IR 05000445/1989023
| ML20006C522 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 06/16/1989 |
| From: | Livermore H, Phillips H Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20006C521 | List: |
| References | |
| 50-445-89-23, 50-446-89-23, EA-88-310, NUDOCS 9002080161 | |
| Preceding documents: |
|
| Download: ML20006C522 (27) | |
Text
j 1.
,
,
-
.
r
,
U.
S. NUCLEAR REGULATORY COMMISSION
'
OFFICE OF NUCLEAR REACTOR REGULATION
,
NRC Inspection Report: _ 50-445/89-23 Permits: CPPR-126
50-446/89-23 CPPR-127 f
Dockets: 50-445 Category: A2 50-446
.
>
^
Construction Permit Expiration Datos:
Unit 1: August 1, 1991
'
Unit 2: August 1, 1992 Applicant:
TU Electric Skyway Tower
'400 North Olivo Street Lock Box 81
Dallas, Texas 75201
.
Facility Name:
Comancho Peak Steam Electric Station (CPSES),
Units 1 & 2
-
Inspection Att-Comancho Peak Site, Glen 1 Rose, Texas Inspection conducted:
April 5 through May 2, 1989
'
0*/4-8'I Inspector:
~ Phillips,' Senior Resident Inspector Dato
-
-
H. S.
Construction Reviewed by:
IA M67 f- -
b "/[-f H. H.
Livermoro, Lead Senior Inspector Date l
l 9002080161 900126 ADOCK0500ggg5 DR
-
m
+
c._
.
c
,
.
.
Inspection Summary:
Inspection Conducted:
April 5 through May 2, 1989 (Report 50-445/89-23: 50-446/89-23)
Areas Inspected: Unannounced, resident safety inspection included:
(1) exit meeting with management, (2) applicant action on previous
-,
findings, (3) follow-up on violations, (4) ovaluation of corrective action on enforcement, (5) review of component cooling water heat exchanger work, (6) repair of diesel generator heat exchangers, (7) vendor services to measure steam generator nozzles, and (8) application /romoval of coatings from dioscl generator tanks.
Results:
Within the areas inspected, one apparent violation was identified:
failure to provido accurato and complete information relative to' corrective action concerning Enforcement Action EA 88-310, paragraph 4.b; and additional examples of violations similar to those identified in EA 88-310, paragraphs 5, 6, 7, and 8.
An enforcement conference will be scheduled to discuss-these findings.
,
N
'
,, -
-
- -
.-
.
,
s a
i
DETAILS 1.
Persons Contacted
- R. W. Ackley, Jr., Director, CECO
- G.
K. Afflorbach, ASM Startup, TU Electric
- M.
Axelrad, Newman and Holtzinger
- J.
L. Barker, Manager, Engineering Assurance, TU Electric
- D.
P. Barry, Senior Manager, Engineering, Stone and Webster Engineering Corporation (SWEC)
- J. W. Beck, Vice President, Nuclear Engineering, TU Electric
- 0.
Bhatty, Issuo Interface Coordinator, TU Electric j
- M. R. Blevins, Manager, Technical Support, TU Electric j
- H.
D. Bruner, Senior Vice President, TU Electric
'
- J.
H. Buck, Senior Review Team, IAG
- J. T. Conly, APE-Licensing, SWEC
- R.
J. Daly, Manager, Startup, TU Electric
- J. W..Donahue, Operations Manager, TU Electric
- D.
E. Devincy, Deputy Director, Quality Assurance (QA),
TU Electric
-
- D. M. Ehat, Consultant, TU Electric
- J.
C. Finneran, Jr., Manager, Civil Engineering, TU Electric
- C.
A. Fonseca, Deputy Director, CECO s
- W.
G. Guldemond, Manager of Sito Licensing, TU Electric i
- P.
E. Halstead, QC Manager, TU Electric
- T.
L. Heatherly, Licensing Compliance Engineer, TU Electric-
.
- C, B. Hogg, Engineering Manager, TU Electric i
- T.
A. Hope, Licensing, TU Electric
- A.
Husain, Director, Reactor Engineering, TU Electric
- R.-T.
Jenkins, Manager, Mechanical Engineering, TU Electric
- J.
J. Kelley, Manager, Plant Operations, TU Electric
,
- O.
W. Lowe, Director of, Engineering, TU Electric
- F. W. Maddon, Mechanical Engineering Manager, TU Electric
- D.
M.fMcAfee, Manager, QA, TU Electric
- S.
G. McBoc, NRC Interface, TU Electric
- J. W. Muffett, Manager of Engineering, TU Electric
- E.
F. Ottney, Program Manager, CASE
- S.. S. Palmer, Project Manager, TU Electric
- P.
W.
Pellette, Operations, TU Electric
- D.
M. Reynorson, Director of Construction, TU Electric
- A.
H. Saunders, EA Evaluations Manager, TU Electric
- A.
B. Scott, Vice President, Nuclear Operations, TU Electric
- B.
J. Sewell, TU Materials Coordinator Manager, TU Electric
,
- J.
C.
Smith, Plant Operations Staff, TU Electric
- R.
L. Spence, TU/QA Senior Advisor, TU Electric
,
- M.
D.
Skaggs, CPE, Mechanical, TU Electric L
- P.
B.
Stevens, Manager, Electrical Engineering, TU Electric l
- J.
F.
Streeter, Director, QA,,TU Electric l
- C.
L. Terry, Unit 1 Project Manager, TU Electric l
- M.
A. Thero, CASE Intern
- -
.
, = - *
'
.
,
.
4
- 0. L. Thero, QTC Consultant to CASE
- T. G. Tyler, Director of Projects, TU Electric
- R. D. Walker, Manager of Nuclear Licensing, TU Electric
,
- R.
G. Withrow, EA Systems Manager, TU Electric
'l The NRC inspectors also interviewod other applicant employees during this inspection period.
- Denotes personnel present at the May 2, 1989, exit meeting.
2.
Applicant Action on Previous Inspection Findings (92701)
a.
(Closed) Open Item (445/8908-0-01):
The documentation file relating to the auxiliary feedwater motor fans being installed backwards contained two nonconformance reports (NCRs) not previously reviewed by the NRC.
The NCRs described arcing between the fans and brass rings on the rotor winding.
The arcing was attributed to the condition of having reversed fans.
The W analysis concluded that the reversed fans would not cause motor failure or reduce the level of safety during operations.
The NRC questioned whether the W analysis included the NCR conditions.
During this inspection, TU Electric met with the NRC and presented additional information.
That is, W reevaluated the NCR conditions in connectior with the fan reversal issue and concluded that their original analysis was not impacted by these NCRs.
This item is closed.
Note:
In NRC Inspection Report 50-445/89-08; 50-446/89-08 the tracking number for this item contained a typographical error.
The number shown above (445/8908-0-01) corrects the number error (445/8808-0-01).
b.
(Closed) Open Item (445/8908-0-03):
No NCR was available on stripped threads in bearing holes for an auxiliary foodwater 'AFW) motor.
The NRC inspector confirmed that operations / maintenance had issued NCR 88-03638, Revision O.
This item is closed, c.
(Closed) Open Item (445/8908-0-04):
QC did not verify temperature contrcl during the welding on the AFW rotor bar assembly.
The maintenance instruction stated that extremo caution must be taken not to concentrate an
-
excessive amount of heat on the rotor bar assembly.
The NRC inspector was concerned that QC had not verified that the instruction was followed.
TU Electric met with the NRC to provide information about this concern.
The NRC inspector asked what type of material was used and what heat input controls were
l
.:
e
,.
.
5 necessary..TU Electric did not have a welding engineer present, so a subsequent meeting had to be arranged.
During that subsequent meeting TU Electric revealed that an electrical engineer had inserted the caution about heat j
input.
The wolding spenislist identified the material as a low carbon steel and provided information about the
!
energy input.
The NRC inspector has no further questions, i
This item is closed.
l d.
(open) Unresolved Item (445/8908-U-02):
TU Electric l
maintenance personnel substituted Grade 5 carbon steel t
bolts for the siliconc bronze bolts that secured AFW fans in the motors.
The NRC inspector learned that a W field
representative had directed this material change 5ecause
past experience had shown that the siliconc bronze bolts
were cracking and failing because of fatigue.
The NRC
-inspector stated that this material change was improperly authorized unless an engineering change had authorized the t
change.
The inspector also questioned if this was a weakness in the maintenance program.
.
During this inspection, TU Electric met with the NRC and made a presentation on this subject.
They admitted that
'
the material change was not authorized.
They were unable to find the W field representative as he was a consultant I
and performod'this work for W.
TU Electric submitted a j
large amount of material on this subject.
The NRC
inspector considers this to be a potential violation; t
however, this item will remain unresolved pending the completion of the NRC evaluation.-
(
3.
Follow-up on Violations (92702)
a.
(Open) Violation (445/8847-V-Ola):
Failure to establish
'
QA and technical requirements-in procurement documents for-coating removal from service water system (SWS) piping.
(open) Violation (445/8847-V-Olb):
Failure to establish b..
adoquato controls for the coating removal process.
[
t c.
(Open) Violation (445/8847-V-Olc):
Failure to provido adequate QA/QC procedures for the coating removal process.
!
d.
(open) Violation (445/8847-V-Old):
Failure to take corrective action relative to coating problems and coating removal.
'
The abovo violations were doucmonted as Enforcement Action (EA)88-310 in an NRC letter to TU Electric dated January 9, 1989.
'
TU Electric's response to the violations is discussed below in paragraph 4.
-
,
..,.,.,
. -,,
,
--
.-
-
,
0".
...
',c
!
.
-
..
j J
I l
4.
Evaluation of TU Electric Corrective Action on Enforcement
]
(35065, 49063, 49065, 92702)
j i
a.
Background
]
NRC Inspection Report (50-445/88-34; 50-446/88-30 for-
.
May 1988) identifies open items concerning the
'
removal of Plasite 7122 from SWS and potential wall thinning by sandblasting.
NRC Inspection Report 50-445/88-47; 50-446/88-42 was
.
_
issued on September 2, 1988, and identified one apparent violation (breakdown in the QA program relative to the removal of the protective liner from the SWS piping).
On September 13, 1988, TU Electric responded to the
.
findings at a public meeting on sito.
i On November 9, 1988, the NRC held an Enforcement i
.
Conference at the NRC's Rockville, Maryland, office.
TU Electric made a presentation and provided a
.
handout.
The handout was attached to NRC Notico of Violation 50-445/88-47; 50-446/88-42 dated January 9,
1989.
The handout stated that problems in the implomentation of QA program requirements occurred but were isolated and were not significant.
The
handout also stated that corrective actions were
.
completed and included (1) evaluating / replacing worst
'
damaged piping, (2) evaluated other spinblast indications with satisfactory results, (3) performed critical self-ovaluation, (4) reviewed other Code V services (other than SWS) procurements with satisfactory results, and (5) reviewed previous CPSES onforcement action and found no precursor events.
.
On. January 9, 1989, NRC issued the NOV for NRC
.
Inspection' Report 50-445/88-47; 50-446/88-42.
It stated that after careful review of information, the NRC decided that four Severity Level IV violations were appropriato instead of the one Severity Level c
III that was initially considered.
It also stated that the NRC was concerned that once it was
recognized that the coating removal process needed to r
t
.be modified, adequate measures were not taken to inspect for damage caused by early process problems.
The NRC letter stated that if the violations were not fully corrected they may lead to more significant ConCOrns.
On February 8, 1989, TU Electric issued their
.
responso (TXX-89070) with one attachment to the NRC.
i
,
-
.
y c; -
.
.
,,4
/
b.
Incomplete and Inadequate Information Provided Concernino
.Concerning EA 88-310 The NRC inspector reviewed the TU Electric Enforcement Document which was docketed with the NRC Enforcement.
Action EA<88-310 and Notice of Violation 50-445/88-47, 50-446/88-42.
TU Electric' Response TXX-89070Lto the
'
enforcement action was also reviewed.
.These documents
,
provided TU Electric's overall response.
At the
'
'
Enforcement Conference information was provided to the NRC which advocated a reduction in the proposed severity level
'
'
from Level II2 to' Level IV and V.
During the enforcement conference TU Electric made several statements, some of c
which are discussed below, to show that QA/QC deficiencies
,
identified by-the NRC were not program breakdowns and,
>
therefore, were not significant. The NRC inspector found that specific information related to the results of TU
>
Electric's review of other. Code V procured services was
!
not included in the information provided to the NRC.
l Thus, the information provided by TU Electric concerning i
the enforcement action was incomplete and apparently e
'
inaccurate. 'Further, the inspector believes that other information provided by TU Electric during the enforcement conference.was misleading.and misrepresented the deficiencies encountered during the SWS coating removal i
project.
NRC' Regulation 10 CFR Part 50.9 requires the i
applicant / licensee to provide accurate, complete, and
significant information to the NRC.
l
'
(1)
TU Electric stated during the enforcement conference that they had "(r]eviewed other Code V services activities with satisfactory Results."
Contrary to the above, the NRC inspector determined
-
that TU Electric failed to provide significant
information concerning the results of their review of six code V service procurements which would have i
shown that these code V procurements for services were not satisfactory.
These deficiencies are described in TU Electric memorandum NE 22156 dated September 30, 1988.
That memorandum indicated that
?
there were deficiencies in the six Code V service procurements.
These deficiencies were similar to the
,
code V procurement for service water system piping coating removal.
Further, this information was not provided to the NRC in the meeting on September 13,
,
1988, in TU Electric Engineering Report ER-ME-19, Revision 0, or in the TU Electric Enforcement
'
Conference Document handout.
L
'
-
, - -
.
..
.
8 The deficiencies docume.nted in Memorandum NE-22156 were:
Except for two purchase orders for vendor
.
services (661-74340 and 661-74038), the
-
procurement documents did not clearly define the relationship between the organizations involved and the TU Electric QA Program.
.
None of the procurements (requisitions or
.
purchase orders) addressed the identification and disposition of nonconforming conditions.
Verification Plans (engineering and QC
.
inspection points) for each requisition lacked detail.
Work on the component cooling water heat
.
exchangers should have fallen under the auspices of ASME Section XI.
Work on the steam generators was performed
.
before the purchase order was approved.
The procurement documents in general were of
.
similar quality to those associated with Service Water coating removal.
(2)
TU Electric's response to EA 88-310 (TKX-89070 dated February 9, 1989), stated in part that "six previous Code V services procurements were identified.
..
review of the associated inspection and surveillance reports showed that the requisitioned work was successfully completed and documented. "
Contrary to the above, the NRC inspector interviewed the TU Electric representative who coordinated the response to those deficiencies.
The NRC ins actor e
questioned the apparent contridiction between the response (TXX-89070) and the internal memorandum (NE-22156).
TU Electric responded that the inspection and surveillance reports showed that the work was successfully completed and documented.
When asked if QA records were reviewed, the TU Electric representative responded that they had not.
Subsequently, during a meeting on May 1, 1989, TU Electric pointed out that procurement documents and some work orders had been reviewed, while reviewing project files.
.
.. _ ______
4 jhkI[T C
~"
~
~
~
_. 'f9fEEi l
?.:
q.
.
,
.)-h ' w ;
,
9'
"
MNJ Electric had not performed'an adequate review'of i
,
'
the other Code V service-procurements to support the
- t
.
conclusions they presented.- In addition, during the
'
course of the review of this material,1 he inspector-t identified additional deficiencin; associated with
- '
the subject procurement, as described in more detail.
in' paragraphs 5'and 7.
(3)
TU Electric stated during.the enforcement conference,
,
in part, that
"[d)amage did not' occur following i
modifications.to spinblaster."
j i
Contrary toLthe above, the NRC determined that damage
.f occurred during coating removal-of Train B after
modifications were made to the spinblaster after damage was found in Train A of the SWS in July,1988.
<
In March 1989, three NRC1 inspectors performed a field
inspection to view video tapes of Train B'after-coating removal.- Defects caused by-the spinblaster were observed.in, Train B (Spool SW-1-SB-7-14A-8 frame
,
1484).,Although the video tapes.of Train A.and i
l L
Train B'had been misidentified during the video
,
review,(blasting marks on the Train B piping were, f
'
. confirmed by the inspectors.
The TU Electric coating specialist was present.when the NRC viewed Train B E
tapescandLthe NRC, pointed to the marks that were apparently made by'the spinblaster.
When directly.
-
asked if they appeared to be.spinbJaster marks,-he
.
agreed that they appeared to be spinblaster marks.
i Th'e three. items described'above are apparent
. violations of.10'CFR Part. 50.9 (441/8923-V-Ol; 446/8923-V-01).
L 5.
. Review of Component Cooling Water Heat Exchanger Work (50073, i
'
50075)'
s During NRC Inspection Report 50-445/89-16; 50-446/89-16, the NRC inspector performed a follow-up inspection to verify the
,
corrective actions taken for Code V service procurements, as
- described in TXX-89070.
Records at the procurement vault,
' construction QA records' vault, and the QA Records Center were
L reviewed.
The QA-Records Center personnel provided the NRC inspector with a computer run which listed all QA records available for the component cooling water (CCW) heat exchangers.
(One of the previous six Code V procurements was for work on the CCW heat exchangers.)
Records for J
cpl-CCAHHX-02 were selected for review.
About March 29, 1989, l
the NRC inspector met with TU Electric to discuss the results j
of'the NRC review.
TU Electric was informed that the available u.
'
.
.
..
.
..
.
....
...
.
,
<
et 92. :
i
,
i
"
p n'=*
[
1 records were insufficient.to demonstrate that the work
activitics were properly _ controlled,-conducted, and documented i
as stated in TXX-89070 and TU Memorandum NE-22156.
TU Electric
'
was provided specific questions regarding how the critoria of-10 CFR 50', Appendix B, were implemented.
TU Electric was. asked
-
to providefadditional records to demonstrate. implementation of
'
the critoria necessary to control work.
TU Electric was unable to locate'or produce additional'QA records-before the exit
'
meeting'on April 4, 1989, and the inspection was not completed.
During this inspection period, the NRC inspector completed the inspection discussed in the previous paragraph.
TU Electric j
nevor-provided answers to the questions concerning which it criteriafwere applicable for each procurement-and how they-
.
complied with those criteria.
As a result, the NRC inspector-performed a comprehensive review to obtain the answers.
During i
March and April 1989, procurement documents, work procedures,
inspection reports, QA contractor surveillance reports, startup work authorizations, work orders, correspondence, and-
-
miscellaneous. records were reviewed to evaluate how work was done, inspected, and documented by TU Electric to show that the
,
QA program was-implemented in accordance_with 10 CFR 50, t
' Appendix B, QA requirements.
The NRC inspector found'that the QA program was not adequately implemented for four.-_ Code V
,
procurements for vendor services for work on the Unit 1 and 2
't CCW heat exchangers.
Multiple examples of inadequacies and program deficiencies similar to those identified for SWS
"
coating removal were identified.
The NRC met with TU Electric on April 28, 1989, and provided the findings similar to those that were provided in March 1989;
'
that is, the QA program was not adequately implemented.
On May 1, 1989, TU Electric requested another meeting during which they concluded that the program for. procured services-was-adequate:and was appropriately implemented except for the
.*
specification issues and the contract for steam generators
.
L which was marked nonsafety.
However, the NRC inspector
~
identified deficiencies in these activities, as follows:
!
(
a.
Chemical. Cleaning of CCW Heat Exchangers In 1985, serious corrosion problems were identified insido l
the CCW heat exchangers (Problem Report 85-302).
Several actions were taken-to' correct-these problems and one action in the process involved chemical cleaning.
Requisitions 6R-282724 and 6R-340403 were processed and respective Purchase Orders 661-74038 and 661-74340 were issued to Haliburton Industrial Services Division (HISD).
i m
-
.
a f : n
.7
- ..
- L 11'
<
,
' Procurement The NRC inspector. identified the following deficiencies with the. code V procurement for chemical cleaning the CCW heat exchangers:-
(1) contracts?did not reference ~or discuss the fact that vendors would be required to comply <
with TU1 Electric's QA program which implements 10 CFR,.
Appendix B, QA Program, and 10 CFR Part:21, defect reporting requirements; (2) the requisitions and contracts-did not address the training of vendor personnel (who must'
,
help: implement TU Electric's QA programs'since the vendor.
.has no Appendix B QA and 10 CFR Part 21~ program);
(3) technical and QA requirements were not explicitly defined;'(4)-activities concerning ASME components were not done under-the auspices of ASME XI; '(5) verification and inspection plans lacked specific. requirements as they generally stated:
"QA shall monitor the vendor's work",.
and (6): vendor work plans and procedures were inadequate.
-Project Plan and Procedures j
The NRC. inspector reviewed the plan and procedure that were:used-for chemical cleaning and found that the same planz and: procedure were used for both purchase orders referenced above for work performed in February-and May
'1987.
. Project' Plan'for Chemical' Cleaning CCW HX CP1, CPZ-CCAHHX-01 and -02, Revision 0, did not adequately describe the.QA and technical-requirements needed to l
control the: process.
The plan failed to:
!
.!
Describe the purpose of fiberoptic inspections,.
ll
.
l Describe the criteria for' determining when the metal
.
.,
surface was clean and corrosion products were removed, and
. Describe the criteEia for and the hydrolazing/ flex
.
lining operation.
~After the chemical cleaning was completed in February 1987 i
such criteria were discussed in TU Electric Maintenance i
Engineering Evaluation (MEE) No.88-003. dated January'13, fj 1988, but were not factored into the plan before the-
,
second. job.
The evaluation stated that the comparison of J
fiberoptic videos of tubes before and after flex lancing
!
would be compared with a second video recorded after
consistently low copper-concentration was reached during
chemical cleaning.
It further stated,
"A comparison of the before and after videos would be the final determination of adequate tube cleanliness."
These
'
actions were not accomplished.
i
!
!
!
J I
..
-
-
-
--
.
. ".
m,, ;w
_
'
,
.
.
-
a--
.
-I
-. =
s,
1<
_12 -
.
.. ;
,
a
-
,
"ProcedureLfor Chemically Cleaning Component Cooling Water
"^
- Heat Exchangers," Revision 2,1was.inadequato in the
,
-following; respects.
i
,
The chemical c1 caning-(vendor) procedure did not j
.
contain information such as'a reference ~section,
'
purpose, scope, responsibility, definition,
'
instruction, or records.
(See TU Electric Startup
.)
,'
procedures for cicaning the diesel fuel / lube. oil pipingffor an example of a good procedure.)
{
"
$
l
' The vendor-procedure does not describe or reference
.
the ASTM standard which governed chemical testing: to
-!
e assure l proper chemical concentration.
.
The-vendor procedure does not describe how Ehe
.
N-blended solution was to.bc mixed.
t
- The vendor. procedure does not describe tho mixing of
<
y
.
nitrogen-gas with the foam solution.
Also, there was
..,
no. description of how much-heat should be added at Step 2 prior.tx) adding the nitrogen.(which is
,
'
Lreferenced.in Note 1).
s Stop 6 of the vendor proceduro did not describe where
.
samples were to-bc taken nor how to.cnsure a representative sample.
Step 8 rcquires an inspection of the CCWHX tubes tx)
.
.
determine the. degree'of scale removal, but does not j
specify the method or any specific criteria.
-
i Stop.9 should read:
" repeat' stops 5-9" instead of
'
.-
steps 7-9.-
-Step 10 states, in part, "that once inspection
.
.
.
reveals the desired degree of scale removal." but gives no description of the desired surface condition
or critoria for inspecting.
This stop did not incorporate the criteria described in Maintenance
. Engineering Evaluation (MEE)88-003.
Step 11 does not specify the quality of the water.
.
Step 13 does not describe specific mixing
.
instructions for the soda ash and sodium
'
tolyotriazole (500 ppm).
L The procedure did not describe the Haliburton
.
L operator's log nor any coquirement to record various data such as type of operation, time, temperature, I
chemical ~ concentration, and pressure.
There were no j
..
,[
-
. - -
N 87't.
-.
-
-
"
4-
.,-
'
.
,.,
j L
'
Haliburton! signatures on the data: forms.to
.
,
authenticate the data.
Only the TU' Electric Project j
~
. Manager signed Haliburton's log.
However,.the-
project manager did'not perform the steps-or-
.
i operations and was not always present to verify each aspect of the operation.
'
The vendor procedure (Attachment i to Work
,
.
,
Order CP7-2347) had no TU Electric' approval on it.
l t
'
The vendor procedure.did'not address the calibration
.
of the gauges and'other measuring devices used for process control.
Pressure and temperature were at
'least'two parameters which should have requiredL measuring cquipment and calibration.
TU Electric
,
,
memorandum'TCP-87027 described this deficiency after
'the-first' job, but'no nonconformance report or
.
. corrective action request was evident.
Note:
The project managor's log for the second job indicated that TU Electric took measurements, but
'there is no record of those measurements.
It was,
also indicated that the project manager was. issued
1 calibrated measuring and test devices that did not
<
work. -In discussions with the project manager, it-was determined.that the vendor's equipment had gauges that were not under an approved; calibration program (Appendix B' requirement).
Finally,;the tegperature as geasured by the project manager was 117 F versus-
,
122 F as-measured by site chemistry.
No' deficiency report was. issued to document and'cvaluato this C
deficiency.
E The procedure did not' address acid spills.
.
L L
The' procedure did not address passivation after-
.
cleaning.
V
,
Support' Procedures - The NRC inspector found that a number of other work activitics-were required to support the chemical cleaning process.
Specifically, three work
,
activities were~an integral part of the cleaning process:.
(1)= fiberoptic examination,-(2) ficx lancing or hydrolazing, and (3) eddy current testing.
Procedures to contro1~these activities were not r'eferenced in the
.,
'
chemical cleaning. procedure.
Work order C870000585 contained a revision to require Hydro Nuclear Company to flexlance the tube side of the heat exchanger, not to exceed 10,000 lbs pressure.
Since l
no procedure was found, it is not clear if quality was
,
sufficiently involved to verify and document this work.
.
..
'
xa <,..
>
-
,
wM6
j t
d
..
-t n
,*.
j
<
.j
,
,
'b i
>
.A
'
Other TU Electric activities-concerned the; fiberoptic examination and hydrolazing performed by Hydro Nuclear.-
',
It is unclear if the flexlancing.and hydrolazing'was the
'
same operation.. Finally, eddy current testing was
- >
,
performed,'but was not discussed in the Project Plan or Ecleaning procedures.
Documented evidence of controls for
these activities.were not provided to the NRC.
s
.
i control of Work Activities - The subject purchase orders
~
resulted in.the chemical cleaning of Units 1 and.2 CCW heat exchangers.
The first work occurred in February 1987
under Purchase order 661 74340.
During.the first cleaning job in February'1987, the vendor experienced a number of-problems'as described in TU Electric office Memorandum
~
TCP-87027:
nonuniform distribution of chemical. cleaner
(which prolonged'the cleaning process), flow rate
'
. considerably below estimated flow rate of 24 gpm, quality
'
. of chemical foam inconsistent; gas flow meter was not
'
calibrated for expected flow rate; long interruptions
occurred while foaming; nozzles were not the correct type.
for most effective cleaning; nozzles plugged up several
,
times; defoamer equipment was inadequate to deliver the.
chemicals and'eaused interruptions; anduthe vendor had
'
+
insufficient manpower for the task causing TU Electric to
'
supplement the vendor's work force.
The memorandum concluded by recommending a penalty lfor poor performance.
'.
This memorandum appears to be in contrast to TU Electric
'
'.
Contractor Surveillance Report CSR-87-002'which concludes
'
that contractor performance was satisfactory (except when i
~
the vendor removed a: red danger tag without l
authorization).
Although the chemical cleaning 1 job for. Units 1 and 2
'
CCWHXS, Train B, (performed in May 1987) was better than Units 1 andL2 CCWHXs, Train A, the NRC inspector determined that no deficiency /nonconformance report or corrective action request was generated to' identify, evaluate, disposition, and correct the_following
~
' deficiencies.
Also, the causes of,these deficiencies were not identified.
The process problems discussed in Memorandum
.
TCP-87027 and surveillance summary 87-022 were not documented in deficiency reports and evaluated to determine if the requirements were adequate.
The QA
.
'
and technical requirements were identical for both
,
requisitions and purchase orders.
Considering the
'
problems discussed in the memo and summary 87-022 chemical cleaning of CCWHXs, it should have been evident that the requirements were either inadequate or the vendor was not meeting the requirements.
'
,
'Y a
p
o
'
O g:9 L.'
y
'
...
.
i
j The chemical cleaning process deficiencies documented-
!
,
in Momo TCP-87027 wore-nonuniform distribution _of
f
'
chemical'cleancr, inadequately'mcasured flow rate,-
t chemical mixing' inconsistency;-process interruptions,-
inadequate equipment,,and; inadequate. manpower,-were not documented as deficiencios-and formally evaluated
'
.
to assure correction before the~ award of the second
"
contract for chemically. cleaning CCWHXs for Train B
!'
and before. work was completed on the second chemical cleaning job.' TU Electric memorandum TIM-870301
estimated a loss of*0.01-mils of metal surface.except; for areas where active pits.wcre and'the loss'there was estimated to be 0.2 mils of. metal.'
Since thc.CCW
'
heat exchanger is an ASME, Class 3 compon'ent,tthe deficiencies in memorandum TCP-87027 should have been formally documented, evaluated, and dispositioned to-a assure the process did not result =in excessive metal attack, especially in active pits.
,
'
After the chemical cleaning was completod-(por the
.
procedure)', two hours worth of chemicals were left
.
-
'over.
Rather than wasto those chemicals, one hour of
.,
additional-cleaning was added to"each heat exchanger.
'
This action was taken without~ obtaining authorization to change the process procedure.
NOTE:
On May 1, 1989, TU Electric stated that the
'
process;was not-continued on the basis of chemicals left over,.but acknowledged the log stated that.
The
>
NRC inspector is of the opinion that additional chemical.use should have been based on inspection
_1 criteria to determine if the surface was cleaned.
+
A projects summaryL(Theimer 6-18-87) listed ton
.
comments / recommendations based on the second chemical
cleaning job in May 1987.
.These comments are further
"
indication that the chemical c1 caning and support i
procedurcs woro.not well developed to achieve an integrated approach which would assure the work was properly controlled.
The main comments discussed deficiencies in these areas:
(1) organizational interfaces, (2) acceptance criteria to avoid unnecessary attack, (3) chemical and point indication, (4) samplo not taken from main tank L'
supply, (5) PH sampling locations, (6) passivation, L
and (7) timely chemical analysis.
The objective and acceptance criteria were not described in the vendor procedure, but this memo stated, "The acceptance y
criteria [ sic) is a visibly clean heat exchanger tube L
surface without unnecessary base metal attack."
This critoria should have been established in February 1987.
One important comment on a support activity
.
l
<a
.
%%.;
,
r 5,,,
34;
,
concerned TU Electrib analyses ofichemicals and de corrosion product (for copper) versus the vendor's-
. analyses.
The comment suggested a time lag had occurred-between the vendor's analyses and TU, Electric's1 chemical analyses.
Also, fit recommended agreement of 1 20% between values.- This suggested a large difference had occurred and after-7-
the fact-correlation'was made.
Since these analyses control-the rate of attack and along with visual inspection, indicate the process.end point, this should=have:been documented as a deficiency.
s OA' Surveillance andcInspection - The NRC inspector-
-
reviewed QA Surveillance Reports CSR-87-002 dated March 2,
'
1987, and backup files for the first chemical cleaning of the CCWHXs. -The checklist for CSR-87-002 included 11 attributes, 4 of which were marked not applicable.
The NRC inspector determined that:
i
.-
Item 1 checklist characteristic was marked.
- satisfactory and required = verification.of contractor prepared procedures reviewed and approved-by appropriate "TUGCO" personnel prior to use..No signatures for review and approval were on.the procedure.
Rather, the surveillance report stated that approval was accomplished by attaching the chemical cleaning procedure to the Plant' Operation organization's work order.
-Item'2 checklist characteristic was marked
.
satisfactory and-'it'requiredLthe verification that contractor prepared procedures for.special process were qualified iri accordance' with industry standards while Item.5._ addressed contractorJpersonnel performing special processes.
This characteristic and finding forJItem 2'is contradicted by Item 5.
That is, Item'5 was marked not: applicable.
As both
"
address special processes, they are either both Eapplicable or not applicable'.
Item 3, checklist: characteristic was marked
.
satisfactory and required verification that
contractor personnel performed'in accordance with
,
procedures.
This finding-does not reflect and is in opposition to' process deficiencies that were identified in TU Electric Memorandum TCP-87027.
Since the' procedure was not properly reviewed, approved, and contained shortcomings, the finding for this characteristic was of questionable value.
Item 4 checklist characteristic was marked
.
satisfactory.
The comment indicated the vendor
$
,
-
,
- _ _ _ _ _. _.. _ _.. _..
.
__
m.
y s
$ --
_y. Y:l e
'
'
l
'
?
-
1.-
.
h
.
,
completed-documentation as specified in tho' purchase-
order.
This' finding;is contradictory as no such
documentation rcquirement was in the purchase order.
(Sec TU Electric Memo NE-22156).
,
Item 7 checklist characteristic was marked not-
.
';
applicabic..
The checklist characteristic required
'
TU Electric to-verify.that safety-related material
supplied met CPSES requirements and material certifications. :The vendor furnished chemicals which l
should have been checked or verified when received.-
'
.
.
Tho surveillance could have verified that appropriate
chemical grado materials were received before uso.
- TU Electric Procedure EC 6.11 requires engineering',
,
construction, and QA to certify that all' contractor.
g?
supplied ~ material, and/or special tools be' received
by the TU Electric QA warehouse and accepted by QA.
There was no reference to the " Contractor Work i
Release Authorization Form" which is required by lEC 6.11.-
'
'
Item.9 checklist characteristic was marked not-
.
' applicabic.
Item-9 required the verification of contractor supplied'mcasuring and test equipment.
The comment on'this item stated that,
... our
"
chemical dept.-provided cal. cquip...
"
This
.
...
statement shows that cquipment. furnished by TU Electric should have been verified as a part of
'the surveillance because TU, Electric assumed all QA'
responsibility.
In addition, the project manager
'
_
indicated that vendor furnished equipment hadl gauges which helped control the'proccss.
The surveillance
.should have addressed the calibration or lack
thereof.
j Item 10 checklist characteristic was marked
.
satisfactory.
Item 10 stated:
" Chemistry provide-periodic oversight of process & take samplos to test
'
for Fe & citric acid concentration."
This was
- followed by a comment " incorrect requirement.
y Checked for copper & nickel."
The procedure
"
misstated which test should have been performed.
The
o satisfactory finding was contradicted by the negative finding.
t Item 11 checklist characteristic was marked
.
satisfactory.
In this caso an attribute was added to y
verify that passivation was done after cleaning and S
before the domineralized water flush.
As this chemical passivation operation was not in the
'
proceduro,'the source of the characteristic is not clear.
If no procedure was established, this i
y
^
-
- - - - - -
-
- -
.
.
- - -
.
- -
- -. -
. -
w%
.
-
- 3
.
,
.!
j 2
..
>
ii
.
.
-
-
,
,
,
'
a n
L
}
^*
p
-
- 18 m
,
a g
. characteristic should have been marked unsatisfactory'
.and a deficiency written because it was not an
-
t'
- approved step in the procedure.
'
The NRCLinspector also reviewed Surveillance Activities
. Summary (SAS)87-022 which was-referenced by CRS-87-002.
The NRC inspector did not find the~Surveillanco-Activity
,
Summary in the QAirocords.
It was~ furnished in a personal t
file and was not signed by the Quality Surveillance ~
.
4, supervisor.. This summary only addressed chemical testing-and-appeared,to contradict the surveillance report.
The summary indicated that the overall: chemical cleaning f
process was not appropriately controlled as analyses of-
- copper concentrations' indicated an' unstable condition, m
verbal agreements allowed acceptance because of cost-e considerations,' discrepancies between times (that chemical-foamswas-stopped) were recorded by project manager and
.Haliburton data' sheets, no. chemical analysis during approximately two hours of continued cleaning, samples Lwere not taken and analyzed, pH values for ammoniated solutions'were not. adjusted for temperature, sample location 7was improper, and largo differences-between
,TU ElectricLand the vendor's chemical analyses.results.
This. surveillance. summary concluded that only two of five
,
findings were deficiencies and reports were written.
The
,
NRC believes the three remaining findings should have been documented as deficiencies.
Finally, surveillance summary 87-022' stated that the chemical cleaning' process is
'
I defined as a special process in paragraph 5.2.18 of-1: -
' ANSI N18.7..while CRS-87-002 stated it was not a special-Lprocess.
The' summary of 87-022_ stated that inconsistent u
cin-process. controls coupled with " cost-effective"~
l decisions in Train.B cleaning activities may.have.a
-
L detrimental effect on the heat: exchangers, at a later time.
W There is no evidence that the potentially detrimental
'
offect' discussed in this report was ever formally y
addressed in a deficiency report.
The NRC inspector reviewed Surveillance Report CSR-87-005 dated.May 13 ~, - 19 8 7, which covered. activities on the second chemical cleaning operation'by Haliburton.
The quality of this surveillance was about the same as CSR-87-002.
L'
Based on the available documentation that was' reviewed, the NRC inspector believes the QA surveillances were not completely adequate.
In addition, all deficiencies identified in surveillance activity sammary SAS87-022 were not documented in a deficiency report to assure L
cvaluation, disposition, and corrective action.
.
The NRC inspector reviewed the records and files, but a.
found no inspection reports for chemical cleaning.
Work
,
'
m
.
,
gyc,
,
.
s
,
,
_
~
,jv (
~
, 3, s
g<
'
.
orde'r C870000585 in'dicated.thatLQC would be involved, but no; inspection report-was required.
The work' order statedL j
>
that QC shall provide personnel and equipment to perform
'
fiberoptic examination, but no inspection report was
)
'
required,
<>
g
'
b.
Cutting Heat Exchanger Tube Ends
'A Code V procurement'for this service resulted in issuing-Requisition R-49642 dated August 18,.1986, and Purchase Order CPF-13593-5.
The purchase order was issued to-Perflex Services.
This work was'a prerequisite to
!
recoating of_the heat exchanger.
The work' involved cutting ~5720 tube. ends to lie flush with the tubesheet, grinding-rough surfaces, preparing ends for coating and-removing-brass plugs.
Such_ preparation was necessary to obtain a quality coating to protect the surface from
' corrosion.
Procurement - The NRC inspector reviewed the procurement files and found that the Code V procurement deficiencies described in: TU Electric' Memorandum NE 22156 generally-
..*
applied to this procurement.
Project Plan and Procedures - The NRC inspector determined that the work activity,(cutting heat exchanger tube _ ends)
was-not;part of a project plan such as STA-TP-87-3 which
- n described a plan for cleaning the CCW heat exchangcr.
No individual project plan was-found.
~
Perflex Services Procedure CPF 13593-S, Revision 2, dated-September'4, 1986,.was approved by-Stone and Webster Engineering Corporation who was project manager for this
'
job.-
The procedure submitted by Perflex was a one page
>
procedure which did not:
Describe how the vendor's personnel would interface
.
with various organizations such as SWEC, Brown and Root, Inc., TU Electric Construction, and Operations.:
Describe the inspection to be performed by Perflex
..
Services personnel and or the personnel ~
,
qualifications.
F
'
Describe criteria for rough grinding tubes after-
.
being cut or.specify a surface finish or generally
.
'
state that burrs, rough edges, and other defects be removed.
Describe the steps to meet DCA 25192, Revision 0,
.
which required that sharp outsido corners be 1/8-inch radius (minimum)Jand inside corners be welded (ASME'
rr
,
.
-
-
- - - - - -
,
,
p y %.:,
'
t
.
' a ts
>I l
'
'
,
,
,
~'
20-i
.
.
..,
.
.
R
.
DivisioniND)cto' build up to this' radius.~
section'III,
'
.Revisioni4 oflthe DCA specified the_ radius l
'
"
' requirement, but this should have been addressed in the subject: procedure'unless the cuttingjcaused'no sharp-corners.
NOTE:
-Interviews with the project manager did not.
~
,
clear up this matter.and no answer was-provided specificnto whether welding' occurred or.not.
..
' '
'However, the QC inspector stated that welding did-not occur.
- Inspection of Work Activities
"
,
' Inspection of the tube cuts and removal was documented in TU Electric Inspection Report 86-0289.
However, only 25_
CCWRX tubes were: inspected for one'CCWHX.
The purchase
,
order stated that 5720 cuts would be'made on-the~CCWHX,
but Form:TNE-PR-3.~2 indicated two1CCWHXs.
It appears that T
,
thc balance of the tubes, about 5670", were not inspected
'
z or_were inspected-by Perflex Services.(who had no QA/QC program responsibilities in the contract).- Such'
-inspections should have been made by inspectors certified
,
e to ANSI N45.2.6.
It appears that the inspection'
"
L characteristic of 0.030 inches maximum protrusion'was not verified by direct measurement with a goLor no-go gauge.
E The-vendor's procedure did not indicate 1how in-process L
work was= monitored and no in-process inspection procedure:
D was evident.- No! documentation was?provided by the applicant to: verify the inspection'for minimum radius of_
- 1/8-inch por: DCA 25192, = Revision 4)
OA Surveillance - The NRC inspector reviewed
,
Surveillance CSR-86-004.
The checklist'was the-same as-a r
others reviewed and it appears to'bc a:gencric checklist.
'
Similar to previous surveillances, a large number (half).
of the characteristics were marked not applicable.
The-summary of this surveillance was not complimentary to the vendor regarding the lack of discipline, tools, and
!
cxperience.
h c.-
Application of Epoxy Coating to CCW Heat Exchanger
"
i
- Requisition 48370 dated July 10, 1986, and Purchase
'
.
Order CPF-13597-S dated August 26, 1986, were issued to
"
,
?
Specialtics Engineering Corporation (SPECO).
,
Procurement - The NRC inspector determined that the L.
general comments in TU Electric Memorandum NE-22156 applied to this procurement.
l-Y
,:2 i
,
, - - -. - -... _. - --. - - - - -.. - - _.. -
,
_ _
_
-
^ ~ '
'
' p[,;
'
t
,,
S s
~
a:
215 -
l Project Plan and Procedures -'The NRC inspector reviewed.
q the plan and procedures.
These were more detailed'and;
'
technically comprehensive than other. vendor' plans and-j procedures.. Work procedures (Attachment A, B, C, D, and E
]
.to'SPECO letter FR-48370) described surface preparation, a
coating of; channels, heads, tube sheets, and tube ends.
l However, theso procedures.were not dated and no-signatures
for~ review and approval were,on the procedures.
j one: procedure required the applicator.to visually-inspect
the coated area'where spark. testing.was not possible, a
~
' practice that-is generally not acceptable because an-individual should not final inspect his own work.
There is no indication that the TU Electric inspector inspected
+
-
areas where a spark test was not possible.
Also,
!
Specialtics Engineering Bulletins dated December 14, 1978, i
for repairs were not described in the procedure and were not in the onsite records.
Inspection-and Test - The NRC inspector reviewed'
!
TU Electric Inspection Report 86-0289 and determined that:
..
The inspection of the coating only addressed the inspection of the final dry film thickness.
Such-a-
-final. inspection would not assure that the epoxy was.
.
applied as required by Attachment.B, " Coating
'
-Application Procedures for Channels Heads."
Attachment B procedure required'three coat b
applications and thickness was supposed to be i
. controlled during-each application.
l:
After after the figal (third) coat it was to be cured a
.
L'
for lg hours at-70 F ambient temperature'or 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> i'
at 60 F ambient.
Inspections of.these
'
'
characteristics, if performed, were not documented on:
'
the. inspection report.
No characteristic was included in the inspection
.
.
report for repairs for SPECO Bulletin 35.
'
The procedure (Attachment B) required-measurements
.
using a Bacharach Sling psycrometer and a Pacific i
'
Transducer Company surface thermometer.
No-i TU Electric inspection showed that the vendor's
'
equipment was calibrated.
The inspection report had:no characteristics to
.
require inspection of the surface preparation or procedurcs (Attachments A and E).
,
QA Surveillance - The NRC inspector reviewed Contractor Surveillance Report SR-86-007 and-the attached checklist.
.
I
.. - - -
-. -
. - - - -
- - - - -
.
. -
.
.
,
.
,
-
?) win %;i '
y
. ;
a'
^
a:
I J
'
t s
22'
(
'
,
,,
g
,
.
L The. checklist: contained 12: characteristics to be verified.
The generic checklist:had-been modified to add characteristics to verify surface-preparation,. spark test,
~
'and 6 inches.of tubes coated.. All three:were marked
,
satisfactory.
The basis of the~ satisfactory _was a-i Troference~to' Inspection Report IR-86-0289~and the:
' inspection of surface preparation.
However, the inspection of the surface was not in inspection report-IR-86-0289.
.Throc characteristics on the checklist, including calibration, were marked not applicable.
This decision
~
-appears questionable considering vendor-personnel were inspecting'with equipment that may or may not have been in
,
the site calibration program.- The surveillance was insufficient to. fill _the inspection gaps described in the paragraph above.
,
6.
Repair of Diesel Generators Heat:Exchangers (50073, 50075)
U The NRC inspector learned that diesel generator _ jacket water;
,
-heat exchangers were examined.
Corrosion was found and Design Chsnge Authorization (DCA) 21981, Revision 6, required repair,
'
corrosion removal, and recoating.
This' involved removing the existingDrubber liner, inspecting surfaces to_be coated with Belzona Ceramic S-metal, welding.to-build up corroded areas.
.'
(ASME III work), sandblasting in preparation for coating
'
application, and coating application.
,
ihe NRC inspector evaluated selected areas where the above-work
- was done.. The quality of the procurement, inspection, QA surveillances, Land corrective action concerning Requisition'6R-345080 and Purchase Order CPF-14220-S to Haliburton for the above work were similar to SWS and CX39 work
activities.
The NRC' determined that similar deficiencies existed with respect to the QA program implementation as described above'in paragraph 5 above.
one exception was the procedures developed by TU Electric ~Startup.
They were a good example'of how other procedures should have been developed and implemented to assure proper controls for work activities.-
y
.7..
Vendor Services to Measure Steam Generator Nozzles (50073, 50075)
L procurement - The NRC inspector reviewed requisition 6R-356251
'
dated July 15, 1988.
The requisition does not make clear
L whether-this was a safety or nonsafety-related activity.
.
L Including the SWS requisition, three of seven Code V E
requisitions cVidenced such confusion.
Had TU Electric QA adequately audited the Code V procurements, this trend may have H
been identified and the problems associated with these procurements could have been identified and corrected.
There
. <
> <
_
--
.,
_ _ _ _. _ _ _ _ _ _ _ _ _, _ _ _ _ _
rm
,
.bl? Jo;.-
,os
,
,
-4
s
-
.is no indication that adequate audits of the Code V E
procurements were ever performed.
A' purchase order (661-74054) dated January 16, 1987, was. issued
]
'
to Nuclear-Services,-Inc.
Sixteen nozzlos on eight steam generators were to be measured and visually inspected to
,
determine each nozzle diameter and radius, height'of flange
' ring to nozzle-and location of the 3/4 - 10 UNC tapped holes.
This was necessary in order for TU Electric to procure > nozzle dams to be: used inside steam generators to temporarily isolate i
the steam generator primary channel head from the refueling pool and permit refueling and testing or repair ~of the steam
,
-generators to occur simultaneously.
'
In July 1988, requisition 356251 was issued to purchase the
!
nozzle dams from: Nuclear Energy Services.
The requisition was marked Code N which meant that no 10 CFR 50, Appendix S, QA program or 10 CFR 21 requirements were applied.
Codo N was e
incorrect because the activity was safety related.
Inspection - The work on stcam generators was in progress before QA was aware that Nuclear Energy Services was on site.
.
Since this was a Code V procurement for services, TU Electric H
was required to provide the QA program and assume 10 CFR Part'21 responsibility for the vendor.
As previously discussed,-QA was required to certify that material and tools
were roccived and personnel were trained prior to work j
(rcquired by EC 6.11).
This was not dono.
By chance, the QA l
organization discovered the work was in-progress and decided to i
verify: access control, survoillance CSR-87-003.
No checklist
,
was attached-to the surveillanco.
TU Electric wrote a i
deficiency report'(P87-0135) because the work was completed
January 14, 1987, but the contract.was not completed and. dated until January 16, 1987, and maintenanco engineering and QA did
!
not roccivo:it until January 19, 1987.
The surveillance
,"
concluded that QA did not know about special requirements until afterEthe fact.- The NRC inspector found that TU Electric Memorandum NE-22156 concluded that this was " acceptable"-
+
because hardware was not changed.
The basis for this conclusion is not evident.-
,
!
'
The_NRC inspector determined that-the procedure for measuring and inspecting the nozzles was comprehensivo; however, the
a
. procedure was not reviewed and approved to incorporato it into l
'
the TU Electric document control system and bring it under i
H their QA program.
Section 3 of Attachment 1, " Steam Generator L
' Nozzle Measurement Procedure," stated that Nuclear Energy
.
-Services would furnish profile gauges and thread gauges to I
verify-location and' condition. -The procedure did not state L
that this equipment would be under the TU Electric calibration program.
There was no QC verification of the calibration of i
this equipment.
'
L
,
L
,;
i.
?
?
10, M.-
,
'
a EU,
,
>
- The NRC inspector determined that 'on December 1987 design ui modifi. cation' request (87-1-237C) was issued to drill and tap cight additional holes in each nozzle to accommodate the nozzle dams when'needed.
Measurements and inspections to implement
.'
'those proposed modifications and inspections were safety
,
related.
'8.
' Application / Removal of Coatings from Diesel Generator Tanks
,
-
"
(51053)
NRC inspectors mot with TU Electric in September 1988 and I
-
pointed out the similarity between deficiencies in the diesel
= generator fuel oil tank coating removal and service water
system coating removal.
TU Electric did not consider them to be similar.bocause the procurement code was different.
The NRC inspector believes-the same lack of QA/QC controls existed, as
,
'
documented below:
a.
Initial coating requirements for the dicsol fuel oil storage tanks were defined in the tank specification
'(2323-MS67A); this document did not require inspection or
documentation of the coating. process.
In May 1979, a design change authorization (DCA 4665).was issued implementing,the provisions of Specification 2323-AS-31 which included requirements for safety-related procedurcs, inspection, and documentation for-protective coating work.
In January 1983, the project recognized that the required
'
documentation was lost and an NCR (C-83-00223) was generated.
It was dispositioned "use-as-is" on the basis that coatings of the tanks were not critical since coating
'
failure could be offset by alternate means=of filling the-day' tanks.- In August of 1983, blistering of the coating
was noted in one of the tanks and an NCR-(C-83-021615) was written and dispositioned "use-as-is" on the basis of
-
insufficient blistering to warrant repair.
>
In mid-1985, the safety-related coatings Specification (2323-AS-31) was reclassified to "Non-Safety Related".
In 1986, during the cleaning of the Unit 2 dicscl fuel oil storage tanks in preparation for startup testing, a band of rust spots approximately two (2) feet in width was-observed in the Train A tank.
DCA 4665 classified the coatings work.as a safety-related activity, but the declassification of the Specification (2323-AS-31) removed the technical basis for implementing a' repair of the
.
safety-related coatings.
This distinction between safety versus non-safety for coatings and similar activitics requires resolution by TU Electric prior to initiating repairs on the Unit 2 tank and investigation of the Unit 1 tank coatings.
.TU Electric letter (TXX-6461) concluded
-
-
y
--
,.c
- - - -
f.afg y
2v
'
-
,
.
I
[# i
<
_f
- --
- =
..
.
'
.
'
-i
.
"
that the' coating was not safety-related and the walls were
. thick enoughito withstand corrosion for.the 40 year, design life.
b. :
Th'e NRC inspector reviewed the background of the issues
<
,
'
and records for the above activities and determined the following:
Similar to the service water coating, Gibbs and: Hill
.
failed:to recognize that'the procurement / application.
of coating is safety-related even though the coating I
may be nonsafety related.
The coating subsequently-
'
failed and the tanks were attacked ~by corrosion.
The initial corrective action-(DCA-4665 dated 1979)
.
attempted to correct the QA program deficiency by changing the specification (2323-AS-31) to require such= controls.
However, this action was reversed in mid-1985 by reclassification to nonsafety-related.
,
'This reversal was incorrect because it did not
>
recognize the adverse effects the uncontrolled work
,
activity could have on the safety-related fuel oil
-
tanks..
,
-Onl September 4, 1986,.TU Electric reported in a
.
10 CFR 50.55(e) report..that the fuel oil tanks-that were coated without.QA/QC controls were. acceptable and the first reclassification to safety-related was
' incorrect.
Therefore, no-QA/QC controls were needed.
It was also' concluded that since it was unlikely that fuel _ lines would,become' clogged with-coating (if.it failed), this item was not reportable.. The final response (TXX-6461). dated May 22,11987, failed tct
,
L'
assure corrective action as follows:
(1)
Failed to consider the. fact tha't activities-H affecting the quality of components must be
!
controlled even though the purpose of the coating is considered nonsafety related..
(2)
Failed to address the fact that the coating I
material was not known for sure, but assumed it was AMERCOAT 395.
J (3)
Failed to address the loss of the documentation
,
of the type of coating and how it was applied.
L TU Electric failed to address the similarity between
.
diesel generaEor and serv.i-e water coating damage in the Enforcement Conference Document.
Both involved L
L the lack of QA/QC controls for coating procurement application, coating degradation, and corrosion of l
5
- - - - - - - -
-U
-
a
- -, - -,. - - -_ _ - - _ - _, - - - - - - - - ----
- - - - - - -, - - - _ - - - - - - -
-
-r
>
bI
.. o
-
'
<
g -.'
.
,
,
.
lY
,*
.
.
'
. v.
,,
,
n
,
- U components.
The NRC had ~poliited out the similarity
]
'
beforetthe Enforcement Conference, i
J The. removal of the coating:fromfASME tanks should
.-
'
have come under ASME XI,for Unit l' tanks. -There was no'-indication-that'ASME XI-was considered.
,
,,
c.
-The NRC inspector found that TU Electri~c failed toLtako adequate' correctivo action as 'follows:
l
'
'
(1)- The specification does not specifically address--the
.
,
controls of activitics affecting the quality of the J
fuel oil tanks.
(2)
TU Electric-did not address why the documentation was o.
'
lost.
(3)
TU Electric did not specifically address the lack of ASME XI involvement..
JThe above workLon the diesel fuel oil tanks was not a service procured under Code V, but the work was performed on sitc by a contractor.
However, the similarity existed
~
'
between work on the service water system and work--
'
previously done on the component cooling water, diesel-l
'
generators, and steam generators, that is, the questioni
'
about'whether thogprocurement and application of coating
-
wereisafety-related.
Other similarities.were that work was:not donc under.ASME XI auspices, and documentation was not readily' retrievable.~
~
.
9.-
Exit Meeting (30703)
-
An1cxit; meeting was conducted May 2, 1989, with:tho applicant's Lrepresentativos 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 as proprietary any of the materials providod'to or reviewed by the inspectors during this inspection.
During this
-
meeting, the NRC inspectors summarized the scope and findings.
~
of the inspection.
,
i'
t L
L t
-~
.
.
.
.
.
h.;k< d(#w
-
J^
e
..,
6 Addendum-to. Inspection Report 50-445/89-23; 50-446/89-23 Paragraph :2.a, b, and c Open Items 445/8908-0-01, 445/8908-0-03, and 445/8908-0-04 were closed in IR: 50-445/89-84.
~ Paragraph 2.d
Unresolved item 445/8908-U-02 was determined to be a violation =
'(445/8935-V-01) and was subsequently closed in IR 50-445/89-85.
Paragraph 3.c
.
Violation 445/8847-V-Olc was closed in IR 50-445/89-76.
Paragraph'4.b.(3)
One of the three NRC inspectors who watched the video tapes of Train B after coating removal (not the author of IR 50-445/89-23) wrote:
1"The_ observed spinblaster marks were superficial in nature and appeared to be the result of only a brief holdup in the i
spinblaster's rotation rate.
The loss of wall thickness
<
associated with the marks appeared to be on the order of a few mils, far less than necessary to create a pipe wall thickness concern.
On this basis, the' applicant may have determined that the marks did not constitute ' damage'."
,
,
i
..
l
'
- - -. -
- - -
- - -
- - - - - -
- - -. -