ML20195F910
| ML20195F910 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 11/18/1988 |
| From: | Livermore H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20195F904 | List: |
| References | |
| 50-445-88-71, 50-446-88-67, NUDOCS 8811220409 | |
| Download: ML20195F910 (18) | |
See also: IR 05000445/1988071
Text
_m-
-
_
_
.
.
7PPENDIX E
U. S. NUCLEAR REGULATORY COMMISSION
OFFICE OF SPECIAL PROJECTS
NRC Inspection Report:
50-445/88-71
Permits: CPPR- 126
50-446/88-67
CPPR-127
Dockets: 50-445
Category: A2
50-446
Construction Permit
Expiration Dates:
Unit 1: Extension request
submitted.
Unit 2: Extension request
submitted.
Applicant:
TU Electric
!
EXyway Tower
400 North Olive Street
Lock Box 81
Dallas, Texas
75201
Facility Name:
Comanche Peak Steam Electric Station (CPSES),
Units 1 & 2
Inspection At:
Comancho Peak Sito, Glen Rose, Texas
Inspection conducted:
October 5 through November 1,
1988
Inspection conducted by NRC cor.sultants:
J.
Dalo - EG&G (paragraphs 2.a and 3.a)
K. Graham - Paramotor (paragraph 5.c)
P.
Stanish - Paramotor (paragraph 2.b - 2.d,
3.b,
4.a - 4.0,
5.a, and 5.b)
Reviewed by:
R F ld M L
fw
N-a-M
H. H.
Livormore, Lead Senior Inspector
Dato
8311220409 881118
PDH
ADOCK 05000445
O
_ -.
x-----__
_ _ - -
- - - - _ _ _ _ _ _ - .
.
'
2
.
Inspection Summary:
Inspection Conducted:
October 5 through November 1,
1988 (Repgrt
50-445/88-71; 50-446/88-67)
Areas Inspected: Unanncanced, resident safety inspection of
applicant actions on previous inspection findings, follow-up on
violations / deviations, action on 10 CFR 50.55(c) deficiencies,
Corrective Action Program for Conduit Supports Trains A & B and
Train C >
2",
piping and pipe supports, and general plant areas
(tours).
Results:
Within the areas inspected, the NRC inspector identified
a strength in the manner in which modification to conduit supports
for Train A & D and Train C > 2" are being installed and inspected.
This is based on the number of modifications inspected by the NRC
inspector versus the small number of inspection errors detected
(paragraph 5.a).
In the PCHVP, weaknesses were identified in the
area of applying appropriate controls for the evaluation of
inaccessible / unverifiable final acceptance attributes, processing
of nonconformance reports (NCRs) related to the implementation of
the PCHVP, and corrective actions for previous NRC inspection
findings (paragraph 5.c).
Additional significant weaknesses were
identified in the Corrective Action Program for piping and pipe
supports in the areas of corrective action to audit findings and
NRC issues related to errors in design calculations, performance
and review of design calculations
as well as procedural and
training weaknesses (paragraph 5.i).
Three violations (paragraphs
5.b and 5.c) were identified documenting:
(1) failure to provide
documented instructions for performing technical dispositions for
inaccessible inspection attributes exempted from inspection by
DCA 58252, (2) failure to properly inspect thread engagement in a
spring hanger, and (3) failure to properly review a design for
suitability of application.
One unresolved item (paragraph 5.b)
related to SWEC design calculations for piping and pipe supports
was identified.
+
.
,-
,?
,
'
.
DETAILS
1.
Persons Contacted
- R.
W. Ackley, Jr., Director, CECO
- R.
P. Baker, Licensing Compliance Manager, TU Electric
- J.
L. Barker, Manager, Engineering Assurance, TU Electric
- D.
P. Barry, Manager, ESG, Stone and Wcbster Engineering
Corporation (SWEC)
- J. W. Beck,'Vice President, Nuclear Engineering, TU Electric
- M..R.
Blevins, Manager, Technical Support, TU Electric'
- H.
D.
Bruner, Senior Vice President, TU Electric
- W.
J.
Cahill, Consultant, TU Electric
- J.
T.
Conly, APE-Licensing, SWEC
- G.
G. Davis, Nuclear Operations Inspection Report Item
Coordinator, TU Electric
- R.
D. Delano, Licensing Engineer, TU Electric
- D.
E.
Deviney, Deputy Director, Quality Assurance (QA),
TU Electric
- G.
L. Edgar, Attorney, Newman and Holtzinger
- G.
E. Grabruck, QA, Impell
.
.
- W.
G. Guldemond, Executive Assistant, TU Electric
- P.
E. Halstead, Manager, Quality Control (QC), TU Electric
- T.
L. Heatherly, Licensing Compliance Engineer,
TU Electric
- C.
B. Hogg, Engineering Manager, Bechtel
- R.
T. Jenkins, Manager, Mechanical Engineering, TU Electric
- J.
J.
Ke.11ey, Manager, Plant Operations, TU Electric
- 0.
W.
Lowe, Director of Engineering, TU Electric
- F.
W. Madden, Mechanical Engineering Manager, TU Electric
- G.
M. McGrath, TS/SP Manager, Startup, TU Electric
- J.
C. Miller, Site Mana_Jr, TENERA
- J.
W. Muffett, Manager of Civil Engineering, TU Electric
- L.
D. Nace, Vice President, Engineering & Construction,
TU Electric
- E.
F. Ottney, Representative, CASE
- S.
S.
Palmer, Project Manager, TU Electric
- J.
D. Redding, Executive Assistant, TU Electric
"D.
M. Reynorson, Director of Construction, TU Electric
- M.
J. Riggs, Plant Evaluation Manager, Operations,
TU Electric
- J.
C. Smith, Plant Operations Staff, TU Electric
- P.
B. Stevens, Manager, Electrical Engineering, TU Electric
- J.
F. Streeter, Director, QA, TU Electric
- C.
L
Terry, Unit 1 Project Manager, TU Electric
- T.
G. Tyler, Director of Projects, TU Electric
- J. R. Waters, Licensing Compliance Engineer, TU Electric
The NRC inspector also interviewed other applicant employees
during this inspection period.
. - - . .
.
.
.
.
-
-
'
.
.
4
- Denotes personnel present at the November 1, 1988, exit
meeting.
2.
Applicant Action on Previous Inspection Findings (92701)
a.
(Closed) Open Items (445/8513-0-10):
During a Comanche
Peak Response Team (CPRT) reinspection, the Evaluation
Research Corporation (ERC) inspector identified the
following to the NRC inspector as subject to evaluation
as potential deviations:
(1) pipe clamp parallelism,
(2) spherical bearing gap, and (3) no locking devices on
pipe clamps.
The NRC inspector verified that all of the potential
deviations were documented on ERC deviation reports (DRs)
and subsequently by TU Electric on NCR M-86-23228N.
The
NCR was dispositioned "Repair / Rework" for the spherical
bearing gap and no locking devices on pipe clamps and
"other" for pipe clamp parallclism.
A safety
significance evaluation and calculation were provided for
the pipe clamp parallelism; these two documents conclude
that pipe clamp parallelism is acceptable as-is for this
installation.
The NRC inspector has reviewed the closed NCR and
supporting documentation and concurs that TU Electric has
properly dispositioned the NCR in accordance with
applicable requirements.
This item is closed.
b.
(closed) Unresolved Item (445/8607-U-27):
CPRT
identified a potential for switching non-ASME and ASME
Code class valve bonnets and concluded that there was "no
implication that switching of non-ASME and ASME valve
bonnets could be safety significant."
The NRC reised
questions as to how the results report considered the
differences between non-ASME and ASME Codo requirements
with respect to material identification and traceability,
welding and weld repairs, nondestructive examinations,
and personnel qualifications.
As stated in NRC Inspcction Report 50-445/88-32;
50-446/88-28, documentation provided by the vendor dated
April 8, 1988, does not support the conclusions presented
in the results report.
In this inspection period, CPRT
provided errata to the ISAP VII.b.2 Results Report which
definos diffcrences between the valve bonnets and
manufacturing process and modified the Results Report
conclusion to "tharo is not a significantly greater
likelihood that a safety significant defect would remain
undetected .
" and ".
it is very unlikely that
. .
. .
interchanging a non-ASME bonr.ct with an ASME bonnot on an
ITT-Grinnell diaphragm valve would be safety-significant."
_ _ _ _ _ _ _ _ .
_ _ _ _ _ _ _ _ _ _ _ _ .
_ _ _ _ _ _ _ _ _ _
_ _ _ _ _
4
.
.
5
The NRC inspector agrees that this conclusion is more in
line with the documentation.
This item is closed.
c.
(Closed) Unresolved Item (445/6656-U-03; 446/8852-U-02):
This item dealt with an inspection method for hand tight
bolts that could not ensure that bolts that were
overtightened would be detected.
During this inspection
' period, the NRC inspector and SWEC civil / structural
personnel witnessed an inspection of two monorails which
had a requirement for the bolts on one end of each
monorail, where it attaches to the building structure, to
be handtight.
The results of these inspections revealed
that several of these bolts were overtightened, which
i
substantiated the concern.
Subsequent to these
inspections, review of the records for the PCHVP
'
inspection effort performed-in accordance with FVM-090,
determined that this condition had been identified by the
inspector and documented on NCR-88-09973 and
NCR-88-10306.
Therefore, the results of the PCHVP
. inspections confirm that these inspections are adequate
to resclve this ccncern.
This item is closed.
d.
(Open) Unresolved Item (445/8856-U-05; 446/8852-U-03):
This item documented three examples of deviations from
the requirements of the applicable wold procedure
specification (WPS).
Two examples had unconsumed filler
material, and one had an excessively large fit-up gap.
,
The applicant has initiated an evaluation of the extent
'
of this condition and its safety significance under
Corrective Action Request (CAR)88-028.
In this
inspection period, SWEC presented its proposed sampling
plan to be utilized as part of the correctivo action for
this CAR.
'
'
I
The SWEC plan calls for a random sample of welded joints
l
to be inspected.
The weld joints will be inspected for
unconsumed filler material and compliance with fit-up
.
requirements of the WPS.
Deficiencies will ha documented
i
and evaluated in accordance with NEO 3.05.
?.'o date,19 wold joints have been inspected by QC, SWEC,
and the NRC inspector.
In this sample, there were no
'
further examples of the use of unconsumed filler metal to
l
compensate for poor fit-up; however, there were two
additional examples of deviations from the fit-up
requirements of the WPS.
These deviations were
documented on NCRs 88-16728 and 88-16729.
This item will
remain open pending completion of the SWEC review and
closure of the CAR.
i
h
l
l
!
, , .
.
.-
'
is
1
e
d
.
6
f
3.
Follow-up on Violations / Deviations (92702)
a.
(Closed) Deviation (445/8601-D-13):
QI-058 requires the
ERC inspector to verify that the lockwire on snubbers is
t
not damaged.
The NRC inspector identified that although
'
the lockwire was broken, the inspection checklist item
-
for this attribute was accepted.
The NRC inspector verified that this deviation was
documented on out-of-scope observation number 289 which
was' subsequently transferred to NCR M-86-23447H and
dispositioned to replace the broken lockwire.
Further,
the requirements of QI-037 and QI-058 have been revised
to include a more comprehensive inspection' checklist and
instructions for the inspection and signoff of lockwires.
A further supplementary reinspection was performed on the
!
pS 7N and PS 42 populations (47 supports) to verify that
f
other lockwires had not been damaged.
The NRC inspector
has reviewed the closed NCR, tha revisions to QI-037 and
QI-058, and the reinspected supports and concludes that
the NCR was dispositioned correctly and that the
revisions to the procedures provide a clearer
understanding of the inspection requirements.
This item
is closed.
b.
(Closed) Violation (EA 86-09 Appendix A Item I.B.5):
The
Technical Review Team (TRT) found that the design
analysis for non-Category I equipment offects on seismic
Category I structures, equipment, or components was
incomplete.
The support installation for
nonsafety-related conduits less than or equal to 2 inches
,
was inconsistent with seismic requirements and no
!
evidence could be found that substantiated the adequacy
<
of the installation for nonsafety-related conduit of any
i
size.
The applicant denied this violation as written, because
l
they stated that there is documented evidence that
!
nonsafety conduits larger than two inches in diameter had
been properly analyzed and adequately supported and
restrained where necessary.
For conduit less than or
equal to two inches in diameter, they admitted the
,
violation.
!
l
During this report period, the NRC inspector reviewed
i
Calculations SCS-1740, Set 1, Revision 4; and SCS-226C,
l
Set 1, Revision 2, performed by Gibbs & Hill (G&H).
These calculations providu documentation for Train C
,
conduit supports and allowable spans between supports for
2-1/2 inches diameter and larger conduit.
Thorofore, the
i
NRC inspector concurs that there is evidence that these
conduits had been analyzed and scismically supported.
t
r
!
- - -
-
- -.
.
.-
r
'
.
7
.
However, since these conduits have been included as part
of the corrective action program and are being
reinspected and reverified, adequacy of these G&H
calculations was not reviewed.
In response to this issue, as it applies to Train C
conduit less than or equal to 2 inches in diameter, CPRT
performed an independent, third-party review of the
adequacy of the installation of these conduits to assure
compliance to regulatory requirements'and FSAR
commitments.
Because the initial reviews performed as a
part of the response to this issue exhibited an
unacceptably high failure rate, the applicant elected to
resolve the concern by performing a 100% seismic
qualification program as part of the Corrective Action
Program (CAP).
The NRC staff has performed multiple
inspections of both the methodology and the
implementation of this phase of the CAP and concur that
this program brings the installed Train C conduit less
than or equal to 2-inches in diameter into compliance
with the FSAR commitments and regulatory requirements.
To ensure that future installations meet the applicable
requirements, the applicant issued Construction
Procedure ECP-19B to provide the criteria and
documentation requirements for the fabrication and
installation of supports in Category 1 buildings for
Train C two inch and under conduit systems.
Thc NRC
inspector reviewed this procedure and concurs that this
procedure provides adequate instruction to ensure that
all applicable requirements are satisfied.
Further, the CAP program for Train C conduit less than or
equal to 2 inches in diameter consists of a multilevel
screening program based on the existence of
safety-related targets and other structures that may have
been assumed to protect safety-related targets from
adverse interactions with Train C conduits.
Therefore,
the program provides for reviews of now installations of
safety-related plant features as potential targets for
Train C conduit which may fail during an SSE event.
It
also provides for reviews of the removal or modification
of existing structural features that may have been used
as a barrier to protect safety-related features from a
potential Train C conduit failure.
These reviews will be
continued as needed after construction completion by
Comanche Peak Engineering (CPE).
The method for
performing these reviews is documented in Impell Project
Instruction (PI)-0210-053-013, Revision 1.
The NRC
inspector reviewed this procedure and finds that it
adequately addresses all the regulatory requirements and
FSAR commitments.
This violation is closed.
,
.-
.
8
.
4.
Action on 10 CFR Part 50.55(e) Deficiencies Identified by the
Applicant (92700)
a.
(Closed) Construction Deficiency SDAR-CP-86-80:
This
item dealt with the fact that the manufacturers
recommended torque values for certain tubing supports had
not been incorporated into the applicable installation
procedures.
This item was initially determined to be
nonreportable, because the applicant felt that the
manufacturer's recommended torque values were met and in
most cases exceeded.
However, stress in the clamp due to
overtorquing was not addressed.
During requalification
efforts in 1987, this deficiency was evaluated and
determined to be reportable and the NRC was notified in
the applicant's letter TXX-88140 dated January 29, 1988,
and further actions will be reported under SDAR-CP-88-21.
SDAR-CP-86-80 is considered closed.
Actions will be
tracked under SDAR-CP-88-21 which is open.
b.
(ClcJed) Construction Deficiency SDAR-CP-85-36:
The NRC
TRT axamined the installation of nonsafety-related
conduit supports in selected Category I areas of the
plant.
According to Regulatory Guide 1.29 and the CPSES
FSAR Section 3.78.2.8, seismic Category II and nonscismic
items should be designed in such a way that their failure
would not adversely affect the function of safoty-related
components or cause injury to plant personnel.
No
evidence could be found that substantiated the adequacy
of the installation for nonsafety-related conduit.
On September 6, 1985, the NRC was notified that
TU Electric had determined that there was an
indotorminacy of their compliance with the Regulatory
Guide with respect to the installation of
nonsafety-related conduit, 2-inch diameter and smaller
and would be evaluated under CPRT Program Plan Item I.c.
On November 5,
1987, TU Electric determined that this
item was reportable under the provisions of
As corrective action, the applicant
instituted the Train C Correctivo Action Program (CAP)
for Unit 1 and Common Arcas.
In previous inspection
reports, the NRC inspection staff, as well as
Headquarters personnel, have documented review of the
CPRT Issue Specific Action Plan I.c,
reviewed the
procedures for physically validating the conduit under
the PCHVP, and performed extensive independent
inspections of the overall CAP for this commodity.
Based
on the above, the NRC inspector concurs that the initial
concern has boon adequately addressed.
To prevent recurrence of this deficiency, Cesign Basis
Document (DBD) DBD-CS-093 titled "Seismic Adequacy of
L
-
_
_ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _
_ - _ - _ _ _
_ _ - _ _ _ _ _ _ .
' . '
u
'
.
'
9
.
.
Train.'C' Conduits (Two Inch Diameter and Less)" was
issued to provido an adequate design basis.
Also,
Construction Procedure ECP-19B was issued to provide
guidance'for future inctallation of Train
'C'
conduit
supports. -The NRC inspector has. reviewed these documents
and concurs that they provide adequate guidelines to
prevent a recurrence of the stated problem.
Further,
PI-0210-053-013 has been written to document the method
to maintain compliance from initial walkdown onward; the
NRC inspector concurs that this procedure is
comprehensive and will aid in assuring compliance to the
Regulatory Guide and FSAR commitments.
i
This deficiency is closed.
c.
(Open) Construction Deficiency SDAR-CP-85-51:
This item
concerned the application of low sulphur content A-588
I
steel in pipe whip restraints.
NCR.M85-20ll25 was
!
written to document that matorial was installed that did
not comply with design requirements.
Specifically, note
No. 4 on drawing 2323-S2-0581-01 states, in part, "The
material shall be A588 Gr. A or Gr. B and the sulfur
This
"
content shall be limited to 0.010% .
. . .
,
'
condition precipitated Stop Work Order (SWO) 85-03 which
documented a material control problem.
Actions taken in
response to the SWO dealt with the specific material
!
referenced and did not address any review of
documentation to determine if thoro are any generic
implications associated with this, or any other loss of
material segregation.
Therefore, this item will remain
,
open pending response to this concern,
i
d.
(closed) Construction Deficiency SDAR-CP-85-15:
This
item involved the equipment hatch on the inside of the
!
Unit 1 Reactor building.
The concern was that the
I-beams that act as guidos for the equipment hatch and
the guides that rest on the I-beams could transmit forces
l
from the hatch to the internal building structure during
a scismic event.
G&H performed an analysis, which was
reviewed by the NRC inspector, which shows that tho
>
l
forces that could be generated are very low and as such
do not constituto a safety concern.
This SDAR is closed.
c.
(Closed) Construction Deficiency SDAR-CP-87-87:
Westinghouse's ovaluation of the location of relief
valves and piping configuration disclosed a possible
problem where the design pressure of a chemical and
volumo control system lino could be exceeded because of
!
the location cf a relief valvo.
This section of pipino
has boon analyzed for offsito radiological doso effects
under the High Enorgy Line Break (HELB) - outside
containment analysis program.
The results of the HELB
i
l
,
-
'
'
.
.
,
.
10
'
-
analysis indicate the release.of radioactivity would be'
well within the design basis.
However, the valve in
question is scheduled to be relocated to comply with ASME
B&PV Code requirements.
This SDAR is closed.
5.
Corrective Action Program (CAP)
a.
Conduit Supports Train A and B and Train C Greater Than_2
Inches
FVM-033 (48053)
During this inspection period, the NRC inspector selected
a sample of six conduit support modifications to verify
the accuracy and correctness of the final QC inspection
data.
The following is a list of the NRC inspected
supports and the corresponding DCA.
Support
C12012285-06
39137 Revision 2
'C13G12831-02
41012 Revision 1
CO2018343-07
33063 Revision 2
CO2018213-20
41775 Revision 1
C13C10414-03
25919 Revision 2
C13G16456-07
25798 Revision 2
The NRC inspection of those supports was performed in
accordance with the requirements of the applicable
section of Specification 2323-SS-16B and the DCA.
The
results of this inspection revealed that the recorded
data were within the tolerances of the specification.
No violations or deviations were identified.
Based on the NRC inspector's review of this activity, it
is apparent that both craft and QC personnel are
cognizant of the requirements of the applicable
proceduros and specifications and are performing their
work in a manner which is producing a quality product,
b.
Piping and Pipe Supports (50090)
During this inspection period, the NRC inspector reviewed
'
SWEC Calculation GENX-315, Revision 0, and Calculation
Chango Notice No. 1 to this revision of this calculation.
This calculation was initiated in response to the
excessive water corrosion found in areas of the Station
Sorvice Water System caused by failure of the internal
!
plasite coating.
The corrosion caused localized pipe
l
wall thinning in various areas of this system.
The
purpose of this calculation was to dotermino the minimum
pipe wall thickr. css required to satisfy the ASME Code
dasign requirements and the system functional capability
i
,_
-
_ _ _ _ _ _ _ ___ _ _ _ _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
- _ _ - _ _ _ _ _ _ _ _ _ _ _ -
_
. - _ . ___ __
__
. ' . -
i[* ,
,,
11
requirements outlined in CPPP-7, Revision-4.
Later, this
'
calculation method was utilized to justify that damage
caused tcr the piping during the coating removal process
r
need not be repaired prior to operation.
Review of this calculation found that the method of
analysis was lcgically arrived at and meets code'
requirements.
However, there were many inconsistencies
found in the results of the analysis.
These
i
inconsistencies, most of which resulted from the preparer
not adt.ering to the procedure outlined in the method of
analysis, resulted in the tabulation of minimum required
pipe wall thickness having many nonconservative results.
Calculation Change No. 1 was-prepared to perform a
similar analysis for the 10-inch portion of the Station
i
Service Water due to the damage caused in the coating
removal process.
NRC review of this change'to GENX-315
,
revealed that for this esiculation effort, the preparer
failed to follow the established procedure in overy
instance reviewed, resulting in nonconservative results.
i.
'
Calculation GENX-315 and the change notice were prepared
by the same person; however, the reviews were performed
by three different individuals all of whom failed to
detect the errors.
Further examples of this type of
'
design. review concern were cited in NRC Inspection Report
50-445/88-11; 50-446/88-09.
The NRC inspector reviewed
five design calculations which all contained errors that
had not been detected in the review process; this
indicates that there is a potentially significant flaw in
SWEC's design review process.
3
l
Further, the NRC inspector reviewed audits performed by
!
the applicant's Technical Audit Program tTAP) on the
'
design efforts at various SWEC locations including the
site.
The audits reviewed were ATP-87-03 performed at
i
the New York office, ATP-87-09 performed at the Cherry
'
Hill offico, ATP-87-14 performed at CPSES, and ATP-87-56
-
performed at the Houston offico.
In all of those audits,
similar errors were detected.
Recommendations made by
the auditors in several instances stato that there is a
need for improvement of the review process.
,
niso, reviewed were soveral of the action items from the
>
,
!
EFE audits.
Many of the concerns documented in the
action items identify the checkor (reviewer) or the
'
!
review process as the focal point of the concern.
The
typical SWEC response to this type of audit finding was
that the preparer was advised of the error that had been
'
i
mado, additional calculatio;.s performed by the preparer
wore reviewed (in some instancos the reviewer's work was
'
r
'
, _ . . _ _ _
,_ _
_ _ _ _ ___
.'
- . -
,
'
12
-
also reviewed - but this appears to be the exception
,
rather than the rule) and no safety significance was
found.
Also, that the concern in question would be
addressed in the final reconciliation phase of the piping
and pipe support effort.
The TAP group accepted this
type of response stating that they would confirm the
effectiveness of the proposed action during the
reconciliation effort.
However, the EFE group accepted
this type of responso and closed out the items.
This
represents a weakness in the EFE audit program.
The audits reviewed by the NRC inspector were performed
in early to mid-1987 and additional concerns by the NRC
were documented in a February inspection report and the
latest example was taken from a calculation performed in
August 1988.
The fact that the same types of errors keep
recurring indicates that SWEC's corrective actions are
not offectivo, and do little to indicate that their
efforts in the final reconciliation will be any better.
Also, reviewed by the NRC inspector in this inspection
period was SWEC'a Engineering Assurance
Procedure EAP 5.3, Revision 3, entitled "Preparation and
Control of Manual and Computerized Calculations (Nuclear
Projects)."
This procedure establishes the requirements
for the preparation, review, and control of all manual
and computerized engineering and design calculations.
The requirements of this procedure for reviewing
calculation results list several options, one of which
allows the reviewer to "cyoball" the results to determine
if they are "appropriate to the objective .
."
This
. .
type of requirement, when viewed in light of the number
of errors found in the sample reviewed, is obviously not
producing a quality product.
Further, this procedure
does not require the reviewer to indicate the type of
review performed; therefore, there is no auditable trail
to the method of review utilized.
A proceduro such as
this, which endorses cursory design reviews, sends a
'
message that the review process is a relatively
,
unimportant task, and dees little to emphasize the need
'
for quality.
Many of the errors that have gono undetected stem from
che preparors failure to follow the applicable
procedures, which for the most part appear to be complete
- though they may not be presented in the most usable
format.
This indicates a potential flaw in the training
program and procedures.
On October 28, 1988, a meeting was held between
representativos of the NRC, TU Electric, and SWEC to
discuss the above concerns.
After the concerns were
I
-.__.r
_ _ _ _
.
_ _ _ ,
_ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . . _
_ _ _ _ , _ . . , _ ,
_
,
-
.
.
'.
13
.
detailed by the NRC, SWEC presented an overview of their
design and design review process.
It was their opinion
that while they strive for perfection as an ideal, that
they recognize that it is not readily achievable;
however, they feel that, even though errors and omissions
are made and are not detected in the review process, the
final product they are delivering is a quality product.
Their position is based on the fact that when errors in
their calculations have been found, they have been able
to perform the corrected, or alternate, calculations and
not exceed the allowable capacity of the support assembly
in question.
However, in many cases the percentage of
allownble stress increases and this could present a
future concern to the applicant.
If this condition were
to go undetected and future piping analyses caused design
loads to increase, the applicant may not be able to
assume that the stress levels presented in SWEC
calculations are accurate and would be required to
reverify the entire calculation for each affected
support.
The NRC concerns discussed above were also discussed at
the inspection exit meeting on November 1, 1988.
Pending receipt of additional information from the
applicant and the results of further NRC reviews, this
will be an unresolved item (445/8871-U-01).
During a plant tour, the NRC inspector observed that a
spring hanger SI-1-336-001-522S, located in room 66, did
not have a sight hole in the load coupling; therefore,
thread engagement could not be verified.
The NRC
inspector reviewed the documentation package for this
support and found that the inspector who inspected this
support during the PCHVP walkdown indicated that the
thread engagement was satisfactory even though the
inspection report form identifies the need for a sight
hole to verify this attribute.
When brought to the
attention of the applicant's personnel, NCR-88-16375 was
written to document this condition along with several
other discrepant items.
This failure to properly inspect
th.is support is a violation of Critorion X
(445/8871-V-02).
Also observed by the NRC inspector during a plant tour
was a sway strut on support SW-1-132-039-A43R that had
very little clearance in the clamp / forward paddle
connection.
This clearance is provided to accommodate
angular rotation in this connection caused by
piping / equipment movement in the unrestrained directions.
(
The sway strut in question was manufactured by NPSI and
is identified as an SRF-08 strut with an SPC-08-100 pipe
l
l
-
+-
'
~.
4
,
,
.
14
.
clamp.
The NRC inspector measured the distance bnLW4en
the ears of the clamp, at the location of the inner
,
spacer and at the load pin, to be approximately 5/8-inch
and the thickness of the forward paddle to be 1/2-inch.
Based on the obserygd dimensions, this sway strut would
ongy accommodate +3
of angular rotation rather than the
+5
that is assumed by design engineers.
Forthgs
support, DCA 47818, Revision 4, states that 3.78 11 s
needed in the direction in question.
This means that
because of this condition (internal interference between
component support parts) there will be forces and moments
imposed on the piping and the remainder of the support
assembly that were not evaluated in the original design
calculations.
Further, the condition cited meets the
applicable inspection criteria of Brown & Root ASME
Quality Proceduro AQP-ll.3, Revision 4, DCN 2; therefore,
appropriate quantitative acceptance criteria have not
been provided to QC inspectors for this attribute.
Further, since this condition exists to a lesser extent
at nominal dimensions, it also appears that there is
little basis for the tolerances established by the vendor
for the dimensional attributes critical to this design
feature.
This condition has potential generic
implications to an approved design widely used on this
project.
This failure to properly translate design bases
into procedures and instructiens is a violation of
Criterion III (445/8871-V-03).
Based on the abovo vic1ations and the concerns stated in
the unresolved item, there appears to be significant
programmatic weaknesses in the pipo support engineering
and inspection program.
c.
piping Systems and Supports - FVM-081; FVM-090 (50090)
The PCHVP by either physical validations or through an
engineering evaluation methodology, assures that each of
the final acceptance attributos defined in the attribute
matrix is validated.
During this report period, NRC
inspection focused on procedures which provido the
guidelinos for performing and documenting tho engineering
ovaluations of piping system final acceptanco attributes
identified by the attribute matrix.
Future NRC
inspections of the PCHVP will provide additional
evaluations of these activities.
The NRC inspector interviewed supervisory personnel from
Quality control, Quality Assurance, SWEC, and
TU Electric's licensing personnel to determine
managements involvement in assuring quality.
These
personnel wore cognizant of the appl' abic procedures for
controlling specific activitics in their area of
,_
,
_
_ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _
_ _ _ _ _
_
__
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ -
Y
<
.
,
.
15
responsibility.
However, as discussed below, NRC
inspection of management's involvement in assuring
quality with respect to engineering evaluations required
by ECE 9.04-05, PCHVP engineering evaluations, determined
that policies were poorly stated and understood in that
procedures do not describe all methods currently being
used to disposition inaccessible attributes.
Therefore,
compliance with ECE 9.04-05 is indeterminate.
The purpose of the PCHVP is to domenstrata that existing
systems, structures, and components are in compliance
with the revised installation and design requirements,
which provide a methodology for the resolution of
technical issues from a safety standpoint.
PCHVP
commitments are implemented by Field Verification Methods
(FVMs).
Procedure ECE 9.04-01 defines an FVM as:
"A procedure prepared to control the activities of field
verification activities to substantiate the design
criteria used for design calculations, design drawings
and specifications and provides the data necessary to
demonstrate compliance with the approved design."
The NRC inspector reviewed FVM-081 and FVM-090 which are
two of the FVMs applicable to piping and pipe support
installations.
These FVMs are implemented by specific
engineering and quality centrol procedures which are
identified in the reference section of the FVM.
NRC review of FVM-090 noted that paragraph 7.7,
Revision 5, states, in part:
"If, during the execution of reinspections required by
this FVM, any components are identified by QC as
unverifiable or uninspectable (e.g., due to item not
being installed, subject to in-process work, hidden or
inaccessible) an NCR shall be initiated."
NRC review of Quality Control Procedure AQP 11.5 and
CP-QAP 12.1, which are implementing procedures for
FVM-081 and FVM-090 noted that certain attributes which
were inaccessible or unverifiable by QC were exempt from
inspection requirements although the attribute motrix
identifies "Reinspection required" for the attrib 2tes.
The NRC inspector requested project personnel to provide
the basis for this departure from program commitments and
was informed that DCA-58252, which altered the
,
i
installation inspection requirements of 2323-MS-100,
l
Revision 9,
provided the basis for changes to the QC
i
procedures.
SWEC engineering personnel stated that
Procedurc ECE 9.04-05 provided a method for evaluating
the attributes exempted from inspection requirements, but
l
, _ _ _ _ _ _ _ _ _ _ _ _ .
. _ _ _ _ _
_ __________
. _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
___ _ _ _ _
_ _ _ _ .
.
C
.
16
were unable to provide the NRC inspector evidence of
appropriate procedural control of the described technical
evaluation.
Changes to the PCHVP which deleted the
requirement to document inaccessible attributes with an
NCR (i.e. DCA 58252) were approved by SWEC Mechanical
Engineering, SWEC Material Engineering, SWEC Pipe Support
Engineering, SWEC Quality-Assurance, TU Electric Quality
Assurance, TU Electric Quality Control, and other
discipline supervisors.
Subsequent to NRC inspection
activities, a SWEC interoffice memorandum dated
October 13, 1988, requested a review to verify the scope
of the exempted attributes and initiation of the required
technical disposition.
NRC review of enforcement history identified a similar
violation (445/8718-V-09) in which the PCHVP program
failed to adequately address the identification of
inaccessible attributes so that an engineering evaluation
would be performed.
A QC inspector failed to identify an
inaccessible attribute with an NCR which precluded the
required engineering evaluation.
Corrective steps taken
to avoid further violations consisted of procedure
revision to specifically address the initiation of NCRs
for inspection checklist attributes which are
inaccessible.
NRC assessment of the applicant's
performance in this area indicates that corrective action
was not completely effective, dealt only with the
identified problem, and was not effective in preventing
recurrence of a similar problem.
NRC review of QC PCHVP
inspection results revealed that in many cases an NCR had
been issued by QC to document specific cases where
attributes could not be verified because the piping was
buried or in close proximity of other components which
prevented physical validation of the associated
attributes.
NCR 87-02743, nevision 0,
is an example of
QC documenting buried piping with an NCR.
This NCR was
dispositioned "use-as-is" with the following technical
justification.
"This is no longer a nonconforming condition.
DCA 58252
does not require the reported condition (s) to be
reinspected."
The NRC inspector asked SWEC engineering personnel to
provide an explanation of the technical justification
supporting the NCR disposition.
Appendix A of the
TU Electric Quality Assurance Manual defines "use-as-is"
as:
"A disposition which may be assigned for a nonconformance
when it can be established that the discrepancy shall
result in no conditions adverse to safety and that the
.'
O
h
=
17
item under consideration will continue to meet all
engineering functional requirements.
This disposition
requires technical justification by Engineering of the
item's acceptability for use."
SWEC personnel were unable to provide technical
justification for the NCR disposition.
Engineering
evaluation of the inaccessible attributes for the buried
piping, per ECE 9.04-04, has not been performed.
The
described conditions had been determined not to be safety
significant without a technical basis and the disposition
was reviewed by the Quality Assurance organization and
closed.
NRC review of enforcement history concerning the
dispositioning of NCRs identified a similar violation.
NOV 445/8832-V-01 identified that during the evaluation
of pCHVP nonconformances, engineering personnel failed to
establish that the discrepancies would result in no
conditions adverse to safety and that the item would
continue to meet all engineering functional requirements.
The applicant's root cause analysis appears to have dealt
only with the evident problem area in that corrective
action did not identify similar nonconforming conditions
which had been dispositioned, approved, and closed.
The NRC inspector noted during review of the applicant's
corrective steps taken to avoid further violations in
response to Notice of Violation (NOV) 445/8832-V-01, with
respect to training effectiveness and qualification, that
adequate training may not have boon parformed.
TU Electric's letter TXX-88522 dated July 1, 1988,
states, in part:
-
"The personnel involved in this occurrence have been
reminded of the need to fully document engineering
justifications when dispositioning NCRs 'use-as-is', no
longer a nonconforming condition."
A formalized training session in the specific
requirements for dccumenting the technical basis for
"use-as-is" dispositions would have boon more
appropriate.
NRC inspectors and applicant personnel met on October 28,
1988, to discuss the NRC inspection findings.
Deficiency
Report (DR) C-88-05182 has been issued to address
inadequacies in the PCHVP evaluation of inaccessible
attributes.
In summary, NRC inspection concludes that documented
instructions appropriate for performing technical
&.,
<
u
.
,
r-
~C
,
18
,
.
dispositions of inaccessible attributes exempted from
PCHVP reinspection did not exist in all cases.
Furthermore,.the conditions identified on NCR 87-02743
and other similar NCRs do not have an adequate technical
justification por NCR proceduro requirements.
.The NCR har been reviewed by QA and closed although
engineering personnel have not established that the
indeterminate conditions are in compliance with
applicable requirements.
These NRC inspection findings
are a violation of Criterion V (445/8871-V-04).
These departures from PCHVP commitments are indicative of
programmatic weaknesses in that the actual construction
process can not be reexamined for inaccessible
attributes.
Therefore, with respect to quality
assurance / quality control for these unverifiable final
acceptance inspection attributes, it is mandatory for the
applicant to establish reasonable assurance that
hese
attributes are in compliance with all applicable Jesign
specifications.
6.
Plant Tours (50090, 48053, 48063, and 52051)
The NRC inspectors made frequent tours of Unit 1, Unit 2, and
common areas of the facility to observe items such as
housokooping, equipment protection, and in-process werk
activitics.
No violations or deviations were identified and
no items of significance were observed.
7.
Unresolved Items
Unrosolved items are matters about which more information is
required in order to ascertain whether they are acceptable
items, violations, or deviations.
One unrosolved item
disclosed during the inspection is discussed in paragraph 5.b.
8.
Exit Mooting (30701)
An exit mooting was conducted November 1,
1988, with the
applicant's representatives identified in paragraph 1 of this
repert.
No written material was provided to the applicant by
the inspectors during this reporting porlod.
The applicant
did not identify as propriotary any of the materials provided
to or reviewed by the inspectors during this inspection.
During this mooting, the NRC inspectors summarized the scopo
and findings of the inspection.