ML20153H168

Insp Repts 50-445/88-47 & 50-446/88-42 on 880702-0802. Violation Noted.Major Areas Inspected:Applicant Actions on Const Deficiencies,Info Notice 85-024,previously Identified Items & Removal of Protective Coating from Piping

ML20153H168
Person / Time
Site:	Comanche Peak
Issue date:	09/02/1988
From:	Livermore H, Phillips H NRC OFFICE OF SPECIAL PROJECTS
To:
Shared Package
ML20153H167	List: ML20153H161 ML20153H168
References
50-445-88-42, 50-445-88-47, 50-446-88-42, IEIN-85-024, IEIN-85-24, NUDOCS 8809090108
Download: ML20153H168 (21)

See also: IR 05000445/1988047

Text

c

'

'

. .

G

l

r

11. S. NUCLEAR REGUL ATORY COMMISSION

OFFICE OF SPECIAL PROJECTS

NRC Inspection Report: 30-445/88-47 Permits: CPPR-126

50-446/88-42 CPPR-127

Dockets: 50-445 Category: A2

50-446

Construction Permit

Expiration Dates:

Unit la Extension request

submitted.

Unit 2: Extension request

submitted.

Applicant: TU Electric

Skyway Tower

400 North Olive Street

Lock Box 81

Dallas, Texas 75201

Facility Names Comanche Peak Steam Electric Station (CPSES),

Units 1 & 2

Inspection At Comanche Peak Site, Glen Rose, Texas

Inspection Conducted: July 7 through August 2, 1988 ,

'

!

Inspector / */d , k4 ///

H.'s. Phillips, Senior Resident Inspector Date

'

Construction

i

Reviewed by: M b /_ a44tEd) b

~

H. H. Livermore, Lead Senior Inspector

MA//#

' D' ate

eso909otos esogor

POR ADOCK 0500o445 '

0 FDC  !

_ _ _ _ _ _ _ _ _ _ _ _ _ _ ___

'

g .

i

2

l

Inspection Summary:

Inspection Conducted: July 7 through August _), 1988 (Report

50-445/88-47s 50-446/88-42)

Areas Inspected: Unannounced, resident safety inspection of

applicant's actions on 50.55(o) construction deficiencios, action

on Information Notice 85-24, previously identified inspection '

items, removal of protective coating from SWS piping, and new open

items concerning Plasite removal.

Results: Within the areas inspected, one apparent violation was .

'

identified (breakdown in the QA program [Critorion II) relative to

the removal of the protective liner from the service water system

(SWS) piping), paragraph 3). The findings that support the ,

apparent critorion II violation concerning the SWS liner removal

are: failure to establish QA/QC program requirements and technical

requirements in procurement documents, paragraph 3.a; failure to

control special processos, paragraph 3.b; failure to provide

adequato QA/QC procedures, paragraph 3.ca and failure to take

corrective action, paragraph 3.d. Two new open items concern the

disposition of defects on the outside of the piping without

consideration of the interior defects, paragraph 3.o(1), and

engincering assigning a quality level rather than QA, ,

paragraph 3.e(2). ,

These findings represent a significant weakness in that project

management did not insure the "up-fron*.' involvement and

implementation of quality at the onset - the pipe liner removal

task.

l

l

{

!

I

h

I

t

I

l

'

i

!

,

r

1

- - -, __--_ -- - - _ ____.

'

s

.

O

3

'

DETAILS

l. persons Contacted

• R. p. Bakor, Licensing Complianco Manager, TU Electric

• W. G. Counsil, Executive Vice President, TU Electric

• G. G. Davis, Nuclear Oporations Inspection Report Item

Coordinator, TU Electric

• D. E. Devincy, Deputy Director, QA, TU Electric

, *W. G. Guldcmond, Exocutive Assistant, TU Electric ,

'

• T. L. Heatherll', Licensing Complianco Engineer,

TU Electric

• R. T. Jenkins, Manager, Mechanical Engineering, TU Electric '

• 0. W. Lowe, Director of Engincoring, TU Electric

• S. M. Matthews, Chief Inspector, State of Texas

• D. M. McAfee, Manager, QA, TU Electric

• D. M. Reynorson, Director of Construction, TU Electric

• J. S. Smith, Plant Operations Staff, TU Electric

',

• J. F. Strcotor, Director, QA, TU Electric

• R. D. Walker, Nuclear Licensing, Manager, TU Electric I

The NRC inspectors also interviewed other applicant employees l

during this inspection period. t

i

• Denotes personnel present at the August 2, 1988, exit

meeting. A complete list of exit meeting attendees is

i included in NRC Inspectiot Report (IR) 50-445/88-51;

i

50-446/88-47.

2. Quality Assurance and Administrative controls (35061, 35065,

92700, 92701)

6. Follow-up on Construction Deficiencies Concerning Service

Water system Piping

4 In NRC IR 50-445/88-34; 50-446/88-30, the NRC documented i

a detailed inspection of construction deficiency

(10 CFR Part 50.55[c)) files SDAR CF-80-07 and CP-86-07.

In 1980 and 1985, respectively, these two deficiency j

reports stated that defects were found in the Plasite r

7121 linar in the SWS piping. Inspection of the SWS .

'

piping is also documented in NRC IR 50-445/48-40;

50-446/88-36. {

J

A Plasite liner, an epoxy coating, was applied to the

inside of the carbon steel piping, ASTM A106 Grade B. In

j the 1975-76 time frame, the specifications originally

j described the coating and application as safety related.  !

l The vendor applied a "Q" coating at their shop prior to  !

I shipping the SWS piping to the site. According to Gibbs  !

l  !

& Hill (G&H) engineering, TWX No. 12523, the coating on

!

,

field wold areas was applied by the site contractor

,

.

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

'

t ,

2

4

without QA/QC program controls. The coating was

intended to protect the piping from the slightly

corrosive water supplied from the Squaw Crook Reservoir.

The Unit 1 system has operated for approximately six

years, half of this time in wet lay up.

The NRC inspector found the applican c's evaluation of

deficiency CP-40-07 inadequate. That is, Gibbs & rill,

Inc., the applicant's design engineer of record, did not

appropriately consider the effect of the coating applied

to SWS piping by a site contractor who did not procure or

apply the coating under a quality assurance / control

(QA/QC) program. The evaluation also failed to assume

that the coating might fail by shoots of coating coming

off the pipe wall. It also failed to address such effects

on equipment other than the component cooling water heat

exchanger. Finally, the evaluation assumed that an

inservice inspection program would prevent further pipe

wall corrosion without specific knowledge that an

edequate inspection program existed or would bo  ;

developed.

l

I

The NRC inspector found that at approximately the same

time (1980) design change authorizations (DCA) 8809 and  :

8810 changed piping specifications MS-43B and MS-100 to l

read: "Plasite No. 7122 and its application are not  !

safety related." The justification assumed that the loss l

'

of the coating was not detrimental to the safety of the

plant. This assumption was in direct contrast to the

later Institute of Nuclear Power Operation's (INPO)

Significant Event Report (SER) 6883 and to NRC

Information Notice 85-24, issued in 1983 and 1985,  ;

respectively, which reported the failure of Plasite 7122

at another nuclear plant. The changes also failed to l

consider adverse effects that could be caused by a

coating procured and applied without 10 CFR Part 50, ,

'

Appendix B controls.

In September 1985. defective coatjng was found again in

the SWS, this time by e, paint foreman, and was documented

on TV Electric Operations Problem Report (PR)85-532,

eventually becoming construction deficiency CP-86-07. '

The inspector found that the NRC was not notified of this  !

'

deficiency until January 23, 1986, and then only after

weld failures due to corrosion were identified in SWS t

!

piping. This led to replacement of some piping and

repairs to Plasite using Belzona coating (PR 85-532 and f

TU Electric memorandum TIM-860454 dated April 1986).  ;

TU Electric's 50.55(e) report SDAR CP-86-07 (TXX-4672)  ;

I

stated that normally the coating failure was not

considered a safety issue and that inservice inspection h

would detect piping integrity loss and that defective ,

areas in the Plasite liner would be repaired with Belzona l

l

!

!

_ ._ ___

'

. .

Io

l 5

1 t

!

,

(a cora;nic coating). NRC IR 50-445/88-34; 50-446/88-30

questioned the control of the application of Bolzona and

l other coatings because all site applied coatings were

still considered nonsafety related.

On February 24, 1986, the initial written report for

deficiency CP-86-07 was sent to the NRC. After numerous

interim reports on Unit 1 SWS, in report TXX 88476 (dated

June 22, 1988) the applicant stated that the entire i

Plasite liner and Boltona coating would be removed. The

NRC inspector reviewed Stone & Webstor Engineering

Corporation (SWEC) correspondence along with a formal

corrosion report. Wisconsin Protective Coating

Corporation examined the Plasite pipe liner and found

extensive failures in the Plasito caused by porosity,  ;

excessive thickness, and lack of pipe surfnce

preparation. The corrosion report described 1/16 to

1/8-inch diameter blisters uniformly distributed and

spaced 3/8-inch apart and a few 1/2 to 3-inches apart.

All of the blisters broke while 7, crapping a putty knife

across them. Microbiologically induced corrosion (MIC)

was found in the carbon and stainless steel piping which

can cause pin holes and drips. The report stated that )

the shop or vendor applied coating was adherent with .

small blisters, but the site applied coatings werc  !

degrading. A few pits were present in Unit 1 piping  ;

'

which had been repaired with Belzona. The report str ..

in part, "Due to the Belzona coating repairs, mease' & nan-

of pits which may have formed previously is not i

possible." SWEC memo SWTU-4770 recommended Plasitt l

l

removal, a corrosion program to baseline ths extent l

corrosion, selected piping replacement, and that all

' piping should be weld repaired where it was less than the

specified thickness. SWEC Corrosion Report (SWTU-7749)

was issued in November 1987 and it extensively reported

on the liner failures, the corrosion mechanisms, and the ,

defects. SWEC letter (SWTU-7749 dated April 8, 1988)

specifically cautioned: "Excessive pipe wall i

thinning . . . must be avoided."

The NRC inspector concluded the followings

'

j (1) The first deficiency (CP-80-07) was inadequately

l

cvaluated relative to not applying QA/QC controls to

l the site procurement and application of Plasite and .

relative to possible sheet mode f ailure which might  ;

!

result in blockage of the coolant flow p:th.

l

'

(2) There is no evidence (in reports TXX-3229 and 3218

l

for CP-80-07) that the piping was extensively  ;

l inspected to determine if failures occurred in other '

!

areas of about 15000 linear feet of SWS piping in i

!

r

i

_ _ _ _ _ _ _ _ __

"

.

.

o

6

Unit 1 and 2. Therefore, the first daficiency was

concluded to be not reportable without knowing the

extent of the coating defects.

(3) There is no evidence that an effective inservice

inspection program was established in procedures

between 1980 and 1985. During the six years of

operation and lay up, the system degraded to the

'

point that the Plasite had to be removed.

(4) Concerning the second deficiency (CP-86-07), the NRC

was not notified in a timely manner. That is, the

NRC was not notified from September 1985 until i

January 23, 1986. On February 24, 1986,

deficiency CP-86-07 was reported (TXX-4711). The *

issue of untimely evaluation of noncenforming  !

'

conditions and reporting of significant construction

deficiencies under 50.55(e) wap addressed by the NRC f

in 1987 in Inspection Reports (IR) 50-445/87-07, t

50-446/87-06; IR 50-445/87-08; IR 50-445/87-10, t

50-446/87-08; and IR 50-445/87-18, 50-446/87-10. An

enforcement conference was held and a violation was -

issued.

(5) "TU Electric Task Description of the Protective

Coatings Program conducted by Ebasco Services, Inc., I

CPE-TD-EB-088," Revision 4, dated October 2, 1987,

'

did not describe the inspeccion of corrosion defects

performed by Ebasco engineers prior to the coating  !

being removed or the mapping of such defects for a l

baseline corrosion program. This is comparable to

TU Electric's failure to adequately preplan for the '

coating removal work. (See paragraph 3.a). (

(6) Neither the TU Electric deficiency reports for l

CP-86-07 nor a Corrective Action Report (CAR) *

addressed the cause of the SWS pipe liner *

deterioration between 1980 and 1985. Neither

addressed the lack of QA/QC controls (for site

4 upplied Plasite 7122), the inadequate evaluation by  ;

G&H (in light of information in SER 6883 and

IEN 85-24 which reported problems with Plasite), and r

the inadequate inservice inspections (which allowed l

extensive coating failure / corrosion of piping to

proceed). l

(7) TU Electric's final deficiency report (TXX-88476)  !

,

did not specifically discuss the localized attack on i

the SWS pipe wall that occurred under the blisters i

,

that were uniformly distributed through out the SWS  !

'

piping. j

.

'

\

!

. _ _

'

, .

.

e

7

The NRC belicycs the poor ovaluation of the coating

defects in 1980 and 1985, the poor evaluations of the

INPO SER and the NRC Information Notico, and the failura '

to take adequate corrective action reficct the problems

this utility had in that timo framo. Furthermore, the

NRC believes this poor work also set the stage for the

poor utility performance described in paragraph 3.

Doficiencies CP-80-07 and CP-86-07 are closed for Unit 1,

and 2 based on this inspection and the violation

identified in paragraph 3.

b. Follow-up on Previous Insp?pf, ion Findings (92701)

on May 2, 1988, the NRC 1: qtor observed work activity

relative to removing the i r. ; from Train A piping. NRC

s

Inspection Report 50-445/N J4; 50-446/88-30 (for May

1988) documented two unresolved and five open items  ;

concerning the coating removal and the identification and

disposition of defects (pitting from corrovien plus wall

thinning from sandblasting). The status of carb item

follows: -

(1) (Closed) Unresolved Item (445/8834-U-01;

446/8830-U-01): Downgrading of coating requirements

in Specifications MS-43B and MS-100 to a non-Q

classification. This concern is included in the

violation in paragraph 3.a.

l

'2) (open) Open Item (445/8834-o-02; 446/8830-0-02): l

Inadequate eveluation of INPO SER 6683 and '

IEN 85-24. Tno NRC inspector believes that the

Institute of Nuclear Power Operation significant

event report and NRC Information Notico should have i

been evaluated in greater depth considering the fact

that Plasits . Alures at Palo Verde had caused ,

clogging of safety-related equipment which was l

essential for safe plant shutcown. TU Electric

evaluations depended too heavily on a 1980 G&H

evaluation. .

The NRC inspector mot with the supervisors l

responsible for such ovaluations. Retrospectively, ;

they agreed that the evaluations, prior to 1986, l

were weak. They stated that they would review such '

notices received prior to 1986 to assure that others

were properly evaluated. This item remains open

pending the completion of this review. IEN 85-24,

however, is considered closed. See paragraph 4.

(3) (closed) Open Item (445/8834-o-03; 446/8830-o-03):

The inspector had identified a concern with the

l

i

i'

' *

.

.

8

Belzona repaired areas and the applicant's

identification of the surface defects inside the

piping.

This concern is included in the violation in

paragraph 3.c.

(4) (closed) open Item (445/8834-0-04; 446/8830-0-04):

NRC follow-up inspection of how residual areas of

Plasite liner were measured to ensure that the

ecceptance criteria were met. This concern is

included in the violation in paragraph 3.c.

(5) (Closed) Open Item (445/8834-0-05; 446/8830-0-05):

Possible blockage of safety-related equipment by

loose residual coating. The NRC inspector was

provided a semiquantitative measurement of residual

Plasite and Belzona. On June 20, 1988, the

inspector was informed that calculations showed that

if all the conting remaining in the SWS system after

the ccating removal failed at once, it would amount

to only a cupful and would be of no concern. The

calculations appear to be consistent with the

observations made by the NRC inspectors while

examining the piping internal surface. This item is

closed.

(6) (Closed) Unresolved Item (445/8834-U-06;

446/8830-U-06): The NRC inspector had identified a

concern that the sandblaster might remain in one

area long enough to cause wall thinning and that

this condition might go undetected. This concern is

included in the violation in paragraph 3.b.

(7) (closed) open Item (445/8834-0-07; 446/8830-0-07):

Need for a FSAR change to duscribe residual Plasite

and Belzona in the SWS. As stated in paragraph (5),

TU Electric found that about only a cupful of

Plasite and almost no Bolzona would remain after

liner removal. The NRC inspectcr determined that

this small amount is negligible and there is no need

to describe it in the FSAR. This item is closed.

3. Removal of Plasite/Belzona from Unit 1 SWS Piping (49061,

49063, 49065)

On May 2, 1988, the NRC inspector first observed O. B. Cannon

Company & Sons, Inc., (OBC) removing the Plasite 7122 liner

from the 10, 24, and 30-inch SWS piping. Section 9.2.1 and

Tabic 17A-1 of Section 17.0 of the FSAn describes and

classifies this system as safety related. The piping and

components are American Society of Mechanical Engineers (ASME)

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

. - .

'

3 .

.

9

Divisi'on 1,Section III, Class 3. The SWS system

operationally supports other ASME III, Class 2 and 3 systems.

That is, the SWS removes heat from the emergency diesel

generators and component cooling water system heat exchangers.

The SWS also supplies cooling water to the safety injection

centrifugal charging pump lube oil coolers and containment

spray pump bearing oil coolers. The SWS supplies cooling

water during normal operations and after a postulated loss of

coolant accident.

OBC began the work on Train A on April 11, 1988, and continued

until Train B was completed on July 8, 1988. OBC demobilized

and left the site on July 13, 1988. At this point TU Electric

considered the work on approximately 7400 linear feet of

Unit 1 SWS piping to be complete and acceptable.

Two types of equipment were used for the abrasive blast

operations: an automatic pull-through spin-blaster and

standard hand-held blasting nozzles. The spin-blaster blows

an abrasive grit through two centrifugally rotating blast ,

'

nozzles 180 degrees apart as the unit is pulled through the

10-inch piping. Standard blast nozzles (of varying sizes) are

hand-held to remove the Plasite from the 24- and 30-inch  !

piping. Respostively, the nominal vall thickness 10 , 24 ,

and 30-inch diameter are 0.365, 0.375, and 0.375 inches. The

manufacturer's minimum wall thickness is 0.875 times the

nominal except for the 30-inch pipe which is no less than

0.375 inches. In order to limit the Amount of metal removed, -

engineering had performed blast tests and concluded that only

0.012 inches of metal would be removed if the spin-blaster

stalled for two minutes. The tests also indicated that 0.005

inches (average) would be removed if the spin-blaster was

pulled at the rate of 3-inches per minute.

'

During this inspection, the NRC inspector reviewed the '

procurement documents, procedures, records concerning the "

removal of the coating, identification of corrosion defects,

and corrective action. In addition, the NRC inspector crawled

through about 900 feet of 24 and 30-inch piping between the

,

service water intake structure and the component cooling water

'

heat exchanger and observed multiple pitting defects caused by

corrosion. The worst defects were mainly at weld joints where

the coating had been applied by the site contractor but were

randomly found elsewhere in the SWS piping. Most of the

pitting was dish shaped. Where the NRC inspector entered the

piping in the Service Water Intake (SWI) structure,

TU Electric pointed out the pipe wall that had severely

corroded and the location of one pit which the depth of

-

exceeded 0.125 inches. Also, another weld which was

approximately 50 foot into the piping had a defect that

l appeared to have lack of penetration, corrosion and crosion.

l

l

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

--

.

• *

. .

.

10

TU Electric identified these prior to the NRC inspector's

inspection and generated a nonconformance .:eport (NCR).

Criterion II, "Quality Assurance Program," of Appendix B to

10 CFR 50 and Section 2.0, Revision 0, of the TU Electric QA

Manual states, in part: "The quality assurance program shall

provide control over activities affecting the quality of the

identified structures, systems, and components, to an extent

consistent with their importance to safety. Activities

affecting quality shall be accomplished under suitably

controlled conditions. Controlled conditions include the use

of appropriate equipment; suitable environmental conditions

for accomplishing the activity, such as adequate cleanness;

and assurance that all prerequisites for the given activity

have been satisfied. The program shall take into account the

need for special controls, processes, test equipment, tools,

and skills to attain the required quality, and the need for

verification of quality by inspection and test. The program

shall provide for indoctrination and training of personnel

performing activities affecting quality as necessary to assure

that suitable proficiency is achieved and maintained."

The inspector found that QA/QC and technical concerns and

requirements discussed in preprocurement meetings were not

addressed in the requisition or purchase order issued to

O. B. Cannon. Work and inspection activities began before

procedures were established, procedures for work and

inspection activities were inadequate, special process

procedures were not established to provide adequate controls,

inspection was inadequate, testing was not properly conducted,

nonconforming conditions were not identified and properly

evaluated, and corrective action was not taken relative to NRC

identified items. The NRC inspector is unaware of a QA or

technical document that consolidates and considers all of the

information, recommendations, and corrective actions in i

problem report 85-532 and the memorandum TIM-860454.

It appears that TU Electric failed to establish adequate QA/QC

and technical controls in procurement documents; failed to

provide adequate control of the special process; failed to

provide adequate inspections and process monitoring of the ,

work activity; and failed to take prompt and effective

corrective action in May, June, or July. This is an apparent

violation of Criterion II of Appendix B to 10 CFR Part 50

(445/8847-V-01). This violation is sapported by the following

examples: ,

a. Failure to Establish OA/OC and Technical Requirements

Criterion IV, "Procurement Document Control," of

Appendix B to 10 CFR Part 50 requires that applicable

regulatory requirements, design bases, and other

i

,

, . . - , - , - -

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

- - - -

enn,,,,_-_..n.,- , , - , - - _ - - , , - , , - ,

'

, ~'.

.

11

requirements which are necessary to assure adequate

quality are included or referenced in the documents for

the procurement of material, equipment, and services.

(1) Paragraph 6 2 and 6.2.8.2 of Engineering Purchasing

Procedure ECE-6.02-03, Revision 1, "Engineering

Review of Procurement Documents," states, in part,

"Figure 7.1 and the procurement requisition must

define all technical, qua '_ty assurance and

documentation requirements clearly and explicitly."

The technical and QA/QC requirements to control the

Plasite removal from SWS piping are discussed in the

minutes of the March 14, 1988, meeting for the

requisition and purchase order (PO) CP-794.

However, these requirements were not adequately

considered and included in the TQAR package,

services review summary, critical characteristics,

and verification plaa. At the subject meeting

concerns were expressed by attendees about not

meeting requirements such as: QA/QC level of

responsibi' y; noldpoints/ inspections; spinblaster

pull rates Other measurements and surveillance,

ultrasonic test (UT) examination of a small number

of points would not give a good overview of entire

system cleanliness and metal removal; decrease in

pressure over distance; UT of accessible areas to

verify metal removal; sample program procedures not

setup to give exact pull rato and pressure; control

of hand-held sandblaster; and need for additional UT

as QA/QC surveillance would not be available during

blasting; UT acceptance criteria; and the

possibility that grease might remain on the pipe

inner surface after blasting.

Purchase requisition 6R-350338 failed to clearly and

explicitly define the technical and QA requirements

in the TQAR package, and services review summary

(which also includes verification plans and critical

characteristic information per Procedure

ECE-6.02-I2, Revision 0).

'

The original QA requirement on page 2 of requisition

6R-350338 had been crossed out, initialed, and dated

(March 25, 1988). The deletion consisted of the

following: "The entire blasting operation shall be

under TU Electric's Quality Assurance Plan." On

page 4, another TU Electric QA responsibility was

,

i

similarly crossed out. On page 5, c note was added

relative to contacting a TU Electric representative

to schedule QA surveillance activities prior to

performing work. The TQAR package was not clear as

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

. _ . . - - _ _ . _.

'

E  :.,.

>.

12

to who had. full responsibility and did not

explicitly explain QA/QC and technical controls.

The TQAR form did not discuss all quality assurance

program requirements. TU Electric QA

responsibilities for the entire blasting operation

were eliminated from the requisition and purchase

order description (except for monitoring coating

removal). Instead, an Ebasco engineering group was

assigned the responsibility only for referenced

holdpoints for engineering inspections, without

clear requirements for quality assurance.

In the requisition package the Verification Plan

form was attached to a critical characteristics

Evaluation form. The verification plan was

orig!nally signed on March 21, 1988, and TU Electric

QA was listed as the responsible group that would

examine coating removal and critical defects.

However, on March 25, 1988, a procurement engineer

crossed out the listed QA responsible group and

replaced it with Ebasco protective coatings

engineer.

The approval block on requisition 6R-350338 had

10 signatures for original revie'; and approval. On

March 25, 1988, important changes were made to the

requisition; however, all of the original

signatories did not review and approve these

changes. The singular initial placed next to the

changes was not identified as one of the original

reviewers. In addition, althoulh Ebasco engineering

was assigned QA/QC and tecnnice..i responsibilities,

no Ebasco signature was found.

(2) Paragraph 3.1.c "Intended Use" of Purchasing

i Procedure 5.0-2, Revision 1, requires the

requisition to list applicable work document numbers

'

(of nonconformance/ deficiency / corrective action

reports and design change / modification documents).  ;

The NRC inspector determined that requisition

l

6R-350338 did not reference Problem Report 85-532;

nonconformance report (NCR) 88-00820, Revision 0, ,

dated January 12, 1988; and Design Modification i

"

Request-Construction Phase (DMRC) 88-1-020 which was

the approval to remove the Plasite liner. The 1985 l

problem report and the 1987 coating walkdowns I

'

contained corrosion defect mapping (qualitative in

some cases as depth was approximated), and other l

,

corrective actions that impacted the evaluation of

corrosion defects after coating removal in 1988.

.

--

- - - - - - - - - - "

__ ---~- .-- - - - - -

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

'^

.

.. .

. .

.

13

(3) Paragraph 6.2.5 "ASME Classification" of

ECE-6.02-03, Revision 1, states, in part, "The

(Responsible Engineer) RE shall determine the

appropriate ASME classification (Section III or XI)

which establishes the requirements to which a

particular service is to be performed."

The NRC inspector found that for requisition

6R-350338, the RE assumed that the coating removal

could not affect ASME Class 3 piping. This

conclusion was incorrect since metal was to be

removed from the pipe by blasting; therefore, the

ASME block on the requisition should have been

checked to inform the authorized nuclear inspector

that such work would be accomplished. Figure 7.1 of

Engineering Procedure ECE 6.02-03 contains three

blocks titled ASME III, ASME XI, and NA. This

figure was attached to the requisition and was

marked not applicable.

The NRC inspector reviewed the ASME Code, consulted

with other NRC inspectors, and held meetings with

the State of Texas and authorized nuclear inspectors

(ANI). The NRC concluded that TU Electric should

have notified the ANI on site of the intent to

remove metal from an ASME piping system to allow

voluntary or required ANI inspections. The

following is the basis for that conclusion.

I Section IWA-2140 of.ASME Section XI, dated July 1,

1974, statos: "The owner shall arrange for an

Inspector to have access to all parts of the plant

necessary for making the required inspection. The

owner shall koop the Inspector informed of the

progress of the preparatory work necessary to permit

! inspections and shall notify him reasonably in

advance when the components will be ready for any

required inspection."

Section IWA-2120 of ASME Section XI, same dato,

i states: "It is the duty of the Inspector to assure

himself that the examinations and tests required for

Class 3 components and systems (IWD-1000) have been

conducted and the results recorded . . . . The

Inspector shall review the repair program to

determino compliance with the requirements of this

Division."

,

I The NRC inspector interviewed the ANI for Section XI

and datormined that he was not informed of

preparatory work; therefore, was not provided an

I

opportunity to perform voluntary or required

I

l

,

W'

.'

.

.

.

'

.-

14

inspections of the sandblasting process to assure

.that excessive metal was not removed along with the

plasite coating. He was also not made aware of a

0.012-inch metal removal allowance. No work order

or traveler was processed. Therefore, no ANI input

or inspection of the recently completed work on

Train A was possibl.e.

This failure to properly establish and define all QA/QC

and technical requirements and reference the work

'

documents is a violation of Criterion IV of Appendix B to

10 CFR Part 50 and Section 4.0, Revision 0, of

TU Electric QA Manual (445/8847-V-01.a).

b. Failure to Control Special Processes

Criterion IX of Appendix B to 10 CFR 50 requires that

measures be established to assure that special processes

are controlled and accomplished by qualified personnel

using qualified procedures in accordance with applicable

codes, standards, specifications, criteria, and other

special requirements.

The NRC inspector found that the measures established to

control the special process of removing the coating from

the SWS piping were inadequate. As a result a 1/2-inch

hole was sandblasted through the pipe wall in one place

and several other deep sandblasting indentations were

made in the 10-inch piping.

(1) O. B. Cannon Procedure, Construction Procedure

QCP-1, Revision 2, "Coating Removal by Abrasive

Blasting of Interior of Station Service Water

Piping," dated April 18, 1988, was originally issued

on March 23, 1988. The NRC inspector found that the

Procedure QCP-1, Revision 2, does not prescribe any

of the specific paramotors or give detailed

instructions concerning required air pressure, type

abrasive, required rate for pulling the spin blaster

l

through 10-inch piping, rate of change of pulling l

versus distance, and qualification of operators l

based o.1 parameters developed by Sandblast Tests.

(2) Construction and engineering Procedures ACP-10.1,

,

Revision 1, "Preparation, Approval, and Control of

Construction Operation Travelers," and ECC 2.13-5,

Revision 4, "Construction Traveler," require the use

of travelers to document the disassembly of

previously completed components by providing

instructions and points that should be inspected,

verified, monitored, and/or witnessed. ihe NRC

(

,

l

.

-

p-

.

.

• .

.

15

(.

.

inspector found that no traveler or work order was

used to control this work.

(3) The NRC inspector determined that paragraph 6.9 of

EME 3.21-08, "Engineering Verification of Protective

Coatings Applied to Steel Surfaces Subject to

Immersion Service," Revision 0, was inadequate

because it did not discuss how the rate of

sandblasting would be monitored. The NRC inspector

found that the engineers who tested the various

techniques had verbally instructed OBC personnel to

contact engineering before exceeding the blasting

time in an area or if the rate of travel dropped

below a minimum rate when traveling through the

10-inch piping or if blaster problems were

encountered. Yet, to the contrary, a TU Electric

memorandum NP-6633 dated March 18, 1988, stated that

no further monitoring of pipe wall thinning was

necessary.

(4) DCA 73794, dated April 26, 1988, changed the

instructions and the acceptance criteria in OBC

Procedure QCP-1, Revision 2 dated April 18, 1988.

TU Electric QA Manual, Section 6.0, Revision 0,

requires that such changes be controlled. The DCA

changed the critoria from "The abrasive blasting

operation shall leave the surface of all piping free

of paint . . . ." to ". . . except slight shadows,

streaks, discolorations from residual paint or small

tightly adherent coated areas; i.e., 10-1/4"

diameter speck 1cs por square foot or areas 1/4" in

dimension of length no longer than the corresponding

internal pipo circumference."

The NRC inspector found that paragraph 9.1 of QCP-1

was not revised to include the now criteria. All

work was accomplished without this criteria being

in the O. B. Cannon procedure.

(5) The NRC inspector learned ehat nondestructivo

ultrasonic testing had been performed through the

paint on the outside of SWS piping in order to

measure pitting depth on the inside of piping

without considering that the thickness of paint had

the potential to cause UT measurement error. In a

June 20, 1988, meeting, TU Electric confirmed that

corrosion defects had been measured by UT through

the paint.

Subsequently, TU Electric selected detects in the

wall previously measured and performed UT to

dotormine the depth of the defects. When

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

(,,--

_

, . .

.

16

TU Electric retested without paint, they found an

error was introduced by the paint. This practice is

similar to a violation that was previously

identified and documented in NRC IR 50-445/88-20;

50-446/88-17 dated April 22, 1988.

The failure to estan..sh adequate controls of the special

process for the coating removal is a violation of

Criterion IX of Appendix B to 10 CFR Part 50 and

TU Electric QA Manual, Section 9.0, Revision 0

(445/8847-V-01.b).

c. Failure to Provide Adequate OA/OC Procedures

Criterion X of Appendix B requires a program for

inspection of activities affecting quality to bc

established and executed to verify conformance with

documented instructions, procedures, and drawings for

accomplishing the activity. It requires inspection and

process monitoring when control is inadequate without

both.

Comanche Peak Engineering Procedure EME 3.21-08 is the

procedure that implements the technical and QA

requirements of Appendix B to 10 CFR.Part 50. It covers

qualification of engineers, verification records,

deficiency reporting, and sandblasting to bare metal.

This procedure was originally developed (September 1987)

to verify the application of coatings to steel surfaces

subject to immersion service but Engineering Document

Change Notice (EDCN) No. 3, dated May 6, 1988, added

paragraph 6.9 to reflect the removal of coating from the

SWS piping. The paragraph concerning the removal of

Plasite is approximately one-half page long. The NRC

inspector determined that the subject proceduro and

implementation woro inadequate as follows:

(1) Requisition 6R-350338 package for OBC services

stated that residual coating and damage to piping

caused by pitting or rust were critical defects.

The NRC inspector determined that paragraph 6.9 of

EME 3.21-98 did not discuss or describe how such

critical defects would be identified and no criteria

for acceptance was provided.

The NRC inspector learned that SWEC engineers

performed visual inspections without procedures.

In addition to having no critoria in a procedure for

inspecting defects inside the 10-inch piping, the

NRC inspector found that SWEC engineers had visually

inspected the pipe for defects using video tapes of

the pipe interior and had accepted the piping and

.

f!T

'

..

i* *

7 ,

17

,

returned it to service. A 1/2-inch hole in tha pipe

wall and several other defects caused by the blaster

were not detected when SWEC engineers reviewed the

video tapes of the inside:of Train A piping

(10-inch). The hole in the pipe wall was identified

when the SWS piping was filled with water. The SWEC

ongineers performing the visual inspection were not

certified to ANSI N45.2.6 requirements in regards to

special inspection methods.

On August 1, 1988, TU Electric stated that they had

found that they had misidentified the video tapes-

for the 10-inch SWS piping entering and exiting the

diesel generator building. The defect locations

were reversed when the video tapes were reviewed.

(2) On July 18, 1988, the NRC inspector discussed the

TU Electric inspection of Train B piping with

TU Electric QC and licensing compliance personnel.

The inspector believes the current inspection of the

piping pits and blister indentations is not in

accordance with ASME Section.XI. This code requires

a visual inspection to identify defects and

surface / volumetric examination to assure that the

extent of the defect is known. To date only

measurements with a depth gauge have been made to

measure possible violations of the manufacturers and

code minimum wall thickness.

(3) EDCN No. 3 to EME 3.21-08 was approved May 6, 1988; i

however, the inspection process was initiated about

a month earlier (April 11, 1988). The NRC inspector

observed work being completed on May 2, 1988, before

the QC procedure was established or implemented. i

(4) Paragraph 6.9 of Proceduro EME 3.21-08 stated the

removal of coating shall be inspected in accordance

with Specification MS-100. The NRC inspector notes

that this specification contains a general section

on QA/QC and an Appendix C which addressos Level D

cleanliness. This Appendix also prescribes the size

of areas of residual coating that can remain on the

inside of the piping.

The NRC inspector determined that during the removal

of coating from Train A piping an accurate method

for measuring the residual coating remains in the

piping did not exist. At a June 20, 1988, meeting

on SWS coating removal, TU Electric made a

presentation and specifically addressed their

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

_ _ _

e ~ O

74 ,

' *

'

.

18

methodology for measuring the remaining coating.

They showed that a clear plastic "mockup" could

simulate areas of residual coating that could be

compared with observations by video camera / tape

inside the 10-inch piping. However, according to

site supervision the methodology to measure the

residual coating inside the 10-inch piping was

developed about 2 1/2 months after work began.

Furthermore, a procedure which would include a.

plastic overlay as a standard to judge the size of

the residual coating areas recorded on the video

tape had not been prepared.

(5) The NRC inspector found no test procedure 1 for the

O. B. Cannon sandblast-tests 1-4. However,

paragraph 5.4 of O. B. Cannon Procedure QCP-1

discusses sandblast testing.

The failure to provide adequate inspection procedures and

process monitoring to assure the quality of the special

process and the piping system is a violation of

Criterion X of Appendix B to-10 CFR Part 50 and

Section 10.0, Revision 0, of TU Electric QA Manual

(445/8847-V-01.c).

d. Failure to Take Corrective Action

Criterion XVI of Appendix B to 10 CFR Part 50 and

TU Electric QA Manual, Section 16.0, Revicion 0, requires

that conditions adverso to quality be promptly identified

and corrected.

The NRC inspector first raised concerns to TU Electric

project management in regards to the adequacy of coating

removal QA/QC controls on May 2, 1988. On May 27, 1988,

the NRC inspector completed a review of the SWEC

Corrosion Report and raised specific technical concerns

about the removal of the coating and corrosion defects:

(1) was a metallurgical analysis of pitting performed to

deterraine whether microscopic cracks exist, (2) woro all

such defects identified and measured to dotormine if

manufacturer's minimum wall was violated, (3) was the

control of the sandblasting process adequate to prevent

wall thinning / damage, (4) was the amount of coating

remaining after gritblasting within specification, and

(5) where were QA/QC controls related to coating

removal / defect identification proscribed? The control of

the gritblasting process was a contral issue.

'

On or about May 30, 1988, the NRC inspector reviewed the

construction deficiency files (CP-80-07 and CP-86-07

previously discussed) and raised concerns about the

.

b.o.

.

.

,

19

downgrading of G&H specifications MS-43B and MS-100.

(See paragraph 2.) Specifically, the inspector's concern

was the effect that nonsafety-related material (i.c., the

introduccion of contaminants) could have on

safety-related systems. The inspector was also concerned

that TU Electric project management seemed to think that

because the coating was non-Q, work activities including

the coating removal were, therefore, non-Q without

considering the effects on Q-systems. This concern

continued on through meetings on June 20, July 13, and

August 2, 1988.

The NRC inspector observed no comprehensive and effcetive

corrective action concerning the lack of QA/QC program

and technical controls previously identified by the NRC.

At all the meetings, the NRC inspector was informed by

TU Electric that the coating removal and defect

identification were properly controlled.

On August 1, 1988, TU Electric notified the NRC that they

had detected a leak in a section of 10-inch SWS piping.

It was removed and sectioned for visual inspectjun. It

appears that the leak was caused by the sandblaster l

because there was one 1/2-inch diameter hole through the  !

0.365 wall thickness and 180 degrees around the

circumference of the insido diameter there was another

defect where the wall was extensively worn. This would

duplicate positions on either end of the spinblast nozzle

if it stuck and ceased te spin. The defect configuration

also suggests that if the blastor nozzlo stuck, the

pulling was stopped or else there would have been a

longitudinal groovo instead of a round hole. By

August 2, 1988, other similar defects were found, but the

total number and locations woro not available.

The failuro to promptly identify and correct the QA/QC

and technical deficienclos identified by the NRC is a

violation of Critorion XVI (445/8847-V-01.d).

Since the end of the inspection period, pieces of Train A

piping have been cut out and sectioned for detailed

examination. See the photos on page 21.

c. New Open Items Concerning SWS piping

(1) The NRC inspector learned at the end of the July

1988 inspection period that minimum wall violations

on the outside of the piping had not been evaluated.

During construction, surface defecta had been

identified on the outside of the SWS piping and

dispositioned "uso-as-is." Now, internal defects

have boon identified. It is not clear how SWEC

( c o o

. .

,

20

calcul&ted the stresses and returned Train A to

service when they did not know if the outside

defects and the inside defects might align and be

additive. SWEC was reviewing this data after the

SWS was returned to service. Since all new defects

were net mapped, it is unclear as to how a corrosion

monitoring program can be established. This item is

open pending the completion of the TU Electric

review of this matter (445/8847-0-02).

(2) Paragraph 3.1.f of Purchasing Procedure NPI-5.0-2

requires the entry of a QA code on the requisition

by engineering. The NRC inspector believes that the

"quality" determination should be the responsibility

of a quality assurance organization, not an

engineering responsibility. If the engineering

organization specifies the level of quality to be

applied, the decision could be affected by cost and

schedule considerations.- Also, the designating

engineer may not have sufficient expertise in

quality assurance to know the appropriate level of

quality assurance required. This is an open item

pending TU Electric consideration (445/8847-0-03).

4. Action on NRC Informatio,n__ Notice (IN) 85-24 - Plasite Liner

(92700)

The NRC inspector reviewed the files and action taken

concerning IN 85-24 which pertained to Plasite 7122 failure in

SWS piping. This item is closed based on the inspection

described above.

5. Open Items

Open items are matters which have been discussed with the

applicant, which will be reviewed further by the inspector,

and which involve some action on the part of the NRC or

applicant or both. Two e s en items disclosed during the

inspection are discussed in paragraph 3.e.fi) and (2).

6. Exit Meeting (30703)

An exit meeting was conducted August 2, 1988, with the

applicant's representatives identified in paragraph 1 of this

report. No written material was provided to the applicant by

the inspectors during this reporting period. The applicant

did not identify att proprietary any of the materials provided

to or reviewed by the inspectors during this inspection.

l

During this meeting, the NRC inspectors summarized the scope

and findings of the inspection.

l

l

('

.

'

.

,

21

'

,s . e

.$

,

• ' 4. .. .

,

'

f \

'

.6  :

l.-

..-- $..

'

,

,-

.'

.

.

.

,

i

.o

...i

• s

s'

..  %

Photo No. 1 ,

l

Hole viewed from outside of l

sectioned 10-inch diameter pipe. '

i

j \- _ l

8

' ,

l .

,

, o

I l

1

'

,

l

!

"

..

/

!

'

~

l

_

--

!

Photo No. 2

i

Hole viewed from inside of

sectioned 10-inch diameter pipe, I

l

l

l

l

_ __ __

_ , ,