqualified coatings, refer t'o Coatings SSER-6, Coating Exempt Log.

6. Actions Recuired

! The Technical Review Team (TRT) reviewed the procedures used and the

] inspection reports (PCRs) and statistical data resulted from the Coatings j -Backfit Te.st Program conducted for coatings applied to the containment I

liner, concrete strctureu,s, and miscellaneous steel.

For 2189 miscellaneous steel items (such as pipe hangers, cabl'e trays, and i equipment and conduit supports) that were' tested under th.e backfit test i

i

.-%.w.7...--. yea #-- --* m , . ._ .p m __ . ,

m.,, . .~~c--.y-

. 7 . . ._

• s t . . - ,

l 1

~

) program from about December 19El to February 1984, the TRT found that TUEC i did not correct any of the coating adhesion test field data.to account for.

the calibration error of the Elcometers used for the tests. Consequently,

'j the field test. data could be in error by as much as 200 psi in a -

nonconservative' direction. Thus, any Elcometer reading less than 400 psi  ;

j ,

1 potentially represents a test that failed to meet the test acceptance _

criterion of 200 psi.

Based on TRT analysis of sampled data coverir? 78 test results for 42 miscellaneous steel items, the appropriate calibration correction could '

increase the adhesion test failure rate from about 1% to 19%. The - ~

following inspection reports (PCRs) indicated acceptable test results Q200 psi)beforecorrectionbutunacceptabletestresults(>200 psi) ,_

after correction.

s. '

, PCR-02103 PCR-02i64 PCR-02105 PCR-02166 -

PCR-02515 PCR-02171 Accordingly, TUEC shall:

. (1) Apply the Elcometer calibration correction to the data for the 4714 adhesion tests covering 2189 miscellanecus steel items teste'd to establish a more reliable estimate of the adhesion' test failure rate.

This revised analysis should include a statistical analysis showing the 95% confidence upper limit of the failure rate for all the miscellanecus steel items inside the containment building. (It is

{ appropriate to assume that the approximately 22% uf the total

-l miscellaneous steel coated surface that was backfit tested is a

] representative sample of the remcining miscellaneous steel surface).

j - .

I w===,_ _

-,p.p, umme,-* m. - w, , og ,

e J- m; , . . N -e * *% g ; y- - - - -

I

- 37*-

~

, z) Analyze the corrected data to establish a more reliable estimate of the fraction of. test miscellaneous steel coated surlace that failed the adhesion test acceptance criterion. The resulting failed areas should be. repaired / reworked or be otherwise dispositioned, such as by. , ,,

entry in the Protective Coating Exempt Log (see item IV.f. below). .j

. .. n:wn 2,w,d"' -

.  :%%%<td5d~xFif"^'%

m-

'None.

8. Attachments None.

. 9. Reference.Dccuments . .

(1) QI-QP-11.4-23, " Reinspection of Coated Stsel Substrates for which ~

DocumentationisMissingorDiscrepant,"Rev.0[-'Nov.19,1981 th' rough Rev. 13, April 18, 1984. -

(2) QI-AP-11.4-24, " Reinspection of Protective Coatings on Concrete Substrates for which Documentation is Missing or Discrepant," Rev. O, Feb. 5,1982 through Rev. 7, April 18, 1984.

(3) TUGC0 letter of July 16,1984, TXX-4225, f' rem J. T. Merritt, Jr. to Richard L. Bangart, Region IV, "CPSES Units 1 and 2, Brookhaven Interim r,epset on Protective Coatings, Fi,le No. 906.1, 10010."

(4) TUGC0 off. ice memorandum of October 2, 1981, QTN-434 from D. N. Chapman to J., T. Merritt, Jr. , "CPSES, TUGC0 QA Audit Report TCP-24, Protective

. Coatings QA Audit File: TCP-24. ,

w-

(. ~ l - . . - - . .

t .

~ .

! (5) TUGC0 letter of Nov. 19,1981, TXX-3439, from R. J. Gray to G. L.14ad'sen,

j. Region IV, "CPSES Response to NRC Notice of Violation, Inspection Report

.j No. 81-15, File No. 10130.

(6) TUGC0 office memoranda, QTQ-107, TUQ-1085, TUQ-1016, QTQ-96, and TUQ-992 .

j bet' ween R. G. Tolson and D. N. Chapman between Oct. 30, 1981 and Jan. 22, 1982.

(7) NRC letter of Jan. 19, 1982, from W. C. Seidle to R. J. Gary of TUGCO.

(8) NRC letter of Oct. 26, 1981 from W. C. Seidle to R. J. Gary of TUGCO. - -

t' ?) NRC letter of Nov. 6,1981 from W. C. Seidle to R. J. Gary of TUGCO.

(10)NCRNo,._C81-01373,R.5,10/19/81 through Rev. O, 10/19/81.

(11) NCR No. C-81-01613, R.4, 11/23/81 through Rev. 0,.11/23/81.-

(12) NCR No. C-81-01567, R.4, 11/16/81 through Rev. O, 11/16/81. .

(13) NCR No. C-83-01903, 7/12/83.

(14) NCR No. C84-00'471, R.2, 3/6/84 through Rev. O, 2/9/84. '

(15)NCRNo.C-83-01986,7/21/83.

(16)TUGC0letterofJune 22, 1984, TXX-4201, from L. F. Fikar to R. L.

Bangart, NRC, Region IV, " Protective Coatings Allegaticns Transmitted by I letter of liay 18, 1984, File No. 906.1." -

e t

?

i 1

i

_h +=,ee 49m*. .m e - 9mv ==eme , mams - * - - e==sw. -

• • t * *P *-%a q. +.= e- + g== o - eeg - em. ,

I

. y:5. =

07) TUGC0 letter of Aug. 21, 1984, TXX-4262, from L. F. Fikar to T. A.

Ippolito, NRC, " Additional Questions Relative to Protective Coatings Allegations, File No. 906.1, 10010.

J *

(18) Imperial Professional Coatings letter of Jan. 19, 1983 to Dear Customer

• from G. A. Arnold, NUTEC System - Film Thickness Controls. .-

(19) Imperial Professional Coatings, Application Instructions, NUTEC 115 Surfacer /NUTEC 11 Surfacer /NUTEC 1201 Top Coat System, Revision 8/7/81, pages 2 through 20.

(20) CCP-40, " Protective Coatings of Concrete Surfaces," Rev. 5, 8/18/82.

(21) QI-QP-11.4-10, " Inspection of Concrete Substrate Surface Preparation and Coatinsi. Application and Repair," Rev. 2, 11/11/81 through Rev. 18, 11/16/84.

(22) BNL report of June 13, 1984 to NRC Region IV, " Status Report on Protective Coatings Allegations."

(23) BNL report of April 25, 1984 to NRC, Region IV, " Interim Report on Protective Coatings."

(24) Memo to File, from V. Lettieri, 8/27/84, "Backfit Procedures and Visual Inspections."

(25) QI-QA-11.4-5, " Inspection of Steel Substrate Seal and Finish Coat Application and Repair," Rev. 5, 11/18/81 through Rev. 17, 11/8/83.

(26) TUGC0 office memorandum, TUQ-1726, 7/6/83, from C. T. Brandt to File,

" Test Report on Adhesion Tests."

r

[ ,

e s- .

~

l l

(27) TUGC0 office memorandum, QTQ-369, 6/23/83 from L. M. Bielfeldt to C. T.

f Brandt,"PatchTestResults,"r'af.NRCIR 83-07/02, NRC IR 83-19/13.

i .

(28) Telex 'to TUGCO from KTA-Tator PGH to M. Foote of TUGCO, 2/3/83. . ,

li (29) Elecmeter Adhesion Tester Manufacturer's Operating Instructions for M,odel 106B.

i l (30) Protective Coatings Specification 2323-AS-31, Rev. 2, March 15,1984 l 10. This statement prepared by:

I V. Lettieri Date S. Xirslis

', . Date i

P. Matthews' Date Reviewed by:

-f l Group Leader Date Approved by:

Project Director Date 1

.I l

\

9

. g *

' s.

,..