IR 05000395/1989200
| ML20006B895 | |
| Person / Time | |
|---|---|
| Site: | Summer  |
| Issue date: | 01/24/1990 |
| From: | Imbro E, Lanning W, Parkhill R Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20006B891 | List: |
| References | |
| 50-395-89-200, IEB-79-02, IEB-79-14, IEB-79-2, NUDOCS 9002060065 | |
| Download: ML20006B895 (27) | |
Text
_ _ _ -
,:
-
-. '
U.S. NUCLEAR REGULATORY-COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION DIVISION OF REACTOR INSPECTION.AND SAFEGUARDS Report No.: 50-395/89-200 Docket No.: 50-395 Licensee: South Carolina Gas & Electric Company P.O. Box 88 Jenkinsville, SC 29065 Facility Name:
Virgil C. Summer Nuclear Station-Inspection at: Virgil C. Summer Nuclear Station Inspection Conducted: November 27 through December 1, 1989 and December 11 through.15, 1989 Inspection Team Members:
R. W. Parkhill,.RSIB, NRR - Team Leader P. S. Koltay, RSIB, NRR A. J. H. Lee, EMEB, NRR J. S. Ma, ESGB, NRR
- D L.-Solorio, RSIB, NRR A. V. duBouchet, Consultant 0. P. Puri, Consultant
,
R. E. Serb, Consultant.
- S. O. Forsberg, Foreign Assignee (Sweden)-
'
- Present Second Week Only Attended Exit Meeting:
S. G. Guthrie, RSIB, NRR J. J. Blake, RII'
L. P. Modinus, Resident Inspector - Summer iLfd Pr/E$
/- M -Po
?
Ronald W. Parkhill, Team Leader Date Signed
['
Reviewed By:
_ tt'bt'l A MMnr-
[v
'
1'/ 9d Eugene V. Imbro, Chief f
Date Signed Team Inspection Development Section B 2f O
Approved By:
l//h
'
Nayrp D. Eanhi~ng, Chief D(te 51tjned
-;
-
Special Inspection Bran l NRR 9002060063 900126
PDR ADOCK 05000395 Q
'
.
,
.
.
.
,
,
SUMMARY INSPECTION REPORT 50-395/89-200 VIRGIL C. SUMMER NUCLEAR STATION j
An inspection team composed of NRC staff and consultants n. viewed South Carolina
,
Electric & Gas (SCE&G) Company's compliance with Inspection and Enforcement Bulletin (IEB) 79-02, " Pipe Support Baseplate Designs Using Coccrete Expansion Anchor Bolts,"land IEB 79-14, " Seismic Analysis for As Built Safety-Related '
Piping Systems." The inspection was conducted from November 27 through
December 1,1989 and December 11 through 15, 1989.
For IEB 79-02, the inspection team was satisfied that SCE&G had demonstrated that the expansion type anchor bolts and baseplates were adequately designed i
and constructed. The team was particularly pleased with the full quality control (QC) utilized during construction in that 100 percent of expansion
!
anchor bolts installation was QC verified. The licensee had complied-with the
'
bulletin. However, the team identified that too low a value of bolt stiffness
!
was utilized in the computer program to verify anchor bolt and baseplate sizing.
t For IEB 89-14, the inspection team was satisfied that SCE8G had complied with i
the bulletin. The inspection team noted-that the. design drawings correctly represented the as-built configuration of the plant and that the calculations were well organized and auditable. However, the inspection team identified inconsistent modeling techniques used by different piping analysis contractors, unconservative piping analysis modeling practices, and unjustified engineering assumptions.
,
Appendix A contains the deficiencies identified by the inspection team.
Appendix B lists the minor discrepancies identified by the inspection team whose individual or collective safety significance is minimal, and SCE&G is
.
. expected to resolve the Appendix B items without the need of additional NRC inspection.
Examples of some of the most significant deficiencies are.briefly-
discussed below:
'
.
1.
An inadequate evaluation of the effect of equipment nozzle loads was performed for the residual heat removal (RHR) pump and emergency feedwater
'
!-
(EFW)pumpturbine. The inspection team identified that the RHR pump nozzle loads exceeded the allowable by 30 percent, however, the' pump.
'
vendor was contacted during the inspection and evaluated and accepted this increased loading. Also, the piping analysis' calculation did not include
.
the extra loadings for the EFW pump turbine steam supply' nozzle associated
'
with the mass of the governor valve and the excessive gap in the first upstream seismic restraint. This raised a concern with regard.to the
'
adequacy of other pump nozzle loadings.
,
2.
The effect of the service water pump on the piping system was not evaluated even though its natural frequency was less than 30 hertz, i.e., flexible equipment. This practice was contrary to gal procedures which required-that flexible equipment effects be evaluated. Other similar configurations were not evaluated by SCE&G to ensure that this practice was not a per-vasive error at Summer.
3.
Various pipe support snubbers located next to the rigid anchors in the suction piping of the RHR pump did not appear to have enough movement to
.i
.
'
-..
.
-
.
.
...
.-.
,
.
.
.
e
perform their. intended function during a seismic event. SCE&G did not confirm that the subject snubbers could lock-up and did not review this concern for other.similar applications, c
f There were four issues icentified by the team which require further evaluation
,
by the NRC and listed ir. Appendix C.
A staff position regarding these issues
'
will be made and communicated to the licensee. These issues included:
,
1.
Zero period acceleration (ZPA) was not uniforn.ly considered at Summer.
.'
One of the four piping analysis consultants, Teledyne, did not include consideration of ZPA within its scope of work (i.e., 33 analyses) when the
'
other three consultants included it in their analyses. This is viewed by the inspection team to be an example of inadequate design control..SCE8G did not recognize the need to ensure that all piping design analyses were performed using a consistent et of design criteria, nor did they require.
their prime architect engineer, Gilbert Associates, Inc. (GAI), to
review the work place methods used by the other contractors.
2.
SCE&G excluded consideration of seismic anchor movements (SAM) less than'
1/8 inch without any quantitative technical basis for this engineering assumption. Also, SCE&G inappropriately combined adjacent structure SAMs
,
by the square root of the sum of the squares rather than by absolute
'
values.
3.
Containment penetration movements were not considered in the piping analysis' for the effects of post-accident pressurization or steady state temperature growth.
Apotentiallynonconservativepipingdecouplingcriterion.wasutiljzedand no justification for using this criterion was available.
Several strengths were identified by the inspection team and are discussed in detail in Section 4 of the report. Areas where strengths were identified included document retrieval, drawings. reflecting as-built condition of plant,
,
'
design basis documents, installation of expansion anchor bolts, and piping Code l
compliance.
(
l'
,
!
l
!
1^
L
.
.
..
.
-.
-
..
-
..
-
1.
l -
. ~..
..
[.
.
Virgil C. Summer Nuclear Station
,
Bulletin 79-02 and'79-14 Followup Inspection November 27 Through December 1,'1989 December 11 Through 15, 1989 i
1.. INTRODUC1 ION f
1. '1 Background Information InspectionandEnforcementBulletin(IEB)79-02, Revision 0,."PipeSupport Baseplate Designs Using Concrete Expansion Anchor Bolts," was: issued on j
.Narch-8,1979 to ensure the adequacy of the design and installation of -
pipe support baseplates using concrete _ expansion anchor bolts.
IEB 79-14, Revision 2, " Seismic Analysis for As Built Safety-Related Piping Systems',"
was issued on July 2,1979 to ensure conformance between the as-built safaty-related piping: systems and the associated seismic analyses.
s As a result of' inspections recently conducted at various facilities, the
-
NRC had indications that the actions requested by IEB 79-02 and 79-14 may not have been properly completed by some licensees. Consequently, the NRC staff decided to review, on a sampling basis, implenientation of these bulletins. The first. plant selected for this review was Arkansas Nuclear One, Unit I and the second was V. C. Suniner.
SCE&G addressed IEB 79-02 and.79-14 by contractin'g with Gilbert Associates; Inc. (GAI) to perform the technical work required to address these-bulletins. GAI was the original plant architect-engineer for Sumer
'
Plant. SCE&G subsequently contracted with Impe11 and Teledyne to assist GAI.
1.2 Purpose of Inspection
.;
The p(urpose of this inspection was to assess South Carolina Electric and'
Gas SCE&G) Company's implementation of.IEB 79-02 and IEB-79-14'. The inspection was performed by a. team composed of members of'the Office of.
Nuclear Reactor Regulation and consultants.
2.
INSPECTION EFFORT AND REPORT ORGANIZATION 2.1 Inspection Effort l
For IEB 79-02 the inspection team reviewed the licensee's program and its-L implementation to verify conformance with the subject bulletin. A walkdown of various pipe supports was performed to evaluate the.acequacy of randomly selected installed baseplates and anchor bolts and to review pipe support designs selected from the piping analyses reviewed by the
team.-
l
-
.
_
.
._
.
.
..
.
. _...
_ _ -_ _ _._______ _ _ _ _ _ _____ _ ___
_ _ _ _ _
_ _. _ _ _ _ _ _ -. _. _ - - _ _ _ _ _ - - _
-
,
>
.
.
..
'
'
u
.
'
For IEB 79-14 five piping analyses were identified by the inspection-teem as potential inspection subjects during_a pre-inspection visit to the.
SCE&G offices on November 13, 1989. These analyses were-identified to
- '
permit SCE&G to assemble copies-of the associated piping calculations and-pipe support drawings such that-the inspection team could begin work.
immediately upon arrival. The five chosen piping. analyses included the following:
(1) Emergency feedwater'(EFW) piping from the pump discharge nozzles to the containment penetrations (Analysis performed by Teledyne,'Subsys-tem Analysis Package EF-02, Revision 4, dated November 9,1982);
'(2) EFW turbine steam supply piping (Analysis performed by GAI, Subsystem.
Analysis Package MS-07, Revision 6,, dated May 6,1987);
'
(3) Residual heat removal (RHR) suction piping from the containment penetrations to the RHR pump suction nozzles (Analysis performed by Teledyne, Subsystem Analysis Package RH-03, Revision 2,-dated
,
May 28, 1982);
(4) Main feedwater safety-related piping-located outside of containment.
- Revision 6, dated March 4,1985)ystem Analysis Package FW-13A, (Analysis performed by GAI, Subs
(5) Service water piping located in the service. water _pumphouse building (Analysis performed by Impell, Subsystem Analysis Package SW-10, Revision 4, dated May 15,1986).
The inspection methodology included comprehensive reviews of design criteria documents, design inputs, drawings, computer models, and other I
applicable design basis documents to ensure-compliance. The piping runs modeled in'the five analyses were walked down on a sampling-basis to L
verify agreement between the installed configuration' and the analytical model.
,
2.2 Report Organization
!
Section 3 lists those practices which were viewed by the inspection team-to be indicative of' licensee weaknesses.
Section 4 provides those licensee good practices that were viewed by the inspection team to be indicative'of areas of compliance and are identified as licensee strengths. Section 5 provi N the overall conclusions reached by the inspection team.
>
Three appendices are attached to the report. Appendix A lists the
l deficiencies identified by the. inspection team.. Appendix B lists the L
minor discrepancies whose safety signiticance'is minimal. Appendix C.
'
lists the four issues that require additional NRC review. Appendix D lists the personnel contacted during the inspection.
l l
l l
t L
!
.
_
--
.
-.
.. -
.
.
___
_ _ _ _ _ _ -. - _ _ _ ___
- _ -._
.
.
.
'
...
3.
LICENSEE WEAKNESSES 3.I Design Control'
During the IEB 79-14 design effort, SCE8G did not ensure that consistent
~
design criteria or techniques were used by their' piping analysis con-
,
tractors. SCE8G did not require GAI, their architect-engineer, to review-
,
- the analysis methods used by the other piping consultants to ensure consistent piping analysis results. SCE&G had contracted directly with
{
l'
Impell and Teledyne to perform piping analysis and GAI's role was relegated
-
L to supplying. basic information such as piping configuration, response spectra curves, pipe sizes, etc. Although SCE&G was aware'of a 20 percent difference-in piping analysis results between.two of the consultants for the same analysis, they failed to question why the results were different t
and take the appropriate action.
'
!.
An example of-an inconsistently. applied design criter. ion involved the consideration of.the effect of equipment flexibility upon the piping analyses [ Appendix A, Open Item 89-200-01(a)]..GAISpecificationNo.
SP-702-4461-00, Revision 4, dated February'11,.1977. required consideration f
of equipment flexibility but Impell did not-include the flexibility of the-service water pump in the service water piping analysis.
This example illustrates that the design criteria / techniques utilized by the various piping analysis. contractors had not been properly coordinated, l
Another example of inconsistently applied design criterion was the con--
sideration of zero period. acceleration (ZPA).
In particular not consider ZPA within its scope of work (i.e., 33 analyses}.Teledyne did
, although.
the other three consultants included it in their analyses (Appendix C, item 1).
'
The inspection team identified two examples ~where equipnent. nozzle load-i L
ings exceeded the vendor specified allowables.
First, an. error was identified in the method utilized to calculate the RHR pump' nozzle loadings.
'
This resulted in the vendor allowa'le loading being exceeded by 30 percent.
o
Although'the pump vendor eventually a'ccepted the increased loading, this isanexampleof.inadequatedesignverification[AppendixA,OpenItem 89-200-02(a)]. The'other equipment nozzle loading issue was that the effect of valve mass [A)pendix A, Open Item 89-200-02(b)])andpipesupport seismic restraint gaps Appendix A, Open Item 89-200-02(c ] were omitted
'.
from consideration for the emergency feedwater turbine steam supply nozzle. The team concluded that, had more rigorous design verifications
been completed, these types of. errors may have been avoided.
L The inspection team identified other findings that-demonstrated weaknesses in the design-control program, including' omission of structural steel stiffner plates required by the governing AISC Code (Appendix A, Open Item 89-200-03), and insufficient seismic movement of the piping to lock-up the snubbers (Appendix A, Open Item 89-200-04).
For IEB 79-02, the inspection team-identified that too low a value of bolt stiffness was utilized in the computer program to verify anchor bolt and baseplate sizing (Appendix A, Open. Item 89-200-05).
Bolt stiffness is defined as the. force per unit axial deformation and includes slippage due to loading.
l l
.
+
--
-
.-wr s.-
w
-.-.----s----w e
.
-*.--w
+- -
.
- -
e
.
....
.
.-.
..
._____-- - ____--.
.
-.
..
- _.. -
,
_
J
,
,
'
.
l
'
L 3.2 Unjustified Engineering Assumptions (
During the inspection, the inspection team _ identified var _ious practices
-
L for which SCE&G oid not have 'an acceptable technical basis.: For adjacent structure seismic anchor movements (SAMs) (Appendix C, item 2) and con--
l-tainment penetration movements (Appendix C,-item 3) no-quantitative
,
technical justification was provided for SCE&G's practice of not coa-l sidering movements of one-eighth' of an inch or less._ Accepted industry
-
l-practice is to neglect movements of 1/161 inch or less.
Both of these
,
!
issues will-receive further review by NRC.
SGE&G did not consider internal service water building SAMs even though
,
i the USAR paragraph 3.7.3.8.1 committed to its evaluation'(Appendix A,'Open l_
Item 89-200-06). SCE&G identified that the directional ~ magnitudes of a
seismic anchor movements between adjacent buildings would be-calculated i
by taking the square root of the sum of the squares of the relative l
building movement rather than the industry practice of using absolute value [ Appendix C. item 2(b)]. Finally, SCE8G used a potentially uncon-servative branch line decoupling criterion without a documented technical
~
basis (Appendix C, item 4).
.
t
'
3.3 Engineering Support for. Piping Analysis The inspection team noted that the SCE&G engineering' organization-lacked personnel experienced in, piping analysis and pipe support ~ design. SCE&G engineering personnel were generally operations oriented and acted as administrators who relied on GAI'and other' contracting organizations, such
,
as Impe11 and Teledyne, to prepare design' modifications' packages for the-plant.
In the piping area, SCE&G lacked experienced engineering personnel to technically review the work product'of the contractors to ensure a uniform design product.
3.4 Modeling Practices
'
In reviewing the piping analyses, the inspection team checked the coding
of computer input against design informatic,n and identified some'modeling
inconsistencies which amounted to incorrect-and underprediction of
design loads and stresses. The representative examples included incorrect modeling of the as-built orientation of struts [ Appendix-A, Open Item 89-200-01(b)],stiffnessesofsupportsnexttonozzleswhichweretoolow to perform their functions [ Appendix A,_Open Item 89-200-01(c)],one thermal load case for the service water system which was omitted
[AppendixA,OpenItem 89-200-01(d)], flexible equipment was modeled as i
L rigid [AppendixA,OpenItem-89-200-01(a)], and the potential programmatic-
'
misuse of pipe support selective stiffnesses (Appendix A, Open Item 89-200-07). Although some of the aforementioned specific errors had'been
shown to be acceptable, the generic impact of these unconservative modeling practices had not been evaluated.
Similiarly, the team identified cantilever branch lines in the SW-system-that had not been modeled in the piping analysis (Appendix A, Open Item 89-200-08).
.
>
4
..
.
.
_
..
.
.
.
.
a
-__
_ _ _ - _ _ _ _ _ _ _ _ _.
._
_ _ _ _ -
--
-
._.
-
,
.
L-
!
'
,
.
.:
,
,
3.5 Updated Safety Analysis Report (USAR)
The piping and pipe support design was verified by the inspection team to:
be generally in compliance with the.USAR commitments.
However, the team
.1 did-identify a few exceptions.
USAR Section 3.7.3.8, Analytical Procedures for Piping. required evaluation of differential piping support movements within a structure. 'USAR Table 3.7-7a tabulated total relative floor displacements-for the-service-water (SW) pumphouse of about 1/4 inch for the. design basis earthquake (DBE) between floor elevations 425' and 436'.
However, GAI did not require these'SAMs to be considered in the seismic-qualification of.the SW pumps, or require Impell-to consider these seismic anchor _ movements:(SAMs)
in the piping analysis of the SW discharge piping. GAI has indicated that j
the relative floor displacements tabulated in USAR Table 3.7-7a represented
'
a mix.of rigid-body rotation and intra-floor deformation.. GAI concluded,
,
but could not demonstrate, that the deformational component of these SAMs
<
l
.
was negligible. The team requested that SCE&G either. provide justification
!
that the displacements tabulated in.the USAR were due primarily to rigid-
<
body rotation or consider these SAMs in the-seismic qualification of the
'
SW pumps'and discharge piping (Appendix A, Open Item 89-200-06).
Also, the last paragraph of USAR Section 3.7.3.8.1' required consideration l
of SAMs between adjacent structures. However, SCE8G did not consider SAMs of one-eighth inch and-less and provided-no_ quantitative technical justi-fication for this practice-(Appendix C, item 2).
GAI procedures included a modification priority criteria which implied that piping and pipe support Code allowables could be exceeded by as much as 50 percent without implementation of'any hardware change. SCE&G indicated that GAI used this modification priority system only.to cate-gorize condition severity for' subsequent consideration and disposition lby SCE&G. The inspection team was concerned that SCE&G had an established policy whereby Code allowables could be routinely exceeded which con-stituted a deviation from their commitment to meet Code allowables-(Appendix A, Open Item 89-200-09). ~ Any design. change which exceeds. Code allowables is required to be submitted to the NRC for review and approval
because it would be outside the design basis of the facility.
!
4.
LICENSEE STRENGTHS
,
4.1 Document Retrieval a
During the period of the inspection, the team-learned that many of the Summe plant's design documents resided at gal, Impell or Teledyne and not.
,
at SCE8G. Moreover, SCE&G-did.not appear to have any comprehensive.
indices, with the exception of a-drawing index, for the design documents:
a L
that the contractors maintained for the Summer Plant. The team found, j
however, that SCE&G was able to obtain in a timely manner all of' the design documents that the team requested and determined'that SCE&G had an l
adequate document control system. These document types included:
l t
'
r
.l
.
.
.
.
.
.-
.-.
--.
.
,
e
'
Piping physical drawings Piping stress isometric drawings (GAI series 314 drawings)
(
' Piping fabrication drawings Pipe support detail drawings (GAI series 321 drawings)
y Design specifications and procedures for piping, equipment.and pipe
.
supports.
.
[
Equipment procurement documents-l Equipment seismic qualification documents Vendor equipmentidrawings
- '
Pipe stress calculations
Pipe support calculations 4.2 Design Isometric Drawir,gs Reflect As-Built Condition
GAI, Impell, and~ Teledyne. re-analyzed all of the safety-related piping
-
systems required to address IEB 79-14, -1.e., as-built. piping and support configurations depicted on'the GAI series 314' drawings.
Because these l
GAI drawings also serve as pipe stress isometric drawings, a check of the
drawings from walkdowns confirmed the as-built piping and support geometry
'
shown on the drawings as well as the geometry used to prepare the piping-
mathematical models. The inspection team was generally satisfied that
)
these isometric drawings accurately documented the as-built configuration of the plant.
4.3 Design Basis Documents.
SCE&G had GAI! prepare system oriented design basis documents (DBDs).
-These documents have been issued for many years and serve as a convenient
'
and accurate source of design information.- TheLinspection' team reviewed portions of.Section 4.0, Component Design Basis, of the DBD and verified that the functional design information which the DBD specified for selected components was also specified in the component; design specifications and
procurement documents. The team also verified that the operating and design temperatures and pressures that the DBD specified were also specifieo in the cssociated system design specification and on the GAI flow diagram.
4.4 Installation of Expansion Anchor Bolts To ensure that all expansion anchor bolts had been properly installed, SCE&G instituted a complete quality control program that witnessed and verified the installation of each bolt. This is an acceptable-and pre-
ferred method of anchor bolt verification as compared to the random-testing methodology.
4.5 Piping Code Compliance The Code of record for V. C. Summer's safety-related piping is-ASME Boiler and Pressure Vessel Code Section III, 1971 Edition..with Addenda-through-Summer,-1973. The inspection team verified that SCE&G had designed the-piping to this Code Edition and Addenda.
s
.
-
..
.
.
.
_..i
,
..
.
__
_ _ _ _
. _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ -. -
.
.
,
'
,
.
-
.
.
5.
CONCLUSION
,
'
For IEB 79-02 the inspection team was satisfied that.the anchor bolts instal-lation had been adequately verified and that the design loading were appro-priate. However, verification of the PRYTEN computer program was not adequately
>
demonstrated to ensure that the bolts and baseplates had been conservatively
,
analyzed.
For IEB 79-14 the inspection team concluded that the licensee had demonstrated that the as-built plant configuration had been adequately documented on design drawings. -However, errors were identified which involved-undocumented engineering judgements, modeling unconservatisms, and design control incon -
'
,
sistencies which had not been resolved by SCE&G.
The inspection-team' concluded that SCE&G complied with IEB 79-02 and 79-14.
The deficiencies identified questioned the appropriateness of engineering assumptions and analysis techniques rather than the adequacy of actions taken by the licensee to comply with these bulletins. These~ deficiencies did not i
. raise any signficant safety concerns regarding the acequacy of safety-related
<
structures, piping, or as-built configuration.
,
i l
l t
i
.
-
-.
.
_
......
.-.
-
.
.
..
.
w l
)
,
\\
-
'
..
APPENDIX A Significant Deficiencies The deficiencies identified in this appendix are considered to be the most
- significant issues resulting from this inspection and are numbered separately.
<
They are categorized by their applicability to the individual bulletins.
SCE8G
-
should individually track all issues identified during the inspection such that
- '
their formal resolution can be verified at a -later date.
1.
Piping Analysis Modeling Errors in the Service Water (SW) System
'
(0 pen Item 50-395/89-200-01)
,
a.
Seismic Interaction of Piping / Flexible Equipment SW pumps 39-A. -B and -C were seismically qualified by the pump manufacturer's consultant -Mcdonald, in calculation'No. ME-298, dated April 19, 1976. The STRUDL dynamic analysis of. the pump indicated that all of the pump natural frequencies in the lateral and vertical j
modes were greater than 20 Hz (the frequency at which the magnitudes I
of the SW pumphouse design response spectral values approach ZPA l
values) except for one lateral frequency mode of 14 Hz.
For such l
flexible equipment, GAI specification No. SP-702-4461-00, Revision
'
No. 4, dated February 11, 1977 required thatJthe equipment vendor develop a new response spectra envelope-at the point of attachment of.
piping or provide a dynamic model of the equipment to GAI. The
'
latter. requirement inferred the incorporation of the equipment
.
dynamic model into the piping analysis. However GAI did not require-
!
I Impe11 to address these requirements in Impell calculation No. SW-10, Revision No. 4, dated May 15, 1986 which analyzed the SW piping on s
Coupling the SW peeps and the discharge side of the SW pump (or. decrease) the computed pipe
'
,
discharge piping could increase
'
stresses and support reactions, b.
As-Built Orientation of Pipe Support Struts Not Modeled Properly Pipe support SWH-4017.was depicted on GAI drawing No. S-321-251, l
Sheets 4017A-E, Revision No. 5, dated May 22, 1981. The as-built
'
configuration of the support was schematically depicted on GA1 Drawing No. C-314-251, Sheet.17, Revision No. 3, datea.
February 24, 1989. Pipe support SWH-4016 was similarly configured..
These pipe supports were installed near the discharge. nozzles of SW pumps 39-A and -C to reduce the axial loads acting on'the pump
!
discharge nozzles due to pump fluid transients, thermal expansion and l
contraction, and seismic loads. These pipe supports were installed
'
in the portion of the SW piping system that was analyzed in Impell calculation No. SW-10, Revision No.'4, dated May 15, 1986. As noted
'
on GAI Drawing No. C-314-251, the paired struts for pipe supports
.
SWH-4016 and -4017 were skewed rather than parallel to the
!
centerlines of SW discharge pipe trains A and C.
As a consequence,
-i
,
i these pipe supports were also capable of providing lateral restraint in the horizontal and vertical directions. However, Impell calcula-
.
i tion No SW-10 only modeled these pipe supports as axial restraints.
'
g Each of these pipe supports was fixed to a concrete wall by two
A-1
.-
__
_
..
.
-.
.
.
.
.J
.
.
_
_
._
_
_
._ _
__ _ _ _ _
_ _ _ _ _ _
y
.
.:
.
.
.
,
l surface-mounted plates with a total of 12 1-1/4-inch'Hilti bolts.
,
k-Because Impe11 stress calculation No. SW-10 did not also model these L
pipe supports as lateral restraints,'the Impe11 pipe support calcula-:
-
tion did not compute the combined tension. loads induced by the
imposed axial and-lateral forces. The GAI calculations which quali-
fied pipe supports SWH-4016 and -4017 in-accordance with the.Impell'
j L
pipe support loads were unconservative.-
L t
L c.
Stiffness of Adjacent Pipe Supports Less than:the Stiffness of the
!
Equipment
]
Pipe supports SWH-4016 and SWH-4017,'as: described in b above,' and the SW pump discharge nozzles shared the axial forces which SW. trains.A-and C induced in direct proportion to their relative axial stiff -
[
-
nesses. Therefore, pipe supports SWH-4016 and -4017 provided sub-
stantial axial restraint only if they were much stiffer than the SW pump. These pipe supports were installed in the portion'of the SW-l
piping system that was analyzed in Impell calculation No.: SW-10,.
Revision No. 4, dated May 15, 1986. The Impell: calculation stated
-
'
but did not verify the assumption that the axial rigidity of the pipe supports " greatly exceeds" the magnitude of.the axial stiffness of
.
the pump (8.6 x E6 pounds per inch) which was modeled;in the analysis.
..
'
However, the inspection team identified that this assumption was-
'
incorrect, because the axial stiffness of each pipe support was about 5 x E6 pounds per inch.
d.
Omitted Thermal Load Case
<
As noted in design specification DSP-544H-044461-000, Revision.No. 9, i
dated August 22, 1984 for the SW system, the temperature.of the water-on the suction. side of'.the SW pumps could vary from 32 to 95"F.
Impell stress calculation No. SW-10, Revision No; 4, dated May 15,-
L 1986 considered the 35'F hydrostatic condition. in computing the
'
l support reactions for the normal and upset conditions, but failed to
'
recognize that 35'F (actually 32*F) was an operating mode, and not a
temporary condition associated with the hydrostatic test. While the 1'
Impell stress calculation computed the pipe support reactions due to the 35 F hydrostatic condition, the stress calculation did not i
i l
consider the 35 F operating mode when tabulating the axial loads
!
which acted on the SW pump dischcrge nozzles and.did-not consider this thermal mode when computing the displacements of the.run pipe at branch line locations. The branch lines were qualified by-consider-
ing the 35 F as the-lower value for temperature range.
]
e.
Fictitious Thermal Loads at Service Water Pump Nozzle l
Impell calculation No. SW-10, Revision No. 4, dated May. 15, 1986
'
'
analyzed the SW discharge piping depicted on GAI Drawing No.
,
C-314-251, Sheet 17, Revision No. 3, dated February 24, 1989. To
.
model the axial behavior of the SW piping in the vicinity of the l
discharge nozzle for SW pump 39-A, the node point at the location of the SW pump discharge nozzle was coded with an axial spring between p
the node point which represented the point of fixity of pipe support
SHH-4017 against axial movement.
These node points were located
'
.
'
'
A-2 i
.
.
_ _ _ _,,_-.
.~
_
__
,, _. _ _
-
.
'
..:
.
'
.
I approximate 1y'six inches apart on either side of a 24-inch diameter 150-pound welded neck raised face flange.- The piping analysis coded'
-
the flange to expand'and contract under the imposed thermal loads..
'l As a' consequence, the 95'F operating mode induced an axial compres-sion of about 5,000 pounds in the flange, and the 35'F hydrostatic test induced an axial tension of about 7,000 pounds in the flange.
j These-loads-were reacted at the. adjacent node points. However, these-
'
loads, while mathematically correct, werb fictitious because they were induced by the infinitesimal axial elongation and contraction of the flange. The train C SW discharge piping in the vicinity of SW
_
pump 39-C was identically coded.
At the end of the inspection SCE&G had not formally. issued a revised
SW piping analysis calculation to demonstrate that consideration the aforementioned discrepancies had not exceeded code allowable stresses.
Also, SCE&G had not performed generic reviews: of deficiencies la,11b,
and Ic to demonstrate that these errors did not occur in.other
'
safety-related piping analyses.
2.
Equipment Nozzle Loadings
.
(0 pen-Item 50-395/89-200-02)
,
a.
RHR Pump The inspection team compared the combined RHR pump suction and
,
discharge nozzle loads to the allowable' loadings identified in Westinghouse pump specification 678815, Revision 2, dated September 6, 1983. The combined _ nozzle loadings exceeded the vendar allowables by 30 percent. After confirming thisimistake, SCE&G requested the pump manufacturer to accept the increased loading. _The pump manufacturer and Westinghouse eventually accepted the' increased loading on the nozzle, however, SCE&G and its_ piping _ contractor did not initially identify and resolve'this error.
-
b.
Valve Mass Omitted from Analysis Model The mathematical model used to analyze the steam. supply piping-to the.
!
EFW turbine excluded the mass of the' trip'and throttle valve-(2865-MS)-
located'close to the turbine steam' inlet. The valve's weight (640 pounds) was significant relative to allowable-nozzle loads ~
,'
specified for the turbine. One effect of excluding this mass was to significantly. underestimate gravity and seismic loads on the turbine nozzle.
During the. inspection SCE8G did not reanalyze or otherwise demonstrate acceptability of the current design with proper' con-sideration of the valve mass.
c.
Excessive Pipe Support Gap Design Pipe Support MSH-336 was a steel frame located approximately six inches upsteam of the turbine trip and throttle valve (Valve 2865-MS). The support was designed to restrain'both static and dynamic pipe loads in the lateral direction.- In the vertical-oirection, gaps of 1/32 and 1/16 inch were specified between the support and the top and bottom of the pipe, respectively. The
,
A-3
.o i
.-
.
..
.
.
l
,
'
.
,
purpose of maintaining these small gaps was to allo'w unrestrained vertical pipe movement'under thermal and gravity loading:while-restraining the pipe seismically. During the inspection, the: team reviewed the pipe support design, the installed configuration, the.
,
- analyzed mathematical model, and the analysis results.. The~ pipe'
support was found to be installed.as. designed. ;Asiintended, the ll mathematical model specified dynamic restraint while permitting.
)
unrestrained.movementiin response to thermal:and gravity loads.
~
Comparison of the gravity and thermal' load resultant displacements ati
-
this pipe support confirmed that for the. current analytical model, the support would not, as:. intended, restrain pipe movement.- However, the inspection team identified that the relatively stiff piping configuration in the. vicinity of this restraint and the adjacent
'
,
turbine nozzle would result in a very.small dynamic ' displacement at the restraint location..Therefore, the; pipe support gaps were too-
,
large to permit it to be loaded and the turbine nozzle would pick-up
the resultant dynamic load...During the. inspection,2SCE&G did not reanalyze the affected piping or otherwise justify the assumption.
a that the support woult' restrain dynamic pipe-loading.
'
In addition to not resolving the specific issuescidentified, SCE&G-J
'
did not provide'any additional assurance that equipment nozzle-loadings have been properly considered at Summer such that errors
,
similar tocthe ones identified above:had not occurred in other safety e
systems.
3.
Omission of. Stiffeners for Pipe Support Structural Steel j-(0 pen Item 50-395/89-200-03)
y GAI. calculation No. MSH-002, Revision No. 2, dated January 29, 1982
~
qualified pipe support SWH-084 'a vertical strut fixed to the bottom of a e W4x13 structural steel shape..The strut rear bracket was welded to the bottom flange of the W4x13.
Based on the requirements of the seventh edition of the'AISC Code (the Code of record for the Summer plant).
Section 1.15.5, page 5-40, the.W4x13 required stiffeners. However, the
pipe support calculation did not evaluate the W4x13 for this Code
~
~
-
requirement. Alto, SCE8G did not provide assurance from a programmatic perspective that the need for stiffeners was appropriately considered in
.'
the design of other pipe supports considering both compression and tension loads.
,
,
4.
Insufficient Dynamic Movement to Lockup Snubbers (0 pen Item 50-395/89-200-04)
In the review of the RHP, suction piping analysis, the inspection team identified various snubber configurations that were located relatively.
- close to rigid anchors. These snubbers were shown at pipe support loca-
'
tions RHH-140, RHH-148, RHH-164 and SIH-068 as shown on isometric drawings C-314-641, Sheet 3, Revision 4; C-314-641, Sheet 3, Revision 4, C-314-641, Sheet 4, Revision 4; and C-314-691, Sheet 2, Revision 2, respectively.
The inspection team believed that these snubbers.did not have sufficient
dynamic snovement to perform their intended function of locking up.during a design-basis seismic event. During the inspection SCE8G did not demon-
>
strate that these components would function as designed, and as a con-sequence, the existing piping analysis might underestimate pipe stresses
,
l h
'
A-4
'
.
.
. _.. - _,
_-,-.., _,
. _ _ _ -. _ - _,
.
__
_. _
_
.
._
_
_
.
,
.
,
,>
'
,
and pipe support loadings.
If any of these snubbers are determined to-l have insufficient dynamic movement;to lockup.the snubber, SCE8G had not evaluated the specific piping arrangement or performed a generic review to ensure that other-safety-related systems do not have similar errors.
l 5.
PRYTEN Computer Program Qualification q
(0 pen Item 50-395/89-200-05)
The computer program, PRYTEN, was developed by GAI to verify the. adequacy of the pipe support anchor bolts and baseplates pursuant to IEB 79-02. A review of a verification document-for the PRYTEN program revealed that'
sample problems wero benchmarked against another computer program, BASEPLATE II, and good. comparisons were achieved.
However, the.-inspection team identified that.the stiffness of bolts used in the sample runs was
'
unreasonably low. -Bolt stiffness is defined as the force per. unit axial i
deformation and includes slippage due to-loading.. As a consequence, SCE&G
. l performed other PRYTEN verification runs for the same sample. supports.
using a realistic bolt stiffness. The results showed that the bolt force
increased substantially with the worst case increasing by a factor of six:
when compared to the benchmark results..This substantial increase in bolt
,
force could not be explained during the inspection. The licensee had not provideo further_ verification to ensure that use of the PRYTEN program did not result in inadequately designed anchor bolts and baseplates.
6.
Building Internal Seismic Anchor Movements (SAMs)-
l (0 pen Item 50-395/89-200-06)
b USAR. Table 3.7-7a tabulated the horizontal and vertical displacements of
~
the SW pumphouse floors due to the OBE. These displacements were a mix of rigid-body rotation and intra-floor deformation.
If these displacements were not' due primarily to rigid-body rotation, then these displacements should be considered in the seismic qualification of the SW pumps and the
- i SW discharge piping which was depicted on gal DrawingiNo. C-314-251,
,.
,'
l Sheet 17, Revision No. 3, dated February 24, 1989. The inspection team requested SCE&G either to provide evidence that the displacements tabulated
!
in the USAR table were due primarily to rigid-body rotation or to consider these SAMs in the seismic qualification.of the SW pumps and discharge'
i piping. Also, SCE&G had not explained how the effect of' internal building SAM was evaluated for safety-related systems in other seismic Category I buildings.
- -
7.
Use of Selective Stiffnesses in Modeling Pipe Supports
(0 pen Item 50-395/89-200-07)
Section 6.3.4-c of GAI Report 2439 indicated that, in order to reduce conservatism, calculated restraint stiffnesses might be specified in i
piping analysis rather than the very high stiffness values automatically
'
specified by the TPIPE computer program. The effect of using lower calculated stiffness could reduce loads and stresses for expansion type loads (i.e.,. thermal or displacement) and increase loads and stresses for
~
'.
dynamictypeloads(e.g.,spectralseismic). The inspection team was concerned that application of calculated stiffness values could be applied 1-i d
A-5
.
.
,...
.,,,
_
. _.,
,
9e,_,
,..%..
-
_ _ _ _ _ _ - _ _ _ - - _
__
_
- _ _ _ _ _ _ -
_
-
.
.
-
.
.;
i
'
.
.
.
>
.
in an inconsistent fashion thereby_ taking advantage of their effect in.
.
selected circumstances. Examples of selective application of stiffnesses-
'
which are considered inappropriate are the use of calculated stiffnesses for the static load case analysis only and. for very local regions within-
!
a piping analysis..During the. inspection, SCE&G did.not respond to the-
,
inspection team's request for~a description of-how stiffnesses had been.
selectively applied.
,
t 8.
Qualification of Cantilever Branch Lines g
(0 pen Item 50-395/89-200-08)
~Impell Project' Instruction No. 2, Revision No.-1, dated October 27, 1980-
required that-branch lines supported entirely by the pipe run be modeled l
l in the piping analysis regardless of the relative diameters of.the run and branch lines..However, the team conducted a.walkdown.of the SW piping:
l depicted on GAI Drawing No. C-314-251, Sheet 17, Revision No. 3, dated'
a February 24, 1989 and identified several cantilever branchE11nes which I
were not modeled in Impe11 calculation No. SW-10, Revision No. 4, dated-
-May 15, 1986.
The team requested SCE&G to demonstrate that these i
cantilever branch' lines have been adequately' qualified. However, SCE&G1 did not provide this information during the inspection nor review the-relevance of this concern for other safety-related systems.
l 9.
Modification Priority System-Exceeding Code Allowables (0 pen Item 50-395/89-200-09).
Section 9.9 and Appendix D of GAI report.2439, " Pipe Stress Analysis Data,"
Revision 2, dated March 15, 1985 and Section 7.0 and Appendix M of GAI Report 2440, " Project' Pipe-Support Design Data," Revision 4, dated July 11,1989 appeared to specify limits of operation for-conditions under which piping and pipe support design criteria may be exceeded.
For example, these report sections' appeared.to accept conditions of pipe-stress which exceeded Code. primary stress limits by 50 percent and which
l were not checked against Code secondary stress-limits. ' The inspection team was concerned that SCE&G had an established policy whereby Code allowables could be routinely exceeded and this deviation from their licensing commitment to meet Code allowables.had not been reviewed and approved by the NRC.
During the inspection, SCE&G stated that the priority levels identified in these report sections have been used by gal only to categorize condition severity for subsequent. consideration and disposition by SCE&G.
In all cases, when Code allowables were exceeded, a safety evaluation was preparea and submitted to the NRC. resident inspector for his information.
During the inspection, SCE&G committed to remove from both reports those sections which appeared to permit continued operation'with Code allowables exceeded.
This item remains open pending revision of the applicable report sections mentioned above.
10. Misinterpretation of STRUDL Computer Output (Closed)
gal calculation No. SWH-257, Revision No. 3, dated January 2,1982 qualified pipe support SWH-257, a box frame configuration with a brace which was fixed to the concrete by three surface-mounted baseplates
,
A-6
-
, - -
_
.
.
....
.
.
-...
.
.. _ __
.
. - _ - _ _ _ _ _ _ _ - - _ - _ _ _ -
.
.
_
_
.
.
.
l and a total'of 14 one-inch diameter Hilti bolts. A structural analysis
'
'
-(STRUDL)_of the box frame computed the internal forces and moments in the frame, as well as, the reacting forces and moments which equilibrate the-E box frame. However..the calculation used the STRUDL reacting forces and L
moments to qualify the Hilti bolts without changing the signs of these-L quantities. As a consequence, compression forces instead of tension-
- .
forces were incorrectly, used to qualify the bolts. SCE&G had initiated a l
file search to determine whether the' specific designer had improperly l
designed other supports.
11. StiffnessofExpansionAnchorBolts(Closed)
Section 3.9.3.8 of the USAR stated that'the stiffness of an expansion
.
anchor bolt used was less than-that for an embedded stud but greater than
,
that which would be determined-directly from the initial slope of. the force-displacement = curve representing a concrete pullout. test of 'a nonpreloaded anchor. The inspection team requested a technical justifi-
cation for the above statement. The licensee-responded that the word
"nonpreloaded" should be replaced by "preloaded" and initiated a USAR change. The team viewed this as a typographical error in.the USAR and the l
licensee's~ action was acceptable,
,
,
12. Anchor Bolt Design Including SSE (Closed)
I USAR Section 3.9.3.8 stated that the governing load combination for bolts-was a summation of deadweight.. thermal, 1.5 operating-basis earthquake (0BE), and occasional mechanical loads which was less than allowable'
,
anchor bolt load.
However, in Item 3 of responses to'IEB 79-02.. Revision 3, dated February 1982,.it appeared to the inspection _ team that the 1.5 OBE factor had been used only during the design stage and not used during
-
verification. The inspection team requested an explanation regarding what
was the actual governing loads for bolts. The. licensee responded that the
'
governingloadsincludedsafe-shutdownearthquake(SSE),andinitiateda USAR change to clarify that SSE loaos had been utilized during verification i
by replacing the l'.5-0BE factor with SSE load.
13. TensileandShearallowablesforA307 Bolts (Closed)-
-
In reviewing Revision 1 of GAI Piping-Engineering Standard on Baseplates, the inspection team noted that the' interaction curve for A307 bolts permitted tensile and shear allowables in excess of AISC criteria. The
.'
licensee responded that Revision 1 was issued on December 18, 1981 and was not used for the Summer piping verification task. Revision 0 was the version used on Summer and was consistent with the AISC criteria. The inspection team verified that the tensile and shear allowables of Revision 0 were acceptable.
'
14.
Incorrect Floor Response Spectra (Closed)
Table 3.7-7a of the USAR indicated that the vertical acceleration at elevation 425 feet of the Service Water Pumphouse was 0.249 g, which was_-
substantially higher than the corresponding values for the floors above'
i and below.
Furthermore, this high acceleration.value had not been used for piping analysis and the average of the upper and' lower finor response
.
A-7
,-
.
.
.
.
. -.
.
~
.
. -.
.
_
- - -
c
- 10.
n
.
spectra had been used instead. The licensee ~ responded that the correct E
value should be 0.173g, and the error was due to a technician's mistake in reading and utilizing the computer output.and initiated a FSAR change.
-
This was acceptable to the-inspection team since the high acceleration t-value of 0.2499 did not agree with the acceleration sequence of the other floors in the service water pumphouse and the 0.173g value was in.
.
q sequential agreement with the acceleration of the other floors. Therefore, no change to the service water piping analysis was necessary.
,
F I
,
.
l l
!
l
l
,
-
>
I i
A-8
.
<w
-
w x
e
- - -, -,,,-
.__
-
.
- --
-
.- --
._ ___
_ _
-.
__
-
. _ _
.-.
.
.
i
-
'
APPENDIX D Minor Discrepancies
,
,
This appendix lists the minor discrepancies or documented inconsistencies identified by the inspection team that appear to be random isolated errors and, j
therefore, have not have included in Appendix A.
These items 6re noted in the report for conipleteness because they were identified during the inspection.
The safety significence of these items taken individually or collectively is j
minimal.
It is the team's understending that these discrepancies will be i
corrected by SCE&G.
1.
Feedwster Regulating Valve Not Seismically Qualified (Closed)
)
_
The feedwater piping from the intermeciate building interface with the turbine building to the inlet of the feedwater check valves 1684A, B, C-FW was specified as being seismically designed to withstand an operating basis earthquake (OBE) and design basis earthquake (SSE) in accordance
'
with the Design Specification DSP-5440 044461-000, Feedwater(Nuclear)
System Piping and Pipe Supports, Ruision 5, datea February 4,1983.
.
However, the feedwater regulating valves IFV-478, 488 and 498-FW exceeded i
their qualified seismic accelerations which was accepted by GAI because the valves were nonsafety-related. Even though the inspection team was not
,
aware of any safety-related function for these valves, it requested SCE&G
!
to describe what programmatic controls were in place to ensure that these valves would be verified operational prior to re-start after a seismic event. SCE&G comitted to revise Section 5.4 of the earthquake response manual to require an inspection of the feedwater regulating-valves after an OBE.
2.
PipingIsometricMissingTwoInchLine(Closed /
During a walkdown of the energency feedwater (EFW) system discharge piping. the inspection team identified that the two inch line used for steam generator wet lay-up was not shown as a branch line off of a four-inch line shown on piping isometric drawing C-315 085 Sheet'3. As a l
result, the licensee initiated a corrective action to update the drawing.
l l
3.
Wrong Thickness Specified for Pipe Support Friction Plate in reviewing RHR suction piping analysis, the inspection team noted that item number 7 on the bill of material (Lubrite plate) for Pipe Support RHH-098, Revision 8, dated August 18, 1980 was listed as one-eighth inch thick. Subsequent field nicasurement indicated the as-built thickness as one-half inch. The licensee indicated that the drawing would be revised.
4.
Wrong Piping Thickness Modeled The inspection team identified that schedule 40 piping was modeled in l
lieu of the heavier schedule 100 for the containment penetrations numbered j
226 and 316 in the RHR suction piping. SCE&G evaluated the deviation and i
determined that it would have negligible effect on the thermal analysis.
The inspection team concurred with the licensee's response for this issue B-1
.
..
.
..
,
l
.
,
I
'
.
.
-
.
but noted that snubber lock-up would be affected which is addressed
!
separately in Appendix A, item 13.
l
5.
Temperature Break Modeled at Middle of Piping Run With regard to the piping analysis thermal analysis load cases No. 2 and 3 for the RHR suction piping, the inspection team questioned the justifica-
!
tion for changing the temperature in the piping run. Normally, tempera-
'
ture changes are made at closed valves.
For this piping analysis, the
!
temperature step change was made in the middle of piping at Node 4702 as
!
shown on isometric drawing C-314-641, Sheet 3, Revision 4 without any
'
justification.
,
I 6.
Use of Wrong Jet Impingement Loads L
For the RHR suction piping, the inspection team identified that the wrong
-
jet impingement loads were considered in the piping analysis.
As a result, hand calculations were performed-to evaluate the impact on the piping analysis.
Use of the correct jet impingement loading resulted in a 5 percent increase in the maximum stresses which were well below the allowable. The inspection team considered this to be an isolated error 6nd was satisfied with SCE&G's answer because the effect of jet impingement
,
had been considered even though the wrong loading had been used.
7.
Service Water Pipe Support As-Built Versus As-Designed Differences
l
'
SCE&G prepared nonconformance etice No. 3499 on December 1, 1989 to
,
j address two discrepancies which *e inspection team identified during a
.-
walkdown of the SW discharge pipM installed in the SW pumphouse at floor
,
,
elevation 425'-0".
The SW discharge piping was depicted on gal Drawing No. C-314-251, Sheet 17, Revision No. 3, dated February 24, 1989.
a.
A zero gap existed between the body of valve XVG-3102A-SW, which was installed in a cantilever branch line supported off service water pipe train A, and pipe support SWH-4014, which provided lateral support for service water pipe train A.
GAI drawing No. S-321-251, Sheet 4014B, Revision No. 2, dated March 2, 1982 for pipe support
,
SWH-4014 contained the following note:
" Notch flanges if necessary for interference w/ valve & bolted flanges".
However, this note was
'
not implemented when the pipe support was installed. The cantilever branch line in which the valve was installed was therefore subject to
an unanalyzed lateral load when SW pipe train A displaced axially with respect to the fixed pipe support.
b.
A single 1 1/4-inch by 7-inch notch was formed in the flange of each vertical leg of pipe support SWH-4014 to install the support.
However, these notches were not shown on the pipe support drawing,
'
and the calculation which qualified pipe support SWH 4014 did not consider the reduced load capacity of the vertical legs of the pipe support.
-;
.
B-2
. -
_
.
.
-
.
.
_
-
- - _ _ _ - - - - _ _ _. _ _ _ _
___
._
'
!
.
i
'
8.
Pipe Support Calculation Assumes Wrong Number of Bolts
gal calculation No. SWH-4012 Revision No. 3, dated February 9 1982,
qualifiedpipesupportSWH-4012,whichwasaverticalstrutaffixedtothe bottom of a slab by a surface-mounted base plate and four 1 1/4-inch i
diameter HILil bolts. However, the calculation incorrectly computed the
!
tension load which was induced in the anchor bolts on the basis of six, rather than four bolts. The inspection team viewed this mistake as an
}
isolated error _which required a correctio:. to the calculation.
'
9.
Contradictory Instructions for Consideration of Rigid Range Seismic Response GIA Report 2439, piping Stress Analysis Data, Revision 2 dated
March 15, 1985 appeared to include contradictory instructions regarding i
the consideration of rigid range seismic response or RIG stresses.
'
Section 6.3.4-b of the report states that. RIG stresses were to be consid-ered. However, Section 6.4.1 of the report stated that they were not to be considered in Code stress evaluations.
During the inspection, gal noted that the Section 6.4.1 direction to not l
consider RIG stresses was intended to direct proper input specification
'
for the TPIPE computer program stress evaluation subroutine. The inspec-tion team determined that this intent was not apparent and, subsequently, SCE&G committed to clarify Section 6.4.1.
Revision of the report to'
delete or otherwise clarify the requirement to consider RIG analysis
results would be appropriate.
10. Combination of Seismic Spectral and Rigid Range Analysis Results (Closed)
Section 6.3.4-6 of GAI Report 2439 specified the procedure for combining i
i the effects of seismic spectral and rigid range analysis results.
For l
analyses performed using the TPIPE computer program, the report stated I
that for future analysis results from the response spectra analysis and the RIG should be combined by SRSS. Discussion with gal during the inspection revealed that the requirement to conbine results by SRSS applied to new piping only.
Reanalysis of existing piping was to be based on the procedure originally employed for the analysis of that piping.
(i.e., enveloped) piping analyses were usually evaluated based on screened Original seismic spectral and rigid range analyses results.
During subsequent discussion of this topic, SCE&G provided the following clarification and expansion of future piping analysis requirements:
,
I The SRSS technique would be employed for analysis of installations.
whichincludetheadditionofnewhardware(i.e.,pipingorpipe
support) and for complete reanalyses.
,
i The screening technique would be employed for other evaluations,
'
incluaing operability evaluations, for systens which were previously qualified using this technique and which do not include the addition
,
of new hardware.
A decision regarding which technique to employ as part of a contem-plated snubber reduction program would be made in conjunction with the development of such a program.
B-3
_
,.
a
!
-
'
)
.-
l l
The inspection team considered this item to be closed based on the above clarification.
'
,
J i
l 11. Missing Documentation for Increased Turbine Allowable Nozzle Loads t
Section2.13.1ofStressAna'lysisPackage(SAP)MS-07, Revision 6, dated
'
May 6,1987 included a record of a conversation between Terry Corporation i
and GAI regarding increased allowable nozzle loads used for the qualifi-J cation of pipe loads on the Terry steam turbine. Section 2.1 of the SAP
noted that a letter confirming the increased allowable loads had not been
received at the time the SAP verification record was prepared. During the inspection, SCE&G was not able to locate written confirmation of the t
increased allowables which had been used for the nozzle qualification.
12. Center of Gravity Verification for Valve 2865-MS
!
!
l The piping analysis mathematical model of SAP MS-07, mentioned above,
,
specified the center of gravity of Valve 2865-MS at 1.777 feet from the
l system pipe and valve body centerline.
This dimension could not be i
confirmeo based on review of SAP MS-07 or available vendor valve drawings of Gimpel Corporation, Numbered HP-1359 and P-4975, Revision IG dated l
March 14, 1988.
j l
13. Jet Shield Weight Calculated incorrectly Pipe rupture protection for the emergency feedwater steam supply piping
-
included three jet shields fabricated from steel plate and supported directly by the system pipe. The shields were numbered IB412MS-F1,
.
IB412MS-M1 and IB412MS-El and depicted on GAI drawings E-513-138, Revision 8, dated August 28, 1983; E-513-139, Revision 7 cated March 1, 1983, respectively. To account for the shield mass effects on
>
system piping, their weights were calculated and included in the piping analysis mathematical model. The inspection team compared the design, installation and analysis input for each shield.
In all cases the installed hardware compared well to the designs.
However the weight of one shield was incorrectly calculated in Section 2.12.8 of SAP MS-07, and the incorrect weight was input to the analysis model. The weight for Shield IB412MS-El was calculated and input as 75 pounds versus the correct weight of approximately 145 pounds. SCE&G had not reconciled the
difference.
As-Built Support Angles
'
The installed orientation of two snubbers, identified as pipe supports ItSH-315 and MSH-317, did not match the orientations specified on the as-built piping isometric C-314-011 Sheet 8. Main Steam Steam Supply to Emergency Feedwater Pump XPP-8-EF Revision 3, dated January 28, 1983.
The as-built orientation for both snubbers was in an east-west and verti-cal plane at 42 degrees from horizcntal. The inspection team measured the
angle from the horizontal as 47 degrees for both snubbers. The 5 degree
!
discrepancy exceeds the 3 degree support angularity tolerance specified in
,
Appendix E of GAI Report 2439, mentioned above. SCE&G did not provide
'
justification for this difference.
'
,
B-4
.
.
-
-
.
- -
..
.
..
.
.
- -. -....-
.
.-
-
--
-
...-.
.
'
-
....
15. Spring Hanger Setting During the walkoown of the emergency feedwater turbine steam supply line the inspection team identified that the spring hanger identified as item 2 on Drawing MSH-236, Revision 7 might be bottomed out. The design loading
'
was specified as 747 pounds hot load and 760 pounds cold load with a 0.113 inch upward movement. SCEAG subsequently verified that the spring was not fully compressed and that it was acceptably loaded.
!
f P
I I
,
l
>
B-5-
.
-, -
._.
_ _
___ _.
_
_
.
_
__.
_ _ __-_ _ _
-
_ __
_ _ _ _ _ _. _ _
_
.
,
-
!
APPENDIX C
J Items To Be Addressed By NRC 1.
InconsistentDesignControlofZeroPeriodAcceleration(ZPA)
l a.
SCE&G cid not consistently consider the effects of ZPA. One of their
-'
four piping analysis consultants, Teledyne, excluded consideration of ZPA from the 33 piping problems that were within their scope of work
for Summer. The other piping analysis contractors, Westinghouse, GAI,
-
and Impe11, all included ZPA in their respective analyses. The inspection team believed that the root cause of this error was that
SCE4G did not recognize the need to ensure that all piping analysis
~
contractors utilize the same design techniques or criteria when
!
performing piping analysis for Summer. When SCE&G awarded the piping contracts to Teledyne and Impell in the 1979 time frame, they assumed that those companies would need no specific analytical _ technical i
direction and just placed the prime architect-engineer, GAI, in the-
,
role of supplying information and drawings rather than in position of i
assuring consistent application of design techniques / criteria.
Therefore, SCE8G had not provided assurance that other design techniques / criteria have been uniformally considered by the piping contractors.
b.
SCE8G indicated to the inspection team that they did not consider it necessary to evaluate ZPA for the piping analysis performed by Teledyne because it was not Teledyne's policy at that time to include consideration of ZPA.
Because the inspection team believed that ZPA needed to be evaluated f rom a design consistency perspective (i.e.,
design control), this deficiency has been referred to the Mechanical Engineering Branch (MEB) within NRR for resolution.
,
2.
Adjacent Structure Seismic Anchor Movements
,
a.
gal Report No. 2439, Piping Stress Analysis Data, Revision 2, dated March 15, 1985 stated on page 6-46 that absolute displacements between adjacent buildings of less than 1/8 inch can be neglected.
- The report attempted to justify this assumption by stating that there was enough " pin" tolerance to allow for the absorption cf this movement without imparting any load into the piping system. The inspection team recognized a general industry practice of excluding
'
movements of less than 1/16 inch, but was unwilling to accept move-
'
ments up to 1/8 inch without a quantitative technical basis.
Because
,
no such justification was provided, this discrepancy has been-l referred to the MEB within NRR for resolution, i
b.
In Section 6.3.4-c of GAI Report No. 2439, adjacent structure SAM I
effects were combined by square root of the sum of squares (SRSS)
method rather than by absolute sum. During the inspection, SCE&G noted that modal frequencies and dominant modes differ for different l
structures and, therefore, the likelihood of simultaneous maximum l
opposite seismic displacements was very low which justifies the SRSS l
combination method. The inspection team did not agree that this l
C-1
l
-
--
..
. -
-
-
.-
-.
- - - -.-
_,.
_
__
_.
_
.. _.
_ _
____ _ _ _ _
_.-._
_. _ _ _
_.
_ _..
_ -_ _
'
'
!
t
-
,
reasoning was adequate justification for the assumption that maximum
'
effects cannot occur simultaneously. During the inspection, SCE&G did not respond with additional justification for of this method or _
-
evaluate the potential for nonconservative design resulting from its application.
Similarly, this deficiency has been referred to the MEB within NRR for resolution.
3.
Containment Penetration Movements
'
The inspection team identified that containment penetration movements were not considered in the piping analyses for the effects of post-accident pressurization or steady state temperature growth.
During the inspection,
,
SCE&G performed a cursory evaluation and determined each of these move-i ments to be a maximum of 1/8 inch and, therefore, they considered them to be negligibic, again without quantitative justification.
For the post-
-
accident pressurization growth, SCEAG correctly noted that this was associated with a faulted plant condition and, as such, only the primary
stress equations apply to this loading. The inspection team agreed that
post-accident pressurization need not be considered for piping stresses, but identified load increases on the supports due to these displacements and the steady state growth displacements (which already exist) were not evaluated. Similarly, this deficiency has been referred to the MEB within NRR for resolution.
.
,
4.
Potential Nonconservative Decoupling Criteria
-
GA1 Report #2439, Pipe Stress Analysis Data, Revision 2, dated i
March 15, 1985, page 6-20 stated that branch lines and' instrument connec-tions may be decoupled from the analysis model of the larger or main
,
piping system provided the moment of inertia ratio of the two lines was equal to or less than 15 percent. The inspection team questioned this l
engineering assumption because the industry practice was to use a decoupling criteria ratio of 4 percent to 6 percent. Use of the larger decoupling ratio would omit consideration of branch lines.of approximately.
up to one-half the main piping size and could possibly result in under-calculated piping stresses ano pipe support loads.
SCEAG did not respond to this item during the inspection. This potential deficiency has been
referred to the MEB within NRR for resolution.
(
.
..
.
l C-2'
!:
o
.e
-
+..e v
,.,.
y w
p
,
a m
.V
'
-
a i
.
.
'
l APPENDIX D i
I Personnel Contacted During Inspection Name Title Affiliation J. Archie.
Systems Engineer SCE&G A. Barth Technical Specialist Level III SCE8G
- 0.
Bradham V. P. Nuclear Operations SCE&G
'
M. Browne Manager Systems & Performance Engineering SCE8G
- R. Clary Manager Design Engineering-SCE&G
-
R. Cox Supervisor Mech / Chem Plant Support SCE&G
- H. Donnelly Senior Licensing Engineer SCE&G G. Hietpas Supervising Engineer Impe11
,
A. Hoffert Project Engineer-Piping GAI t
- S. Hunt Manager Quality Systems -
SCE8G
'
- K.
Kassam'
Section Manager I:npell
- A.
Koon Manager Nuclear Licensing SCE&G
>
D. Malkmas Engineer SCE&G
'
- G. Meyer Engineer (Mechanical)
SCE&G D. Moore General Manager Engineering Services SCE&G
- K. Nettles General Manager Nuclear Safety SCE&G T. Ogburn Technical Specialist Level I SCE&G l
- C. Rentschler Engineer GAI
!
M. Richard Engineer Teledyne r
C. Rickert Engineer GAI K. Sandman Designer GAI
- J. Skolds General Manager Nuclear Plant Operations SCE8G l
G. Soult General Manager Operations & Maintenance SCE&G
,
'
W. Stuart Engineer SCE&G
- J. Todd Engineer SCE&G J. Turkett Engineer SCE&G
,
J. Waters Technical Specialist 11 SCE&G
- S. Wharton Senior Engineer Teledyne
'
- M. Williams General Manager Nuclear Services SCE8G G. Vaughan Engineer Impe11
- Attended Exit Meeting I
e s
D-1
-
.
-- -
..
-
-
..
--
-
.