IR 05000324/1993048
| ML20059K662 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 11/05/1993 |
| From: | Casto C, Whitener H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20059K617 | List: |
| References | |
| 50-324-93-48, 50-325-93-48, NUDOCS 9311160138 | |
| Download: ML20059K662 (7) | |
Text
.
p ?si arc %,*
UMITED STATES
.
o NUCLEAR REGULATORY COMMISSION
'
.*
d
'
- 4
- REGloN il
S 101 MARIETTA STREET, N.W., SUITE 2930 a
y ATLANTA, GEORGIA 303234199
'
- %.....j
.
,
Report Nos.: 50-325/93-48 and 50-324/93-48 Licensee: Carolina Power and Light Company P. O. Box 1551
Raleigh, NC 27602 Docket Nos.: 50-325 and 50-324 License Nos.:
,
Facility Name: Brunswick 1 and 2 l
Inspection Conducted: October 4 - 8, 1993 Inspector:
hM
\\/
hv)
d#m I
[h)-
i H. Whitener (/
-
Date Signed
/
/
/
-
Approved by:
h g/5/f3
C. Casto, Chief U
/'/
/Dats Signed Test Programs Section
Engineering Branch Division of Reactor Safety l
SUMMARY Scope:
'
This routine, announced inspection was conducted in the areas of reviewing the licensee's commitments for Unit I restart and closing out inspection followup e
items.
Results:
In the areas inspected, violations or deviations were not identified.
The licensee has completed corrective actions related to several valve leakage
issues and a drywell. head leakage issue. The_ inspector reviewed the licensee's performance and. determined that the extensive, detailed and thorough investigation and repair of manufacturing deficiencies in Anchor Darling double disc gate valves was a strength in the area of Engineering and Technical Support. - Also, Licensee ranagement demonstrated positive oversight
}
in support of the technical resolution of Anchor Darling valve issues.
i i
e
!
,
9311160138 931104 PDR ADOCK 05000324
'
G PDR -
,,
'
,
_
~
'
.
.
)
REPORT DETAILS
'
1.
Persons Contacted
'
Licensee Employeer,
- G. Barnes, Manager h it 1 Operations
- M. Bradley, Manager NAD
- J. Cowan, Plant Manager Unit 1
- J. Crider, ISI/IST Manager
- L. Gard, Manager Engineering Projects
- R. Godley, Manager Regulatory Programs
- C. Hinnant, Director Site Operations
- W. Levis, Manager Regulatory Affairs
- R. Lopriore, Maintenance Manager Unit 1
- J. Lyash, Manager Operations Support
- A. Padleckas, ISI Engineer
,
- C. Robertson, Manager E & RC
- S. Tabor, Senior Specialist, Regulatory Program
- G. Thearling, Senior Specialist Regulatory Program Other licensee employees contacted during this inspection included craftsmen, engineers, operators, mechanics, security force members, technicians, and administrative personnel.
NRC Resident Inspectors
- R. Prevaette, Senior Resident Inspector
- P. Byron, Resident Inspector
- Attended Exit Interview i
Acronyms and initialisms used throughout this report are listed in the
,
last paragraph.
2.
LER l-92-023 The inspector reviewed the results of the licensee's investigation into
'!
the cause of Anchor / Darling (AD) double disc gate valve leak rate failures. Documents reviewed included the Root Cause Investigation of
Anchor / Darling double-disc gate valve Local Leak Rate Test (LLRT)
!
failures for both Unit 1 and 2,-February 26, 1993; Brunswick Information Report, December 27, 1991; LER l-92-023, October 26, 1992 and
Supplements 1 and 2 dated December 14, 1992 and March 31, 1993
'
respectively; and valve repeatability test results. Although the LER l-92-023 addressed specifically failure of Reactor Core Isolation Cooling (RCIC) system inboard and outboard steam isolation valver in-August / September 1992, it also related to an on-going investigation by the licensee of multiple AD double disc gate valve LLRT failures. Due-l to the various manufacturing deficiencies identified in the root.cause
!
investigation, the licensee considered that a 10 CFR 21 report was
.
appropriate. This report, submitted as an attachment to Supplement _1 of-LER 1-92-023, addresses both Unit 1 and Unit 2 valve failures.
Because a valve design study identified a potential for thermal / pressure binding
.. _
.
,
)
,
I the licensee replaced eleven valves in each unit with AD double disc gate valves in the 1989-1991 time frames. These valves are vented so that only the downstream (low pressure side) provides a sealing surface.
A wedging mechanism between the upstream and down stream valve discs acts to force the discs onto the body seais. The eleven valves replaced with AD double disc gate valves in each Unit are as follows:
1/2-B21-F016/F019 Main Steam Drain Valves 1/2-E41-F001 HPCI Turbine Steam Admission Valves
'
1/2-E41-F002/F003 HPCI Steam Line Isolation Valves 1/2-E41-F006 HPCI Injection Valves i
1/2-E51-F007/F008 RCIC Steam Line Isolation Valves
1/2-E51-F013 RCIC Injection Valves 1/2-G31-F001/F004 RWCU Inlet Isolation Valves
,
All but the HPCI Turbine Steam Admission Valves (E41-F001) are leakage barriers in the Primary Containment Isolation System (PCIS).
LER 1-92-023 and Supplements 1 and 2, the 10 CFR P rt 21 report, and Root Cause
,
Investigation documents provide a detailed description of the type of
manufacturing discrepancies identified in these valves and the effect on i
the valve function. These problems and the valves affected are summarized in the Root Cause Investigation Report as follows:
[
Valve Deficiencies Identified
.
1-B21-F016 3,7,8 1-B21-F019
1-E41-F001 4,6,9
-
1-E41-F002 4, 8, 10 1-E41-F003 3, 10 1-E41-F006
1-E51-F007
1-E51-F008 1, 2, 3, 7
!
1-G31-F001
'
2-821-F016
-
2-B21-F019
2-E51-F007
2-E51-F008
!
2-E51-F013
2-G31-F001 1,2,5,8
,
2-G31-F004 2, 3
Deficiencies l
1 - Uneven Stanchion Length on Lower Wedge i
Casting Flaws in Wedge Surfaces
'
-
,
b
.
.
.
.
,
l Non-uniform Contact of the Upper and Lower Wedges l
-
'
4 - Improper Centering of the Valve Stem Improper Wedge Orientation
-
Cracked Stellite on Wedges
-
Sharp Edges on Discs
-
8 - Low Spots in Seats
,
Disc Scraping Body Casting
-
10 - Stellite Seat Face Cracking Items 1, 2, 3, 4, 5, and 8 above have been noted as causes or contributing causes to LLRT failures. The first five deficiencies listed could result in inconsistent seating of the valves and _non-repeatable sealing when subjected to low differential pressures.
To provide confidence in the sealing characteristics, each_ double disc gate valve in PCIS applications was subjected to at least two LLRT's during the current unit outages. The second LLRT is performed to dem.onstrate repeatability.
Corrective actions for the identified problems were diverse and
numerous.
Extensive maintenance and repair of the valves also was
.
required. These actions were described in some detail in the documents mentioned above but can be summarized as action to custom fit valve internals to obtain proper wedging action, seating and even force distribution through the valve. Based on review of corrective actions and discussions with licensee engineering personnel the inspector concluded that corrective actions were appropriate and due to the nature of the problems (manufacturing defects) they should prevent recurrence of valve leakage (except from normal usage and wear causes). The licensee's performance in the AD double disc gate valve issue indicates responsible management oversight and a strength in the area of
.)
Engineering and Technical Support. Attributes considered in this evaluation included:
-
Positive Management Oversight - In response to an industry-problem management directed a valve design review including suitability of valve application for valves with a potential for the thermal / pressure binding as identified in-SOER 84-7.
Management allocated resources to replace the suspect valves with a vented valve design (i.e., AD double disc gate valves). Also, management's commitment of resources in pursuing and resolving the technical issues in the valve leakage problem was positive.
i
-
Good cooperation between corporate engineering (NED) and onsite i
Technical Support engineering and strong involvement in support of plant maintenance - Both corporate and Technical Support engineering were involved in valve examination and analysis of information to resolve the valve leakage problem.
-
Good effort in the use of industry information and experience -
The licensee reviewed Nuclear Plant Reliability Data System, conducted surveys with the Boiling Water Reactor Owner's Group
.
.
.
Valve Group; requested information over the Nuclear Network; made
,
direct contact with several other Boiling Water Reactor licensees; and involved _ vendor representatives in the technical issues.
,
-
Procedure Review - The licensee reviewed all leak rate test procedures for AD valves to ensure that test methodology tested
,
the downstream disc leak tightness.
-
10 CFR Part 21 Report-The licensee issued a Part 21 Report to
'
identify the manufacturing deficiencies.
LER l-92-023 is considered closed.
3.
Action on Previous Inspection Findings (Closed) Unit 1 Startup Item F.1.
In response to NRC Inspection Report 50-325,324/92-12, CP&L submitted a plan on July 23, 1992, for improving plant management and operations (NLS 92-160).
Enclosure 3 of that submittal was a punch list of activities to be completed before Unit restart. The inspector reviewed Item F.1. in a previous inspection (NRC Report 50-325, 324/93-12) and closed the portion involving the feedwater inboard isolation check valve leakage issue and the drywell cooler nitrogen leakage issue. However, the Main Steam Isolation Valve (MSIV), (1-B21-F0228) packing leakage issue remained open to examine the licensee's corrective action. The licensee believed that the cause of the valve packing leak was due to steam galling. Observation of the galled area indicated that galling may have slightly increased in length but not in width or depth since 1987.
Previous repair efforts had only dressed out the high spots of metal on the accessible areas. During.the forced outage the licensee developed a special tool and dressed out the galling in the stuffing box area. With the vendors approval, enough metal was removed to blend out s
the rough areas. The licensee believes that this action will prevent i
future damage to the packing. The packing leak was a startup rather
than a safety problem. The safety consequences of a leak developing is minimal since the leakage is in the containment, and no safety important i
instrumentation or equipment is near enough to be affected. Also, valve stroke time or seat leak tightness are not affected.
'
i The inspector reviewed the completed work orders and concluded that the corrective action was sufficient. This issue is considered resolved.
(Closed) IFI 50-325/92-38-01, Review licensee analysis and corrective action for MSIV 1' B21-F028B and drywell head seal leakage problems.
-
MSIV 1-B21-F028B has a record of failure in meeting the leak rate limit.
.]
During the forced outage the licensee engineers determined that the i
cause of the leakage was the valve stem to actuator shaft misalignment.
The licensee performed certain vendor recommended actions to mitigate-the affects of the misalignment.
These actions included:
-
Installation of longer junk rings to provide additional stem
'
support as it travels through the stuffing box.
.,
.
i
,
-
Oversizing of the bonnet backseat bore to reduce the chance of stem galling.
-
Installation of lighter main disk assemblies to reduce the chance
,
of body galling.
!
These vendor recommended actions are expected to mitigate the effects of
misalignment of stem to actuator shaft and provide improved disk to seat i
contact.
Longer term action, if needed, includes the possibility of j
valve actuator modification. The licensee is evaluating the valve manufacturer's recommended modification which involves bigger and
'
stronger spring guide tubes and stronger springs. Also, a recommended fix by the actuator manufacturer is being evaluated. This fix would involve installing a yoke on the valve to support the actuator.
!
The inspector reviewed the completed work orders and concluded that the corrective action was reasonable. The added support to relieve the misalignment partially resulted in two successful leak rate tests in
,
August 1992 and September 1993. This issue is considered closed.
i A second leakage problem reviewed during this inspection was the root-cause analysis and corrective action involving the drywell head seal
leakage.
In response to IEB 78-09 the licensee specified that a
torquing process was adequate to obtain a leak tight drywell head seal.
Torque values and sequence are specified in the head installation
'
Procedure OSPP-RPV502, " Reactor Vessel (And Associated Components)
Reassembly Following Refueling". This procedure further specified that the seal be marked at 45 degrees increments and inserted in the flange groove at these points first to reduce stretching or compressing the i
seal in any one area. The head seal design is common to both Unit I and i
Unit 2.
The drywell head seal leak rate test failures have occurred in-l recent years on Unit 2 in January 1988 and September 1989 and.on Unit 1 in September 1992.
In 1988 it was concluded that the Unit 2 seal failure occui.ed because the material had not been completely formulated. As a result the seals in stock were hardness tested and inspected for uniformity. When a second unit failure occurred in 1990, the seal was sent to Materials Evaluation - Metallurgical Lab at the Harris Plant for testing and analysis. A breakdown in the gasket material seemed to be occurring.
Extensive tests indicated that temperature, air, and lubrication (Nickel
,
Never - Seez) had acted synergistically to degrade the material. The tests indicated that this degradation of the seal material did not occur
'
if Dow Corning High Vacuum grease was used as a seal lubricant.
However, in 1992 the Unit I drywell head seal failed in a manner that appeared similar to the Unit 2 failure.
Further investigation indicated that portions of the seal appeared to have been pinched between the head flanges and the seal material extruded through these areas. Lab tests were able to duplicate the pinched condition that resulted in a similar i
type of seal failure.
In tests where the seal was not pinched and was restrained in the retaining groove there were no seal failures.
It i
I
~
,
.
.
t appeared that the seal failures may have been a combination of operating
environment and inadequate seal installation.
The licensee has revised Procedure DSPP-RPV502 to require benchmarks at 15 degrees to further reduce the potential to stretch or compress the seal material. The licensee also considered establishing an inspection and physical method, if possible, to ensure the seal is not intruded
between.the flanges and gcod metal-to-metal, 360 degree flange contact i
is achieved during seal installation.
Since lab tests have shown that if the proper lubrication is used and the seal is properly restrained in the flange groove, failure should not occur, the inspector concluded that the licensee's action was reasonable and this item is closed.
r 4.
Exit Interview i
Tt.e inspection scope and results were summarized on October 8, 1993,
,
with those persons indicated in paragraph 1.
The inspector described
the areas inspected and discussed in detail the inspection results.
Proprietary information is not contained in this report. Dissenting comments were not received from the licensee.
,
.
5.
Acronyms and Initialisms Ar
-
CP8
-
Carolina Power and Light HPCI
-
High Pressure Coolant Injection
IEB
-
Inspection and Enforcement Bulletin LER
-
Licensee Event Report LLRT -
Local Leak Rate Test MSIV -
Main Steam Isolation Valve NED
-
Nuclear Engineering Department
,
PCIS -
Primary Containment Isolation System RCIC -
Reactor Core Isolation Cooling RWCU -
Reactor Water Clean Up
!
!
!
I i
l
i
)
f c