IR 05000443/1983023
| ML20206F377 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 06/18/1986 |
| From: | Grimes B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| To: | Harrison R PUBLIC SERVICE CO. OF NEW HAMPSHIRE |
| References | |
| NUDOCS 8606240446 | |
| Preceding documents: |
|
| Download: ML20206F377 (8) | |
Text
- o p il Robert Public Service Company of New Hampshire Post Office Box 330 Manchester, NH 03105
Dear Mr. Harrison:
SUBJECT:
Integrated Design Inspection, 50-443/83-23, Supplement 2 J. N. Grace's letter dated January 29, 1985 conveyed the results of a' November 7-9, 1984 follow-up reinspection to the Seabrook Integrated Design Inspection (IDI).
As a result of this reinspection, nineteen items remained open.
Subsequently, Yankee Atomic provided additional information on open items in response to questions raised by the team and the Office of Nuclear Reactor Regulation (NRR).
In addition, the team, NRR and Region I have performed inspections pertinent to the open items at Seabrook and the offices of United Engineers. The enclosed
)
report indicates that no IDI items remain open.
.
In accordance with 10 CFR 2.790(a), a copy of this letter and its enclosure will be placed in the NRC's Public Document Room.
Sincerely, Original signed by Brian K. Grimes, Director Division of Quality Assurance,. Vendor, j
and Technical Training Center Programs Office of Inspection and Enforcement Distribution:
.
ACerne, RI QAB Reading VNoonan, NRR DQAVT Reading VNerses, NRR
'
JMTaylor, IE CRossi, NRR RHVollmer, IE GHarstead, WESTEC BKGrimes, IE LStanley, WESTEC HJMiller, IE AduBouchet, WESTEC GTAnkrum, IE EVImbro, IE DPNorkin, IE TMurley, RI RStarostecki, RI
-
i SEbneter, RI EWenzinger, RI i
b l
M AB:DQAVT:IE QAB:DQAVT:IE C:QAB:DQAVT:IE
.DQAVT:IE AVT:
%
DPNpr,86 '
kin EVImbro GTAnkrum HJMiller 6/// rimes G
6/gy 6//9/86 6/ /86 6//f/86
/86
.
860
[6240446860618ADOCK 05000443
.PDR
.
-..-.
...
-
-.
.--
. - -
-
-.
. -
.
-.
_
,
,
,
i8
,
lt SEABROOK IDI FOLLOW-UP REINSPECTION
-
.
1.
Background The NRC conducted an Integrated Design Inspection of the Seabrook Unit 1 Nuclear Power Plant between November 1 and December 21, 1983.
The inspec-
.
tion report was issued on April 2, 1984.
The applicant responded to the
!
report on June 29, 1984.
The NRC reviewed the response and an NRC letter
!
dated October 5, 1984 identified IDI items to be addressed in an NRC rein-
!
spection which was conducted from November 7-9, 1984.
The applicant's letter dated November 16, 1984 provided revised or additional responses to
'
the IDI findings, which were discussed during the reinspection.
The rein-spection report, dated January 29, 1985, indicated that 19 IDI items remained
,
open.
For most of the 19 open items, the applicant provided additional in-
,
!
formation addressing NRC concerns.
The paragraphs below indicate that all
!
19 open items have now been closed out and the basis for the close out l
actions.
2.
Mechanical Systems j
(Closed) Finding 2-12, Coolant Velocity at Sump Screen f
This concerned whether the sump inner screen meets the requirements of i
Regulatory Guide 1.82, dated June 1974, i.e. "the design coolant velocity j
at the inner screen should be approximately 6 cm/sec (0.2 ft/sec)".
United i
Engineers calculated a velocity of 0.34 ft/sec through the screen and 0.192 l
at a plane approaching the screen.
United considers that it is appropriate i
to calculate the design coolant velocity at this latter location in order to meet the intent of R.G. 1.82, and therefore the design meets the 0.2 ft/sec
,
j criterion. We agree with this interpretation. -It is noted that Regulatory
'
Guide 1.82, Revision 1, dated November 1985 resulted in a wording change with respect to the design coolant velocity reference point from "at the
'
inner screen" to " ahead of the screen." This change apparently clarifies the intent of the regulatory guide rather.than changing the reference point.
]_
(Closed) Finding 2-17, CBS Pump Motor Seismic Qualification
'
The concern was that a purchase specification requirement was not fulfilled l
for a seismic analysis of the stator end turn insulation support system.
!
Westinghouse confirmed that the specification requirement was not applicable i
to the CBS pump motors because they have a different configuration than that addressed in the specification.
However, the applicant made a commitment
'
to review all safety-related pump motors for the above specification require-
ment in the generic pump motor purchase specification.
!
'
[
,
i t
j-1-
,
!
. _.,. -.
. - -
.
_.,.._ _.-.~..__,______ _ _ _ _ __- _ -__ _ _._ _ _ _ _,, _ _ _ _ -, _ _ _ _ _ _
.
.-
._
.
-
-..
-
-
-,
..
.
.--
-
...
.
,,
!
The results of the review were as follows:
,
,
(1) For other Westinghouse designed motors the specification requirement does not apply either because there is no end ring support system or there l
is a coil support system which is part of an assembly whose attachment to the frame is seismically analyzed.
(2) For Reliance designed motors, there.is an analysis which indicates that
the seismic contribution is a small fraction of yield strength and when conservatively combined with worst case operating stresses provides an 8.58 to 1 safety factor.
!
(3) For GE motors, the basis for not performing a seismic analysis is that the winding and its end turns are designed to withstand radial and tangential magnetic forces associated with normal motor starting (in the order of 15
g's).
United Engineers memo number M 28944A dated February 26, 1986 indicates that testing to confirm this design has enveloped Seabrook speci-
!
fic seismic accelerations pertinent to these motors.
i l
(Closed) Finding 2-19, Jet Im)ingement Analysis
'
(Closed) Finding 2-21, Pipe W11p Evaluations (Closed) Finding 2-22, Cracks in Moderate Energy Piping
'
.
!
These findings concern the fact that work had not been accomplished to
!
assess whether there was adequate protection of essential components from postulated breaks and cracks in high and moderate energy piping.
The team performed a reinspection on May 21, 1986.
Procedure TP-3, Revision 2, i
dated January 10, 1986 defines the methods for conducting Failure Modes and Effects Analysis (FMEA) for addressing postulated piping failures.
!
The FMEA analyses for Seabrook are subdivided on the basis of 83 zones.
l As of May 21, 1986, work for 82 zones had been completed, documented, and j
signed-off by management.
The remaining zone (56A, containment annulus, j
including the regenerative heat exchanger cubicle) had an open item to be
!
resolved with Westinghouse concerning a break-target interaction potentially
causing a rupture in a 3-inch steam generator blowdown line.
letter NAH-U-3571 dated May 29, 1986 confirmed the acceptability of this
interaction.
l The FMEA analyses are based on both drawing reviews and field walkdowns.
The latter serve to verify the drawing reviews as well as to address field
'
t run targets, such as instrument tubing, conduit, and.small bore piping.
l The FMEA group has a program for assessing the impact upon completed zone
packages by "as-constructed" stress analyses, which could have different postulated break points than those analyzed by FMEA.
This program focuses i
on lines where either the piping configuration or support location / type i
have changed or where transient loads could be a significant factor.
!
Moderate energy line rupture analysis is addressed in TP-7, Revision 5,
'
dated May 9, 1986, which evaluated flooding and spray impingement effects on systems or components important to safety. With respect to water spray,
-
motor control centers and terminal boards are most vulnerable.
Other safety-i related electrical components are generally specified with some form of pro-i tection from water spray. TP-7 indicates that there are no motor control j
centers or terminal boards in the 8 areas reviewed, i.e., those having safety-related components and potentially affected by water spray due to moderate
energy line rupture.
i j-2-
- -
_.__._,_. _._.__ _
.
..
. _ - - _ _. _. _.. _ _ _, _ _ _. _ _ ~
,. _ _ _
_,
. _ _ _ _
.
.
The team concludes that the FMEA program and moderate energy line rupture
,
study provide assurance that Seabrook plant arrangement and design features provide the necessary protection of essential systems and components.
(Closed) Finding 2-20, Basis for Analysis Standard Review Plan 3.6.1, Appendix B, Footnote 1 states that "the energy level in a whipping pipe may be considered as insufficient to rupture an impacted pipe of equal or greater nominal pipe size and equal or heavier wall thickness." Because of the differences in the nature of the loads from a whipping pipe and jet flow from a break or crack, the staff has not permitted the above guidelines for whipping pipes to be extended to jet impingement loads without specific justification.
This finding identified that the applicant had extended (without specific justification) the above criteria to jet impingement in its design of some balance of plant piping.
The applicant submitted SBN-1023 date April 29, 1986 which provided justi-fication for applying the above criteria to jet impingement.
The applicant's position is that jet impingement loads induced in equal or larger size target piping and its supporting system would be less severe and provide a greater margin against failure than the criteria provided in SRP 3.6.1, Appendix B regarding the pipe-upon pipe impact loading cases.
To support its position, the applicant analyzed two jet-target interactions which represented worst-case situations (e.g., based on jet source and proximity and orientation of the target to the jet source), which effectively bounded all jet-target interactions at Seabrook.
The applicant's program demonstrated that jet impingement loads on target piping of equal or larger sizes has no severe consequences to the target piping.
3.
Mechanical Components (Closed) Finding 3-20, Connection Design Finding 3-20 addressed the effect of torsional moments induced in a support beam by vertical and lateral seismic loads.
United Engineers performed a stress analysis which confirmed the structural integrity of the configura-tion.
Yankee Atomic performed a series of walkdowns to identify any addi-tional configurations where significant torsional loading could be imposed on framing connections.
The screening criteria for the walkdowns involved
,
i cantilevered pipe support attachments that could be perceived as applying a significant torsional moment within a distance equal to the depth of the i
main supporting beam from the framing connection.
The walkdowns did not identify any additional cases meeting these screening criteria, i.e. where t
significant torsion loading could be imposed on framing connections.
(Closed) Unresolved Item 3-1, Equipment Preoperational Test Containment spray pump CBS-P-98 had been immersed in water due to flooding caused by a break in hydro test equipment.
This resulted in the need to change some parts.
The IDI recommended that the pump should be monitored during preoperational testing to confirm the acceptability of the repairs.
Region I Report No. 50-443/85-17 (approved on July 29, 1985) indicates that this item can be closed out based on results of testing in accordance with TPI-62-F02, Revision 2, Test No. 1-PT-12.1, Revision 1, " Containment Spray System Operational Test." The report indicated (1) the NRC inspector observed pump CBS-P-98 in operation at approximately 325 psig with piping-3-
.
.-
,
.
--
_ _.
- - -.
-
. - -. -.
,.
_
.
_
.
.
and valving in the above test configuration, (2) the pump motor had been
'
dried and meggered and found to be satisfactory, and (3) the pump was
.
operated for approximately 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> without adverse effects.
(Closed) Unresolved Item 3-2, Valve Qualification Test Report Test results for the seismic qualification of 8" containment isolation valves CBS-V11 AND V17 indicated frequencies below 33 Hz, contrary to the requirements of the United Engineers active valve test guidelines; the associated stress analyses had assumed valve natural frequencies greater than 33 Hz.
The Mechanical Analysis group found the test results con-ditionally acceptable, subject to review under a verification program which requires that valve frequencies below 33 Hz be reconciled with the pipe stress analysis in accordance with United Engineers procedure DEDP-2607.
United Engineers letter SBU-93187 dated March 29, 1985 transmits calculations for the above containment isolation valves which demonstrate reconciliation of correct natural frequencies (less than 33 Hz) with the pipe stress analyses.
(Closed) Unresolved Item 3-3, Interaction Between Support Steel and Pipe This item concerns the effects of coupling between pipe lines supported at a common support.
United Engineers performed a coupled stress analysis which confirmed the structural integrity of a configuration identified by the team as being potentially susceptible to dynamic interaction. Although the pipe support seismic reactions increased by an average of 522%, this increase was still within allowables.
In order to address similar situ-ations where dynamic interaction may exist, Cygna Energy Services performed walkdowns of the large bore piping and identified one additional piping and support steel configuration for evaluation.
The differences in the magni-tudes of the support reactions computed for the coupled and uncoupled stress analyses proved to be negligible for this configuration.
The above effort confirmed that dynamic interaction between large bore pipes and supporting structural steel was properly addressed, and the uncoupled seismic analyses yielded sufficiently conservative support loads and pipe stresses.
4.
Civil / Structural (Closed) Finding 4-6, Tank Farm Building Stiffness (Closed) Finding 4-7, Structural Steel Bracing These items concerned errors in the mathematical model for tank frame seismic loads, which had the potential of misrepresenting the fundamental frequency and consequently affecting the location of peak frequencies as well as the value of acceleration in the amplified response spectra.
United Engineers reanalyzed the tank farm and, as a result, made revisions to amplified response spectra and seismic inertial loads required for design and analysis of. structural steel framework.
This resulted in significant redesign of the structural steel framework.
Previously the tank frame enclosure was designed as Seismic Category 1 for OBE load.
The structure is now designed as Non-Seismic Category 1 for SSE loads, which is consistent with the FSAR.
(Closed) Finding 4-2, Live Loads (Closed) Finding 4-18, Equipment Vault Live Load These items concerned not including live loads (i.e. moveable loads during plant operations) in structural design equations involving seismic loading.
-4-
_
_
_
_
__
-.
.
.
The applicant proposed an addition of a pre-established live load value i
(2 kip concentrated load to be applied on applicable structural elements in a conservative manner) to the seismic design review of Seismic Category 1 floor slabs and beains.
NRR reviewed the basis and the effect of the pro-posed 2 kip concentrated load, and concluded that this loading condition represented all possible moveable live loads that govern the structural design.
The two structural verification programs (the Concrete Verifica-tion Program and the Beam Verification Program),'vhich focus on critical structural elements, were judged by the staff as workable and reliable.
It is understood that design calculations for Category 1 structural com-ponents will be documented in accordance with procedure AP-22.
In conclusion, the staff found that the approach of adding a pre-established live load value (2 kip concentrated load) to the seismic design review of
,
seismic Category 1 floor slabs and beams is acceptable.
(Closed) Finding 4-16, Refueling Water Storage Tank (RWST) Stiffness This issue indicated a concern that shear stiffness was neglected in the calculation of tank stiffness used to determine tank lateral resonant frequency.
A reduction of this frequency may lead to an increased com-
,
pressive stress value which exceeds ASME B&PV Code Sec. III allowable stress
'
for uniform axial loading.
A recalculation by the applicant taking into consideration shear deforma-tion resulted in a horizontal fundamental frequency reduced from 10.303 Hz to 7.478 Hz. However, the calculated maximum stress based on the revised fundamental frequency is much less than the critical stress for the loading condition of combined axial loading and bending by an accepted analytical method (" Structural Analysis of Shells" by E. H. Baker, T5 accordance with et al.).
The applicant has also performed an alternative calculation the NRC staff's recommended approach based on NUREG-CR-1161 and the paper
" Dynamics of Fixed-Base Liquid-Storage Tanks" by A. S. Veletsos and J. Y.
l Yang.
On March 18, 1986, a meeting was held at the Seabrook plant site
'
where the staff audited the above described tank calculations and raised several questions regarding the methodology of water storage tank design.
During a telephone conversation on March 19, 1986, the applicant provided its response to the questions and the staff was convinced that the NRC recommended alternate approach had been properly executed.
The staff concluded that the design of the RWST is acceptable.
5.
Electrical There were no remaining open items in the electrical area.
6.
Instrumentation and Control (Closed) Finding 6-14, PCCW System Failure This item concerned the primary component cooling water system.
Contrary to position C.4 of Regulatory Guide 1.75, Revision 2, analysis had not been conducted to ensure that failure of unqualified loads on associated circuits would not result in degradation of safety related circuits below acceptable levels.
To address this issue as well as its generic implica-tions for potential degradation of IE components due to interactions with non-1E components, United Engineers provided the following documentation:
(1) Calculation No. 9763-3-ED-00-64-F, Revision 0, dated 10/29/85, " Inter-l action Between Associated and Class 1E Circuit for PCCW System." This-5-
.
__ - - _.
-
-.
._
- - -
-
-.
-..
.-
-
_.
-
.. -.-- --.- - -.
.l
}
-
,
.
I confirmed that the maximum credible short circuit current due to failure of i
'
an unqualified current-to pneumatic converter would not exceed ratings of a Class 1E PCCW system selector switch.
The IDI finding concerned the effect l
,
that failure of the switch would have on operation of the heat exchanger outlet and bypass valves.
,
i (2) Calculation No. 9763-5-ES-00-1-F, Revision 0, dated 2/19/86, " Analysis of Effects of Failure of Non-Safety-Related Equipment on Safety-Related i
Equipment." Nineteen systems were reviewed to assess the impact of failure of non-safety related mechanical components upon safety-related systems and upon non-safety-related systems used to support safety-related systems.
Solutions to concerns raised by IDI Findings 6-13 and 6-14 were implemented
!
by DCNs 65/258A and 68/282A.
In other cases the existing design was j
considered satisfactory, t
(3) Calculation No. 9763-3-ED-00-63-F, Revision 1, dated 2/19/86, " Associated Circuit Study of IE-Non IE."
This analysis covered approximately 45 systems and addressed the impact of failure of non-Class 1E electrical components on Class 1E circuits.
United Engineers identified detrimental interfaces
'
in six systems and identified appropriate circuit modifications for their resolution in BER 779A, DCN 63/0109B.
,
,
j (4) CQ-D1200, Revision 1, dated 1/22/85, " Certificate of Qualification for i
Safety-Related Instrumentation." This document demonstrated that Westing-i house 7300 series safety-related Class 1E equipment, furnished for Seabrook, l
met the requirements of IEEE 323-1974, IEEE 344-1975 and UE&C Specification j
9763-006-174-2, Revision 11.
The team had no questions on this document, j
Based on the above documentation, the team concludes that the specific
!
issue raised by the team for the PCCW system and the generic issue of IE
.
and non-1E interaction have been adequately resolved.
l (Closed) Finding 6-15, Temperature Control Circuit Isolation
!
This item concerned the fact that, contrary to United Engineers Specifica-
',;
tion 174-2, isolation devices were not used to isolate non-safety-related
'
instrumentation circuit TYY-2271-2 (for the PCCW heat exchangers) or its associated card frame from safety-related card frames and instrumentation
within panel CP-1528.
United Engineers had not performed an analysis to demonstrate that this situation would not result in safety-related circuits
'
'
being degraded under accident conditions.
,
i Calculation No. 9763-5-ES-00-2, Revision 0, dated 2/11/86, " Degradation of l
CP-1528 by Unqualified Components" identified all non-Class 1E circuits in
CP-152B, and indicated for each (1) whether qualified input and output l
isolators were used; (2) the impact of postulated component failures where
!
qualified isolators were not used; and (3) justification for the absence of
'
I isolators. The calculation notes that DCN 68/282A-BER 748A resulted in the i
upgrading of six non-Class 1E components to Class 1E.
As a result of dis-cussions with the team, the applicant decided to upgrade an additional eight i
l non-Class 1E components (isolator cards) to Class 1E.
These are identified in Yankee memorandum SBP-86-483 dated June 4, 1986.
The team agrees with i
I the calculation's justifications for the absence of isolators.
i
!
I j-6-
!
l
- -
-.-
_
.
.-
- -..-
- - --- - - -
~
.
(Closed) Finding 6-17, Circuit Breaker Fault Current Qualification
,
This item concerned the fact that the E22 and BQ circuit breakers had not been tested for fault current interruption capability during or after a seismic event, despite the fact that these breakers are used in associated circuits in order to prevent non-1E loads from causing harmful effects to class 1E circuits.
The Seabrook SER (page 8-20) indicates that, based on operating experience of protective devices subjected to actual and simulated seismic conditions, it is the staff's judgement that these protective devices will maintain their structural integrity and perform their power removal function when subjected to a seismic event and failure of a load.
This statement applies to the E22 and BQ circuit breakers used in associated circuits.
The appli-cant indicated that, since these breakers have been shown to maintain their integrity during seismic events, there is reasonable assurance they will function under fault current during a seismic event.
For this reason, the team concludes there is no need for seismic qualification of E22 and BQ breakers under fault conditions.
(Closed) Finding 6-30, Conduit Marking Seabrook Class 1E conduits are not marked to identify the separation group at intervals not to exceed 15 feet, as required by IEEE Std. 384 and Regu-latory Guide 1.75.
The FSAR takes exception to this requirement.
Instead conduit is required to be " identified at each end where the conduit termi-nates and at both sides of walls, floors, and in-line boxes." The applicant considers this marking system is sufficient to (1) ensure that cables are pulled into the correct conduit (not one of a different separation group)
and (2) allow inspection to ensure a minimum of one inch separation between conduits of different separation groups.
QC inspections verify that conduit is properly marked and identified and that train separation is correct.
The team concludes that the intent of conduit identification is to ensure that cable is routed into the correct conduit without the need to consult reference material to distinguish between redundant Class 1E raceway systems.
The above method coupled with QC inspection after installation provides this assurance.
.
-7-
,
I