ML14181A973
| ML14181A973 | |
| Person / Time | |
|---|---|
| Site: | Robinson  |
| Issue date: | 12/22/1997 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML14181A971 | List: |
| References | |
| 50-261-97-12, NUDOCS 9801210300 | |
| Download: ML14181A973 (34) | |
See also: IR 05000261/1997012
Text
U.S. NUCLEAR REGULATORY COMMISSION
REGION II
Docket Nos:
50-261
License Nos:
Report No:
50-261/97-12
Licensee:
Carolina Power & Light (CP&L)
Facility:
H. B. Robinson Unit 2
Location:
3581 West Entrance Road
Hartsville, SC 29550
Dates:
October 12 - November 22, 1997
Inspectors:
B. Desai, Senior Resident Inspector
J. Zeiler, Resident Inspector
E. Girard, Region II Inspector (Sections
E1.1, E1.2, and E7.1)
H. Whitener. Region II Inspector (Section M1.2)
Accompanying Personnel: T. Scarborough, Nuclear Reactor Regulation
Approved by:
M. Shymlock, Chief, Projects Branch 4
Division of Reactor Projects
9801210300 971222
Enclosure 2
ADOCK 05000261
G
EXECUTIVE SUMMARY
H. B. Robinson Power Plant, Unit 2
NRC Inspection Report 50-261/97-12
This integrated inspection included aspects of licensee operations,
maintenance, engineering, and plant support. The report covers a six-week
period of resident inspection; in addition, it includes the results of
inspections by two Region II based reactor safety inspectors.
Operations
The conduct of operations was professional and safety-conscious.
(Section 01.1).
Overall, operator response to the failure of a condensate pump shaft and
subsequent automatic reactor trip was appropriate. Plant shutdown and
startup activities were satisfactorily conducted. Ineffective
corrective actions from a similar condensate pump shaft failure that
occurred in 1991 resulted in the subsequent failure on November 16, 1997
(Section 01.2).
Two Non-Cited Violations (NCVs) were identified involving operator
failure to follow procedures while implementing Technical Specification
(TS) surveillance requirements. While individually, each of these items
had low safety consequence, they indicated weaknesses in operator log
keeping accuracy and operator attention to detail (Section 01.3).
Overall, licensee implementation of cold weather protection activities
was satisfactory and the system engineer was familiar with plant cold
weather protection equipment. Additionally, a Nuclear Assessment
Section (NAS) audit of the plant cold weather protection activities was
thorough, and condition reports were appropriately initiated to address
several deficiencies that were identified by the NAS audit (Section
02.1).
The licensee's implementation and transition to the Improved Technical
Specifications was good (Section 06.1).
The onsite review functions of the Plant Nuclear Safety Committee (PNSC)
were conducted in accordance with TS. The PNSC meetings were well
coordinated and meeting topics were thoroughly discussed and evaluated.
NAS continued to provide strong oversight of licensee activities. A NAS
assessment of the Robinson Motor-Operated Valve program was thorough and
resulted in important findings (Section 07.1 and E7.1).
The failure to properly implement Regulatory Guide commitments as
described in the Updated Final Safety Analysis Report (UFSAR) associated
with the Loose Parts Monitoring Program was considered similar to
deficiencies and inconsistencies in the UFSAR and other licensing
documents previously identified during the NRC's Architectural
Engineering (A/E) Team Inspection (Section 08.1).
2
Maintenance
In general, routine maintenance activities were performed satisfactorily
(Section M1.1).
Emergency diesel generator (EDG) maintenance management indicated
advanced planning and careful attention to detail.
EDG post-maintenance
testing was performed in a thorough and professional manner. The
licensee thoroughly researched potential sources of vibration
experienced on the "B" EDG and planned further evaluations during an
upcoming planned diesel maintenance outage. The "B" EDG vibration data
exceeded the in-service limit but was well below the vendor recommended
shutdown limit (Section M1.2).
Engineering
An Inspector Followup Item (IFI) was opened related to an event
involving the lifting of a relief valve on the Safety Injection (SI)
discharge piping during an Inservice (ISI) test. The licensee's
evaluation of the condition concluded that it did not adversely impact
the SI system nor did it impact accident dose. The inspector concluded
that the condition did not significantly compromise safety; however,
several issues are still under review (Section E1.1).
The licensee had not met its original commitment date for correcting
adverse findings regarding its implementation of Generic Letter (GL) 89-10 identified during NRC Inspection 50-261/96-12. Progress toward
correcting the findings by the new date proposed by the licensee was
generally satisfactory. Some concerns were identified. These concerns
and the outstanding findings from NRC Inspection 50-261/96-12 will be
re-examined in a future inspection which will assess the licensee's
completion of GL 89-10 implementation (Section E1.2).
A Violation was identified for the failure to adequately control post
modification testing related to ESR 9500783. The post-modification
testing failed to confirm that the cooling performance of the
Containment Air Recirculation Cooling system was not adversely impacted,
and that the system was capable of meeting its intended safety function
as described in Section 9.4.3.1 of the licensee's UFSAR (Section E8.3).
Plant Support
The inspectors concluded that radiation control and security practices
were proper (Section R1.1 and S1.1).
Report Details
Summary of Plant Status
Robinson Unit 2 began the inspection period operating at full power. The unit
tripped from 100 percent power on November 16, 1997, due to mechanical failure
of the "B" Condensate Pump shaft. The condensate pump was repaired and the
unit went critical on November 19. Synchronization to the grid occurred on
November 20. The unit reached essentially full power operation the next day
and continued at essentially full power for the remainder of the inspection
period. Up until the day of the trip, the unit had operated at power for 390
continuous days.
I. Operations
01
Conduct of Operations
01.1 General Comments (71707)
The inspectors conducted frequent control room tours to verify proper
staffing, operator attentiveness and communications, and adherence to
approved procedures. The inspectors attended daily operation turnovers,
management reviews, and plan-of-the-day meetings to maintain awareness
of overall plant operations. Operator logs were reviewed to verify
operational safety and compliance with Technical Specifications (TSs).
Instrumentation, computer indications, and safety system lineups were
periodically reviewed from the Control Room to assess operability.
Frequent plant tours were conducted to observe equipment status and
housekeeping. Condition Reports (CRs) were routinely reviewed to assure
that potential safety concerns and equipment problems were reported and
resolved.
In general, the conduct of operations was professional and safety
conscious; specific events and noteworthy observations are detailed in
the sections below.
01.2 Reactor Trip and Subsequent Unit Startup
a. Inspection Scope (71707 and 93702)
On November 16, 1997, at approximately 1:33 a.m., Robinson Unit 2
experienced an automatic reactor trip. The unit was operating at full
power prior to the transient. The Senior Resident Inspector was
notified and responded to the site. A 10 CFR 50.72 (b)(2)(ii)
notification was appropriately made to the NRC. The inspector reviewed
the circumstances associated with the reactor trip, as well as the
subsequent plant startup activities.
b. Observations and Findings
Robinson Unit 2 experienced an automatic reactor trip due to low (26
percent) Steam Generator (S/G) level coincident with Steam
Flow/Feedwater flow mismatch. All control rods inserted into the core
and electrical power transferred to the startup transformer without any
2
problems. The transient was initiated when the "B" Condensate Pumps'
stub shaft sheared. The condensate pumps are deep draft centrifugal
pumps and have a long shaft connecting the motor to the pump. The stub
shaft is a coupling that connects the long shaft to the pump shaft.
S/G levels trended down due to the loss of the condensate pump. The
control room entered Abnormal Operating Procedure (AOP) 010, Main
Feedwater/ Condensate Malfunction, upon noticing decreasing S/G levels.
The "B" Condensate Pump motor continued to run and the stub shaft shear
was initially unknown to the operators since there is no motor ammeter
or individual condensate flow/pressure indication in the control room.
Due to the ongoing transient, the operator initiated a Turbine runback.
However, S/G level continued to decrease and the runback caused a
Reactor Coolant System (RCS) Temperature (T)-average/T-reference
mismatch. This mismatch caused the condenser dump valves to open,
causing steam flow to be greater than feedwater flow, resulting in a
Plant post trip response was normal with some exceptions. Source Range
(SR) Nuclear Instruments (NI) 31 and NI 32 did not automatically
energize at permissive P-6. Consequently, NIs 31 and 32 had to be
manually energized. Additionally, Intermediate Range (IR) instruments
NI 35 was noted to have failed and IR NI 36 was noted to be slightly
under compensated. P-6 is actuated when both IR NIs 35 and 36 reach 1E
10 amperes. With the failure of NI 35, P-6 was not received causing
failure of NIs 31 and 32 to automatically energize. The problem with NI
36 was later attributed to a loose power supply cable connection, and NI
36 was returned to operable status following corrective maintenance.
The problem with NI 35 was identified to be a failed detector. The
detector was replaced and NI 35 was also returned to service prior to
reactor startup.
Initially, the control room was not aware of the stub shaft failure.
However, during the investigation after the reactor trip, an abnormal
noise was heard at the pump. The motor was noted to be running;
however, the local ammeter reading as well as pump discharge pressure
were noted to be abnormally low, indicative of a shaft failure. The
condensate pump was subsequently disassembled and the stub shaft failure
was confirmed. The licensee replaced the "B" pump with a spare pump and
installed a new designed stub shaft.
The inspector examined the failed stub shaft and reviewed the licensee's
preliminary investigation into the cause of the shaft failure. The stub
shaft had sheared in the vicinity of the split ring grove at the top
portion of the keyway grove. This failure was considered similar to the
one that occurred in September 1991 on the same pump. The 1991 failure
was attributed to a faulty design of the stub shaft keyway area. The
keyway was machined into the split ring grove area and contained sharp
corners that created stress riser locations. In 1991, following
discussions with the pump manufacturer, it was discovered that a new
stub shaft keyway design existed that eliminated the stress riser
problem. The licensee ordered three stub shafts with the new keyway
design with the intent to replace each of the stub shafts on the three
3
condensate pumps. In the interim, an old design stub shaft (with slight
modifications to the keyway area) was installed in the "B" Condensate
Pump following the initial failure.
In December 1991, a refurbished pump was installed in the "B" Condensate
Pump, however, a stub shaft of the new design was not installed, and
instead, the old stub shaft was cleaned and re-installed. The licensee
attributed the failure to replace the stub shaft with the new design,
due to poor wording on the work ticket and lack of effective engineering
coordination. During the September 1996 refueling outage, the "B"
Condensate Pump was again removed and replaced with a refurbished pump
as part of preventative maintenance; however, again, the old stub shaft
was reused. The inspector determined that ineffective corrective
actions from the previous stub shaft failure incident in 1991
contributed to the subsequent stub shaft failure on November 16, 1997.
Following completion of a post trip review, which addressed the cause of
the trip and corrective actions, the licensee initiated plant startup on
November 19, 1997 in accordance with General Procedure (GP)-003, Normal
Plant Startup From Hot Shutdown to Critical.
The plant was placed on
line on November 20, 1997.
The inspector reviewed plant logs and post trip report, discussed the
event with key licensee personnel, and walked down plant equipment,
including maintenance activities associated with the condensate pump
replacement. The inspector also monitored plant startup activities.
c. Conclusions
The inspector concluded that overall, operator response to the transient
was appropriate. Plant shutdown and startup activities were
satisfactorily conducted. With the regard to the condensate pump, the
inspector determined that ineffective corrective actions from a similar
stub shaft failure that occurred in 1991 contributed to resulted the
subsequent failure and reactor trip on November 16, 1997.
01.3 Failure to Follow Procedures Implementing TS Surveillance Requirements
a. Inspection Scope (71707)
The inspector reviewed the circumstances surrounding two examples where
operations personnel failed to follow procedures related to implementing
TS surveillance requirements during the startup from the November 16,
1997, reactor trip.
b. Observations and Findings
1. Failure to Log TS Surveillance for Inoperable Control Rod Insertion
Limit Monitor
On November 18, the operators observed that the control rod insertion
limit monitor (RILM) recorder, TR-409, was not printing all of the rod
4
insertion limits. As a result of this condition, the RILM was declared
out-of-service. On November 19, 1997, after entering Mode 2, operations
recognized the need to perform TS Surveillance Requirement (SR) 3.1.6.2,
which is required when the RILM is considered inoperable. This
surveillance required each control bank insertion to be verified within
its limit as specified in the Core Operating Limits Report (COLR) every
four hours.
On November 20, 1997, during a review of the Reactor Operator (RO)
narrative log, the inspector noted that the first four hour surveillance
had been logged as complete at 7:35 p.m. on November 19. The next log
entry documenting the surveillance completion was at 3:32 a.m., on
November 20. The inspector noted that there was no log entry indicating
that the surveillance had been performed as required at 11:30 p.m. on
November 19. The licensee initiated CR 97-02328 to address the
discrepancy.
The licensee's preliminary investigation indicated that the operators
had recalled performing the four hour surveillance at 11:30 p.m.
However, the operators failed to properly log completion of the
surveillance in their log. The inspector discussed the impact of the
original problem with the RILM recorder with the system engineer and
agreed with the licensee's conclusion that this problem would not have
impacted the RILM's alarm function. Therefore, although the monitor had
been declared inoperable, it would have still been capable of alerting
the operators to an insertion limit problem during this time period.
The inspector reviewed Operations Surveillance Test (OST)-020, Shiftly
Surveillances, Revision 2, which provides instructions for performing TS
required shiftly surveillances. OST-020 requires that when the RILM is
inoperable, the control rod bank insertion limit be verified within the
limits specified in the COLR once within four hours and every four hour
thereafter, and the performance of this activity entered into the RO's
narrative log. The inspector determined that the operators failed to
follow OST-020. Prior to this incident, the inspector's had noted an
increase in errors and inconsistencies in operator logs which indicated
a need for improvement in attention to detail and log keeping accuracy.
The inspector determined that increased management attention to the
quality of log keeping was warranted.
TS 5.4, Procedures, requires in part that written procedures be
established, and maintained covering the activities recommended in
Appendix A of Regulatory Guide 1.33, Revision 2, dated February 1978,
including procedures for conducting surveillances listed in the TS. The
inspector determined that the failure to follow OST-020 was a violation
of TS 5.4. This failure constitutes a violation of minor significance
and is being treated as a Non-Cited Violation (NCV), consistent with
Section IV of the NRC Enforcement Policy. This issue is documented as
NCV 50-261/97-12-01:
Failure to Log TS Surveillance Completion in
Accordance with OST-020.
5
2. Missed TS Surveillance of Reactor Coolant Dose Equivalent Iodine-131
On November 20, 1997, at 5:38 p.m., the RO noticed that reactor coolant
dose equivalent Iodine (I)-131 specific activity had not been verified
as required by TS SR 3.4.16.2. SR 3.4.16.2 requires that within six
hours after a thermal power change of greater than 15 percent power in a
one hour period, reactor coolant dose equivalent 1-131 be verified less
than or equal to 1 microcurie per gram. Reactor power was increased
greater than 15 percent between 10:00 a.m. and 11:00 a.m. on
November 20, requiring SR 3.4.16.2 to be completed by 5:00 p.m. Upon
identifying the missed surveillance, immediate operator actions were
implemented to request Environmental and Radiation Control (E&RC)
personnel to sample the Reactor Coolant System (RCS).
The sample was
taken at 6:05 p.m. and the results obtained at 8:18 p.m. The results
indicated that the reactor coolant dose equivalent 1-131 specific
activity was much less than the TS limits.
The inspector reviewed GP-005, Power Operation, Revision 53, which
provides instructions to permit normal plant startup from Mode 2 to
Mode 1 at 100 percent power. Section 5.4.33 of GP-005 requires the
operators to contact E&RC personnel following any load change greater
than or equal to 15 percent power in order for dose equivalent 1-131 to
be verified with six hours. The inspector determined that operations
personnel failed to follow GP-005 resulting in SR 3.4.16.2 not being
performed within the allowed timeframe.
The licensee initiated CR 97-02342 to investigate the missed
surveillance requirement. The licensee determined that the surveillance
was missed due to lack of operator attention to detail and inadequate
supervisory oversight during periods of high control room activities.
The inspector noted that one of the significant contributing factors
identified was a weakness in GP-005's control of the activity. The step
for performing the surveillance was placed at the point where reactor
power would have just reached 15 percent power, as opposed to being in a
continuous monitoring section of the procedure, such as the precautions
and limitations. Once the step was signed off at 15 percent power,
there was no further mention of the surveillance requirement during
subsequent power ascension activities. The licensee planned to revise
GP-005 to ensure that the surveillance requirement was reiterated in the
precautions and limitations section of the procedure.
TS 5.4, Procedures, requires in part that written procedures be
established, and maintained covering the activities recommended in
Appendix A of Regulatory Guide 1.33, Revision 2, dated February 1978,
including procedures for conducting surveillances listed in the TS. The
inspector determined that the failure to follow GP-005 was a violation
of TS 5.4. This non-repetitive, licensee-identified and corrected
violation is being treated as a NCV, consistent with Section VII.B.1 of
the NRC Enforcement Policy. This issue is documented as NCV 50-261/
97-12-02:
Failure to Verify Dose Equivalent 1-131 in Accordance with
GP-005.
6
c. Conclusions
Two NCVs were identified involving operator failure to follow procedures
for implementing TS surveillance requirements. While individually, each
of these items had low safety consequence, they indicated weaknesses in
operator log keeping accuracy and operator attention to detail,
warranting licensee attention.
02
Operational Status of Facilities and Equipment
02.1 Cold Weather Preparations
a. Inspection Scope (71707)(71714)
The inspector assessed licensee preparations and contingencies for cold
weather conditions, as well as performed a walkdown of the related
freeze protection panels and temporary cold weather protection measures.
b. Observations and Findings
The inspector reviewed Administrative Procedure (AP)-008, Cold Weather
Preparations, Operating Procedure (OP)-925, Cold Weather Operation, and
Electrical Distribution Procedure (EDP)-009, Freeze Protection Panels,
to assess licensee preparations and contingencies associates with cold
weather, and existing CRs associated with cold weather protection
problems. The inspector also performed a walkdown with the system
engineer of freeze protection panels and other selected components,
including: the service water intake area, deepwell pumps, installed
temporary heaters and enclosures, and temperature sensing instruments
that control the thermostats associated with the freeze protection
circuits. Additionally, the inspector reviewed existing CRs related to
cold weather preparations and a program audit that was performed by the
Nuclear Assessment Section (NAS).
c. Conclusions
The inspector concluded that the system engineer was familiar with the
system, the NAS audit was thorough, and overall licensee implementation
of the cold weather protection measures was satisfactory. The licensee
appropriately generated condition reports to address several
deficiencies that were identified by the NAS audit.
06
Operations Organization and Administration
06.1 Improved Technical Specification Implementation
a. Inspection Scope (71707)
The inspector reviewed licensee activities associated with the
implementation of Improved Technical Specifications (ITS).
The ITS was
implemented at Robinson on November 13, 1997, following issuance of
License Amendment No. 176.
7
b. Observations and Findings
Concurrent with the ITS, the licensee also implemented the ITS Bases, as
well as the Technical Requirements Manual(TRM), through Plant Program
Procedure (PLP)-100. PLP-100 incorporated requirements from the
previous Technical Specifications that were no longer part of ITS. The
inspector observed licensee utilization of the ITS during evolving plant
issues, including the recent reactor trip and subsequent startup
activities. The inspector observed that, overall, the licensee was
familiar with the requirements of ITS, and as necessary, sought
appropriate guidance, in view of the change in the overall format from
the previous TS. The licensee also stationed, in the control room, an
individual who had participated in the development of ITS. This allowed
the operating staff to ask questions as they related to the ITS to this
individual, as necessary. Additionally, the licensee canceled all
existing TS interpretations.
c. Conclusions
The inspector concluded that licensee implementation, as well as
transition to ITS, was good. The licensee did a good job of
implementing the new ITS surveillance requirements following a reactor
trip only three days after putting them in place.
07
Quality Assurance In
Operations
07.1 Plant Nuclear Safety Committee and Nuclear Assessment Section Oversight
a. Inspection Scope (40500)
The inspector evaluated certain activities of the Plant Nuclear Safety
Committee (PNSC) and NAS to determine whether the onsite review
functions were conducted in accordance with TS and other regulatory
requirements.
b. Observations and Findings
The inspector periodically attended PNSC meetings during the report
period. The presentations were thorough and the presenters readily
responded to all questions. The committee members asked probing
questions and were well prepared. The committee members displayed a
good understanding of the issues and their potential risks.
Further,
the inspector reviewed NAS audits and concluded that they were
appropriately focused to identify and enhance safety.
c. Conclusions
The inspector concluded that the onsite review functions of the PNSC
were conducted in accordance with TSs. The PNSC meetings attended by
the inspector were well coordinated and meetings topics were thoroughly
discussed and evaluated. NAS continued to provide strong oversight of
licensee activities.
8
08
Miscellaneous Operations Issues (92901)
08.1 (Closed) Unresolved Item (URI) 50-261/97-10-01, Complete Review of LPMS
Testing and Maintenance Activities:
This URI involved the completion of
a review of the licensee's Loose Parts Monitoring System (LPMS) program.
The inspector previously identified the potential that the licensee's
LPMS program was not implemented in accordance with their commitments
associated with Regulatory Guide (RG) 1.133, Loose Parts Detection
Program for the Primary System of Light Water Cooled Reactors,
Revision 1. Specifically, Section C.3 of RG 1.133 recommended the
performance of 31-day channel functional tests and 92-day background
noise and channel false signal checks. In addition, the inspector noted
from review of the LPMS vendor manual that the backup battery, which
provides display and memory power, was recommended to be replaced every
six months; however, the licensee had implemented an 18-month
replacement frequency. At the end of the previous inspection period,
the licensee had not demonstrated that these tests were being performed
or had justified the impact of the increased battery replacement
schedule.
On November 20, 1997, the licensee initiated CR 97-2331 to address the
inspector's concerns regarding the testing discrepancies. In addition,
the licensee initiated an engineering self-assessment of the LPMS
testing program. The results of the self-assessment confirmed that the
above mentioned testing was not being performed as recommended by RG
1.133. Licensee corrective actions included plans to revise LPMS
procedures to incorporate the periodic channel functional testing and
background noise/false signal checks. At the end of this report period,
the licensee was still evaluating whether the RG recommended test
frequencies were appropriate. Following resolution of the test
frequencies, the licensee planned to update UFSAR Section 1.8 to reflect
any difference in their position from the recommendations in RG 1.133.
With regards to the backup battery replacement frequency, the licensee
contacted the LPMS vendor supplier and determined that the 18-month
frequency was adequate to ensure that the equipment power requirements
would not be compromised. In addition, the vendor indicated that the
system was designed to alarm on low backup battery power, which would
alert the licensee to a low battery power condition. The inspector
determined that the licensee had adequately addressed the inspector's
concerns associated with the LPMS testing and maintenance.
Section 1.8 of the licensee's Updated Final Safety Analysis Report
(UFSAR) indicated that Robinson was in full compliance with RG 1.133
with the exception of two items. These two exceptions were unrelated to
the recommended testing associated with the LPMS. The inspector
determined that the licensee had not fully implemented LPMS testing in
accordance with their UFSAR commitments. Furthermore, this problem
raised questions regarding the adequacy in the implementation of other
UFSAR commitments related to RGs. During the NRC Architectural
Engineering (A/E) Team Inspection conducted in May 1997, similar
inconsistencies and deficiencies were identified in the UFSAR. In the
licensee's letter dated November 3, 1997, replying to the results of the
9
A/E Team Inspection, the licensee committed to review all of their RG
positions included in the UFSAR to ensure proper implementation. The
inspector determined that, in view of the weaknesses already identified
in the licensee's RG implementation, and proposed licensee corrective
actions to address this problem, no further enforcement action was
warranted for this issue. The adequacy of the licensee's corrective
actions will be reviewed during the closeout review of the A/E Team
Inspection findings. This URI is closed.
II.
Maintenance
M1
Conduct of Maintenance
M1.1 General Comments
a. Inspection Scope (61726 and 62707)
The inspector reviewed/observed all or portions of the following
maintenance related work requests/job orders (WRs/JOs) and/or
surveillances and reviewed the associated documentation:
WR/JO 97-AEYU1
Perform Vertical Drive Inspection on "B"
Emergency Diesel Generator (EDG)
WR/JO AAPV-002
Inspections and Overhaul Maintenance on "B" EDG
WR/JO AAKC-007
Inspections and Overhaul Maintenance on "A" EDG
OST-409-2
EDG "B" Fast Speed Start
b. Observations and Findings
The inspector observed that these activities were performed by personnel
who were experienced and knowledgeable of their assigned tasks.
Procedures were present at the work location and being followed.
Procedures provided sufficient detail and guidance for the intended
activities. Activities were properly authorized and coordinated with
operations prior to starting. Test and maintenance equipment in use was
calibrated, procedure prerequisites were met, and system restoration was
completed. The inspector frequently observed supervisors and system
engineers monitoring the "A" and "B" EDG maintenance activities, and
good support was noted from the EDG vendor.
c. Conclusions
The inspector concluded that routine and corrective maintenance
activities, as well as surveillances, were performed satisfactorily.
M1.2 EDG Vibration Investigation and 18 Month Inspection
a. Inspection Scope (62700, 62707 and 92902)
The inspector reviewed licensee activities to monitor, trend and
investigate a diesel generator vibration problem. Additionally, the
inspectors reviewed the licensee's methods for control of a complex
10
project, 18 month EDG surveillance, and observed portions of an
overspeed trip test.
b. Observations and Finding
The licensee removed the "B"
EDG from service on October 20, 1997 to
investigate vibrations and to conduct an 18 month surveillance. The
Robinson EDGs are Fairbanks Morse opposed piston engines. The method
used by the licensee to control the EDG maintenance activities was to
develop a detailed maintenance schedule. One hundred and seven specific
tasks were identified and scheduled by group, day, and hour. Each task
was implemented by either the PM route or a work package. Completion of
the tasks was tracked by the automatic maintenance management system and
work status was reviewed daily at management and plan of the day
meetings. The inspector considered that the project was well planned
and careful attention to detail was noted.
The inspector discussed the condition of EDG vibrations with licensee
personnel. Actions taken during this diesel testing period included
taking data to develop actions for the planned EDG outage in the
January-February 1998 time frame to specifically address the vibration
problem. Additionally, licensee work activities included checking the
upper and lower thrust bearings for free play, checking the total run
out on the engine to generator coupling for unbalanced mass, and
checking the bolt torque and delivery stroke of the fuel injection pumps
to ensure that no unbalanced firing was occurring. In addition, four
injection pumps were replaced.
From the data taken, the licensee determined that the vibration appeared
to be load rather than speed related. Several areas were identified as
potential sources of vibration to be examined at the planned EDG outage
in early 1998. These included the vertical drive spring pack, torsional
dampers on the crank shaft, alignment of engine to generator coupling
and the generator outboard bearing.
The inspector observed portions of the overspeed trip test performed as
a part of OP-604, Diesel Generators "A"
and "B," Revision 40, during
post maintenance diesel start-up. The pre-job briefing was adequate and
the crew was knowledgeable and adequate resources were available.
Maintenance, engineering, and operator responsibilities were reviewed.
The engine was inspected for leaks or problems as the speed was
increased. The licensee stated that data taken during the post
maintenance run indicated that there was a slight improvement in the
vibration level.
The inspector reviewed the vibration trended data from October 1996 to
October 1997. The data are taken monthly at thirteen locations. There
was an increase since June 1997 in vibration levels at several locations
which exceeded the in-service limit. However, these data were well
below the vendors recommended shutdown limit. Unless there are
significant changes in the increase of vibrations, the shutdown limits
are not expected to be exceeded before the planned diesel maintenance
outage.
c. Conclusions
EDG maintenance management indicated advanced planning and careful
attention to detail. The licensee thoroughly researched potential
sources of vibration to be evaluated at the January-February 1998
planned diesel maintenance outage. EDG post-maintenance testing was
performed in a thorough and professional manner. The "B" EDG vibration
data exceeded the in-service limit but was well below the vendor
recommended shutdown limit.
III. Engineering
El
Conduct of Engineering
E1.1 Pressure Testing of Safety Injection Pump Discharge Piping
a. Inspection Scope (37551)
The inspector reviewed an event where relief valve SI-857A, located on
the Safety Injection (SI) discharge piping lifted during the performance
of an Inservice Inspection (ISI) pressure test.
b. Inspection Findings
On November 12, 1997, the licensee performed Engineering Service Test
(EST) procedure, EST-078, ISI Pressure Testing of Safety Injection Pump
Discharge Piping. This ISI was performed in conjunction with Operations
Test Procedure (OST) 151-1, Safety Injection System Components Test Pump
"A". During the performance of EST-078, the SI system was placed in
service in accordance with procedure OST-151-1 and the system was
visually examined for leakage within the boundaries established by
procedure EST-078. This boundary including piping, valves, and
instrumentation taps, from the SI pump discharge up to the containment
penetration. During the performance of the ISI, a Quality Control (QC)
inspector noted a pencil size stream of water discharging from the tail
pipe of relief valve SI-857A. This leak was estimated to be
approximately one gallon per minute (gpm) and it was terminated when the
piping was depressurized following completion of the EST.
Relief valve SI-857A is physically located in the Boron Injection Tank
(BIT) room in the auxiliary building. The SI cold leg discharge piping
downstream of the BIT includes a single line with relief valve SI-857A.
Then the line splits into two, and includes two normally closed MOVs
SI-870A and SI-870B. The line downstream of SI-870A and SI-870B combine
and then splits into three lines, which include normally open valves
868A, 868B, and 868C. The three SI cold leg injection lines then enter
containment, and include a relief valve SI-857B, which reliefs to the
Pressurizer Relief Tank (PRT).
Each of the three SI injection lines
then have two check valves in series (873A and 873D, 873B and 873E, and
12
873C and 873F).
The three SI lines then tie into the three
Accumulator/Residual Heat Removal (RHR) injection lines. These three
lines, each have a check valve (875A, 875B. and 875C), prior to the
three lines tying to the RCS cold leg loops. The SI piping downstream
of valves 868A, 868B, and 868C is rated as class 2500 (capable of
withstanding full RCS pressure) and the piping upstream of valves 868A,
868B, and 868C is rated as class 1501 (capable of withstanding 1750 psig
at 300 degrees F and 1900 psig at 150 degrees F).
Relief valves SI-857A
and 857B are set to relieve at 1767 +/- 51 psi.
During the performance of EST-078, the A SI pump was run discharging at
a pressure of approximately 1500 psig. When valves SI-870A and SI-870B
were opened to allow pressurization of downstream SI piping, relief
valve SI-857A lifted. The licensee considered that the RCS check valves
868A, 868B, and 868C leaked into the pipe upstream of valves 870A and
870B. With the relief setpoint for SI-857A slightly lower then SI-857B
(within the allowed tolerance of +/- 51 psi), relief valve SI-857A
lifted to relieve pressure buildup.
Engineering Service Request (ESR) 9700594 was initiated to address
potential adverse issues related to this problem. First, the ESR
assessed the implications of relief valve (857A) lifting during an
accident and the subsequent radiological release. Specifically, during
a small break Loss of Coolant Accident (LOCA), where valves SI-870A and
SI-870B would open on an SI signal and the SI pump would start; however,
due to high RCS pressure, SI discharge pressure would be below RCS
pressure. During this time, relief valve 857A would be subject to
similar pressure conditions from backleakage through the check valves,
causing the relief valve to lift. The licensee concluded that the relief
valve leakage would not present a significant radiological consequence,
and under worst case source term assumptions, was still bounded by
existing analysis.
The licensee also evaluated the potential degradation of SI flow in view
of the potential for relief valves 857A and 857B lifting following a
small break LOCA. Further, the leakage from 857A would be outside
containment, and therefore unavailable during recirculation flow. The
relief valves are designed with a 3/4 inch inlet and a 1 inch outlet.
An internal nozzle further restricts the flow to that of a 1/4 inch
diameter pipe. The licensee concluded that existing calculation RNP
M/MECH-1556, Attachment G, demonstrated acceptable system performance
with losses from an open SI test line, and it effectively bounded the
above postulated condition. The inspector had not reviewed the
calculation as of the end of this report period.
The licensee evaluated the acceptability of class 1501 piping upstream
of normally open valves 868A, 868B, and 868C. The licensee concluded
that the lift setpoint of 1750 psig to 1900 psig for a temperature range
of 300 degrees to 150 degrees F adequately afforded the design limits of
the piping from being exceeded. The inspector had not completed the
0s
review associated with this aspect of the evaluation.
13
Additionally, the inspector is reviewing several additional related
issues. These include: the differential pressure testing and
requirements associated with valves SI-870A and SI-870B, the
acceptability of a single isolation valve to afford protection against a
potential leak outside containment, and the check valve backleakage test
methodology and acceptance criteria. Pending completion of the reviews
associated with this problem, the item is identified as Inspector
Followup Item (IFI) 50-261/97-12-03: Review Safety Injection Discharge
Piping Configuration.
The licensee has placed caution tags on the three SI pumps' control
switches in the control room. Periodic testing of SI-870A and SI-870B
requires them to be opened. This presents the potential for lifting
relief valve 857A. The caution tags allows the operator to be cognizant
of this potential and to place an observer in the BIT room or to take
other appropriate actions, including entry in appropriate TS action
statement, as necessary. The licensee is also considering additional
actions through the resolution of the CR system.
c. Conclusions
The inspector opened an IFI related to an event involving lifting of a
relief valve on the SI discharge piping during an ISI test. Licensee
evaluated the condition and concluded that it did not adversely impact
the SI system nor did it impact accident dose. The inspector concluded
that the condition did not significantly compromise safety; however,
several issues still are under review.
E1.2 Implementation of Generic Letter (GL) 89-10,"Safety-Related Motor
Operated Valve Testing and Surveillance"
a. Inspection Scope (Temporary Instruction 2515/109)
The licensee's implementation of GL 89-10, was previously reviewed and
determined inadequate during a previous inspection as documented in NRC
Inspection Report 50-261/96-12. Two violations were identified (see
E8.1 and E8.2) which addressed the principal deficiencies in the
licensee's implementation of GL 89-10. The licensee responded to the
violations in letters dated January 15 and February 28, 1997, and stated
that full compliance with requirements would be achieved by
September 29, 1997. In addition, the licensee indicated that a self
assessment would be performed by that date to confirm adequate
implementation of GL 89-10 and readiness for an NRC closure inspection.
In a letter dated June 30, 1997, the licensee changed the completion
date for evaluations and calculation revisions being performed as part
of its corrective actions, but did not change the date for full
compliance with requirements. Subsequently, in a telephone call on
September 24 and a letter dated September 26, 1997, the licensee
informed the NRC that it would not be ready for the closure inspection
due to issues identified in its self-assessment. The letter stated that
Robinson would be ready for a GL 89-10 closure inspection by
February 16, 1998.
14
The current inspection assessed the procedures, calculations, and
evaluations which the licensee was revising or preparing to address the
adverse findings of Inspection 50-261/96-12 and complete implementation
of GL 89-10. It included a review of the self-assessment referred to in
the previous paragraph, which is described in E7.1. The inspection also
re-examined the licensee's implementation of a GL 89-10 recommendation
to periodically examine motor-operated valve (MOV) data for trends.
This recommendation had previously been examined and found
satisfactorily implemented during Inspection 50-261/94-06. It was re
examined to verify that it remained satisfactorily implemented and,
especially, to determine if the related records indicated any adverse
trends in MOV operability.
The inspection was conducted through reviews of documentation and
interviews with licensee personnel. The documents reviewed included:
CP&L GL 89-10 Corporate Improvement Plan, Revision 1.
CP&L Nuclear Generation Group Standard Procedure EGR-NGGC-0101,
"Electrical Calculation of Motor Output Torque for AC and DC Motor
Operated Valves (MOVs)," Revision 1.
CP&L Nuclear Generation Group Standard Procedure EGR-NGGC-0203,
"Motor-Operated Valve Performance Prediction, Actuator Settings,
and Diagnostic Test Data Reconciliation," Revision 3.
Engineering Service Request ESR-9700328, "Determination of MOV
Rate-of-Loading Factors," Revision 0, dated August 6, 1997.
Engineering Service Request ESR-9700330, "Determination of MOV
Valve Factors," Revision 0, dated August 7, 1997.
Sample MOV calculation packages.
Summary tabulations of MOV information and calculation results,
including a list of MOV "available valve factors" calculated using
formulas described in previous inspection reports.
Other licensee documents were reviewed, in part, as described in
subsequent paragraphs.
b. Observations and Findings
1. Procedures, Calculations, and Evaluations Revised/Prepared to Address
Findings From Inspection 50-261/96-12
MOV Sizing and Switch Settings
The licensee had revised Procedure EGR-NGGC-0203 to provide updated
guidance on MOV sizing and settings. Licensee personnel stated that
further revision was planned, to reflect additional information obtained
in ongoing reviews. The inspectors' review of the current procedure
(Revision 3) and found it generally satisfactory but identified various
concerns which will be evaluated further in a GL 89-10 closure
inspection for Robinson. Examples were as follows:
Section 3.34 of EGR-NGGC-0203 defined valve factor as the valve
disc-to-seat friction factor. The inspectors noted that valve
15
factor can also be influenced by guide-to-disc friction and metal
interference. The licensee personnel agreed and stated that the
definition of valve factor would be revised.
Section 9.3.3 of EGR-NGGC-0203 in Paragraph 1 provided guidance
for the determination of the minimum required actuator diagnostic
torque for butterfly valves. The inspectors noted that the
guidance did not include the consideration of potential spring
pack relaxation. Licensee personnel agreed and stated that spring
pack relaxation will be addressed.
A note in Paragraph 1 of Section 9.3.4 of EGR-NGGC-0203 stated
that, in certain applications, it may be justified to assume a
minimum available motor torque based on voltage conditions other
than design-basis conditions. The inspectors noted that the
assumption of degraded voltage different from design-basis
conditions must be carefully justified.
Section 9.4.2 of EGR-NGGC-0203 described the setting of MOV limit
switches based on handwheel and wormshaft revolutions. The
inspectors noted that a more precise method of verifying limit
switch settings (such as diagnostic trace analysis) may be needed
if the actuator is controlled by the limit switch setpoint.
Section 9.4.5 of EGR-NGGC-0203 provided guidance for the
calculation of valve stroke time. The inspectors noted that the
method in EGR-NGGC-0203 might not provide accurate predictions of
valve stroke time for dc-powered MOVs. Licensee personnel
understood the concern, but stated that there are currently no dc
powered MOVs at Robinson.
Section 9.5.2 of EGR-NGGC-0203 provided guidance for the
evaluation of motor stall failures for MOVs. The inspectors noted
that the guidance did not discuss evaluation of the effect of the
stall on motor performance or key/yoke structural integrity.
Licensee personnel agreed, and stated that the guidance would be
clarified.
Paragraph 4 of Section 9.6.2 of EGR-NGGC-0203 discussed the
determination of actual valve factor in the reconciliation of
dynamic test data. The inspectors noted that the determination of
a valve factor from test data up to the point of flow isolation
will be difficult to apply to other valves because of design and
manufacturing tolerances. The inspectors also noted that the use
of valve inlet pressure when the valve is fully closed (rather
than pressure at flow isolation) might result in underestimation
of the valve factor determined for the point of flow isolation.
The inspectors noted similar guidance provided in Paragraph 5 of
Section 9.6.2 on thrust extrapolation.
Paragraph 5 of Section 9.6.2 of EGR-NGGC-0203 discussed the
extrapolation of the thrust requirements from dynamic test data to
16
higher differential-pressure conditions. A note in this paragraph
specified a minimum test differential pressure of 80% of the
design-basis value. The note referenced guidance from the
Electric Power Research Institute (EPRI) on the extrapolation of
test data, but did not discuss the minimum absolute value of
contact load considered in the EPRI guidance. In response to
inspectors' questions regarding this condition, the licensee
performed a preliminary evaluation to ensure that no present
concerns existed with the extrapolation of test data at Robinson.
Paragraph 6 of Section 9.6.2 of EGR-NGGC-0203 provided guidance on
the calculation of thrust margin for an MOV to perform its safety
function. The inspectors noted that diagnostic error was not
considered. The licensee personnel indicated that the guidance
would be removed or revised to resolve this concern.
The licensee provided guidance for the calculation of motor output
torque for ac-powered and dc-powered MOVs in Procedure EGR-NGGC-0101.
The licensee was also preparing ESR-9700380 to provide guidance on
determining torque output for MOVs. The inspectors noted that these
documents indicated that the effect of ambient temperature on motor
torque output would be considered. The inspectors identified these
documents for further review, including consistency with guidance from
the manufacturer on ambient temperature effects on actuator output,
during a closure inspection for GL 89-10.
Valve Factor and Groupings
In Inspection 50-261/96-12, the NRC found that the licensee had not
adequately justified the valve factor assumptions which it used in
determining thrust and torque requirements for valves that had not been
dynamically tested. This was considered to represent inadequate design
control and was identified as Example 1.a of Violation 50-261/96-12-05,
Unjustified Design Assumptions and Incorrect Stem Rejection Load. In
developing its response to this violation, the licensee had prepared
ESR-9700330 to separate GL 89-10 valves into groups based on valve
manufacturer, model, size, seat, guide, wedge, disc materials, design
basis conditions, and installation configuration. In ESR-9700330, the
licensee established 16 groups for its GL 89-10 gate valves. The
licensee established only one group for its GL 89-10 globe valves
because all five valves were Velan 2-inch globe valves. The licensee
considered the three Allis Chalmers 16-inch butterfly valves in its GL 89-10 program also as one group. The licensee was justifying valve
factors for non-dynamically tested gate and globe valves based on test
data from other valves in the applicable group or from outside sources,
or on thrust and torque predictions that were being calculated using the
EPRI MOV Performance Prediction Methodology (PPM).
The licensee indicated that its justification for valve factors and
grouping had not been completed. The inspectors considered the
licensee's ongoing approach to be consistent with the intent of GL 89-10
and its supplements. The NRC will perform a final review during a
17
future closure inspection for GL 89-10. From their current review, the
inspectors identified concerns for evaluation during that inspection:
For some valve groups, the licensee indicated that the highest
valve factor obtained from dynamic testing of valves in the group
would be applied to other non-dynamically tested valves in that
group. It was not clear that the licensee had reviewed the test
data to ensure that the valve factor selected as the design value
for the group was consistent with the potential statistical
variation of the valve factors determined from the tested valves.
For some valve groups, the licensee indicated that the valves had
only to provide flow isolation rather than a leak-tight seal. As
already mentioned above, the thrust requirements to achieve flow
isolation are difficult to apply from one valve to another because
of design and manufacturing tolerances. Further, in Inspection
Report 50-261/96-12, the inspectors described the difficulty
experienced by licensee personnel in conservatively selecting the
point of flow isolation in the dynamic data trace for MOV AFW-V2
14A, Steam Drive Auxiliary Feed Water Pump Discharge Valve to
Steam Generator A. The inspectors also questioned whether the
licensee had reviewed the flow requirements for each valve in
these groups to ensure that a thrust requirement to achieve only
flow isolation was applicable to the valves.
The licensee initially indicated that the actual valve factor
determined from a dynamic test of some valves might be applied in
establishing the design thrust and torque requirements for those
valves. The inspectors were concerned that this might result in
the selection of a valve factor that was too low. They noted that
valve factors for MOVs can increase with age and service, as
observed in repetitive dynamic tests performed on certain of the
licensee's valves. Where the assumed valve factor (or other
parameters, such as load sensitive behavior or stem friction
coefficient) for a valve is based on a specific MOV test, small
changes in MOV performance may result in the need to evaluate MOV
operability and update the valve's setup calculations. Licensee
personnel indicated that conservative minimum design valve factors
would be selected to help address this concern.
The inspectors noted that the licensee was relying on EPRI
separate effects data and test results from double disc valves to
justify the valve factor used for two Copes Vulcan 14-inch
parallel disc gate valves (in
Group DD7). The inspectors
questioned whether this information was sufficient for these
valves. Licensee personnel indicated that the safety function of
these valves and the justification for the design valve factor
were still under evaluation.
The inspectors questioned the limited test data from another
facility that was relied on in justifying the design valve factor
for the Power Operated Relief Valve Block Valves at Robinson.
18
Licensee personnel indicated that use of the EPRI MOV PPM was
being considered to help justify the thrust requirements for these
valves.
The licensee relied on test data from the EPRI MOV test program to
justify a design valve factor of 1.1 for its globe valves. The
inspectors were concerned that the EPRI data might not be
applicable to the licensee's valves and noted that more
appropriate data might be available from other industry sources.
During Inspection 50-261/96-12, the inspectors found that the then
current version of EGR-NGGC-0203 (Revision 0) provided an incorrect
equation for calculating valve factor in the opening valve direction.
At that time, the inspectors also found the dynamic test evaluation
packages for several MOVs used this equation and incorrectly calculated
the open valve factor. The licensee's incorrect calculation of opening
valve factors was identified as Example 2 of Violation 50-261/96-12-05.
In the current inspection, the inspectors verified that the licensee had
corrected the equation in EGR-NGGC-0203, Revision 3. The licensee was
presently updating the dynamic test reconciliation data packages which
might have used the equation. The NRC will review the calculation of
open valve factor for specific valve tests during a GL 89-10 future
inspection.
Load Sensitive Behavior
During Inspection 50-261/96-12, the inspectors found that the licensee
had not adequately justified its assumption for load sensitive behavior.
This was considered to represent inadequate design control and was
identified as Example 1.b of Violation 50-261/96-12-05. In response,
the licensee prepared ESR-9700328 to justify the load sensitive behavior
assumed for MOVs not dynamically tested at Robinson. Using test data
from 13 gate valves at Robinson, the licensee calculated load sensitive
behavior as the ratio of the thrust at static torque switch trip minus
the thrust at dynamic torque switch trip to the thrust at dynamic torque
switch trip. Typically in the industry, the thrust at static torque
switch trip is used in the denominator for this calculation. The
licensee statistically analyzed the load sensitive behavior data and
determined that the mean was 5.6% and two standard deviations were
23.8%. The licensee found that the load sensitive behavior determined
for Robinson compared favorably with that determined for its other
plants and to the values obtained by EPRI using the same lubricant.
The licensee did not have dynamic test data from its globe valves for
use in determining their load sensitive behavior. In ESR-9700328, the
licensee analyzed information from globe valve testing by EPRI. Based
on its analysis, the licensee determined a mean value of load sensitive
behavior of 12.4% and a two standard deviation value of 18.6% for its
globe valves at Robinson.
The licensee indicated that load sensitive behavior would only be
applied in calculations for torque switch controlled operation. In ESR-
19
9700328, the licensee indicated that the stem friction coefficient
obtained under dynamic conditions would be applied in situations where
MOVs were controlled by limit switch setting.
The inspectors found that the approaches used by the licensee in
determining and justifying load sensitive behavior values were generally
satisfactory. The NRC will further evaluate the values selected and
their application in a future GL 89-10 closure inspection.
Stem Friction Coefficient
During Inspection 50-261/96-12, the inspector found that the licensee
had not adequately justified the stem friction coefficient assumed in
the MOV setting calculations at Robinson. This failure to justify stem
friction coefficient represented inadequate design control, which was
identified as Example 1.c of Violation 50-261/96-12-05. During the
current inspection, the inspectors found that the licensee was preparing
ESR-9700331 to establish appropriate assumptions for stem friction
coefficients at Robinson. The licensee had not completed its
justification during this inspection.
MOV Setup Calculations
The inspectors found that the licensee was updating the setup
calculations for its GL 89-10 MOVs to address the findings of a recent
self-assessment and an NRC inspection at its Brunswick facility.
Design-Basis Capability
During Inspection 50-261/96-12, the inspector reviewed the results of
tests which the licensee performed to establish the design basis
capabilities of its valves. In three examples, the licensee's
evaluations of these test results were found inadequate. These examples
were considered to represent inadequate test control and were cited in
Violation 50-261/96-12-06:
Inadequate Evaluation of Test Results.
During the current inspection the inspectors found that the licensee was
addressing the violation examples as follows in (corporate) Procedure
EGR-NGGC-0203:
Example 1 involved failures to recognize that the dynamic test
results were not consistent with the perceived test conditions.
Further, the test procedures did not ensure that the intended
design-basis test conditions were achieved during the tests. In
response to the violation, the licensee provided guidance in
Section 9.6 of EGR-NGGC-0203 to help improve the adequacy of MOV
diagnostic tests. The inspector will review the licensee's
evaluations of specific MOV diagnostic tests relative to this
violation example during a future inspection.
Example 2 involved the licensee's failure to adjust open thrust
measurements to account for measurement uncertainty identified by
the licensee's VOTES diagnostic equipment vendor, Liberty
20
Technologies, in its Customer Service Bulletin 31 (dated
November 19, 1993). In response to the violation, the licensee
revised EGR-NGGC-0203 in Section 9.1.3 to address the
consideration of VOTES diagnostic measurement uncertainty in
accordance with the applicable vendor guidance and customer
service bulletins.
Example 3 involved the licensee's failure to evaluate the
significance of a severe anomaly in the diagnostic trace of a
dynamic test of MOV MS-V1-8B, Steam Generator B Steam Supply to
Steam-Driven AFW Pump. In response, the licensee revised EGR
NGGC-0203 to improve the evaluation of diagnostic test data. The
inspector will review the adequacy of the licensee's evaluation of
specific MOV diagnostic test results during a future inspection.
The licensee also responded to the above violation examples through
changes to the Robinson plant procedures and other actions, as described
in Section E8.2 of this report.
The inspectors considered the licensee to be addressing the findings and
violation on the performance of MOV tests and the evaluation of MOV test
results identified in Inspection Report No. 50-261/96-12. The inspector
plans to complete his review during a future inspection.
In this inspection, the inspectors noted that the licensee's
determination of available capability revealed several MOVs to have
negative margin in their non-safety directions. In response to
inspectors' questions, the licensee prepared preliminary evaluations to
demonstrate no concern with the performance of the safety function of
these MOVs despite the negative margin.
2. Periodic Examination of MOV Data for Trends
Requirements for periodic examination of MOV data for trends were
described in Technical Management Manual Procedure TMM-032, "Motor
Operated Valve Program," Revision 9. This procedure required
documentation of a periodic review of maintenance and test data at least
every two years. The inspectors found that the licensee issued detailed
periodic trend reports of MOV maintenance and testing information every
two years and trend reports of MOV failures and problems every six
months. The following examples of the reports were verified by the
inspectors:
Trending of MOV Maintenance and Testing Information Reports dated
July 7, 1995 and May 1, 1997
MOV Maintenance Trending Reports for July 1 through December 31,
1996 and January 1 through June 30, 1997
The inspectors reviewed the database compiled by the licensee for the
period of October 1996 to October 1997 and found that the entries in the
database and the evaluations presented in the trend reports were
21
consistent. The recommendations of GL 89-10 for trending were
satisfactorily implemented. The inspectors found no evidence of adverse
trends in the performance of Robinson GL 89-10 valves for the period.
c. Conclusions
The licensee had not met its original commitment date for correcting
adverse findings regarding its implementation of GL 89-10 identified
during Inspection 50-261/96-12. Progress toward correcting the findings
by the new date proposed by the licensee was generally satisfactory.
Some concerns were identified, as described above. These concerns and
the outstanding findings from Inspection 50-261/96-12 will be re
examined in a future inspection which will assess the licensee's
completion of GL 89-10 implementation.
E7
Quality Assurance in Engineering Activities
E7.1 Motor-Operated Valve Program Self-Assessment (TI 2515/109)
An assessment of the Robinson MOV program was completed in August 1997
by the licensee's Nuclear Assessment Section (NAS) and was documented as
Report File No. R-ES-97-01. From a review of the assessment report, the
inspectors found that the assessment had been thorough and had resulted
in important findings. Several findings were similar to findings
independently identified by the inspectors. The inspectors verified
that the findings were appropriately identified for resolution by the
licensee in CRs 97-01822, -01824, and -01848. The corrective actions
had been or were being scheduled and licensee personnel stated that a
matrix describing the status of the NAS assessment findings would be
prepared.
7.2
Special UFSAR Review (37551)
A recent discovery of a licensee operating their facility in a manner
contrary to the UFSAR description highlighted the need for a special
focused review that compares plant practices, procedures and/or
parameters to the UFSAR descriptions. While performing the inspections
discussed in this report, the inspector reviewed the applicable portions
of the UFSAR related to the areas inspected. The inspector verified
that for the select portions of the UFSAR reviewed, the UFSAR wording
was consistent with the observed plant practices, procedures and/or
parameters.
E8
Miscellaneous Engineering Issues (92903)
E8.1
(Open) Violation (VIO) 50-261/96-12-05, Unjustified Design Assumptions
and Incorrect Stem Rejection Load: This violation identified that
important assumptions used in setting and capability calculations had
not been justified and that calculations used in determining opening
valve factors contained errors involving the stem rejection loads
selected. Section E1.2 describes the licensee's response letters and
schedule to resolve this violation and discusses some of the corrective
22
actions that had been taken. The inspectors found that the related
corrective actions identified by the licensee were specified in CR 96
03179. The corrective actions and the inspectors' findings as to the
status of each were as follows:
(1) Corrective action:
Review past test evaluations and procedures.
Then, based on the results of the review, revise set-up and
reconciliation calculations.
Inspectors' findings: The licensee had reviewed past test
evaluations and procedures. As discussed in E1.2, the inspectors
found that the licensee was preparing justifications for the
assumptions used in its set-up and reconciliation calculations and
that the calculations were in revision. The assumption
justifications and revisions to the set-up and reconciliation
calculations will be evaluated in a future inspection.
(2) Corrective action: Revise procedural guidance for calculating
opening valve factors and train personnel on the revision.
Correct previous calculations.
Inspectors' findings: The inspectors reviewed the applicable
procedure, EGR-NGGC-0203, and verified that it had been revised to
provide the correct guidance. This involved specifying use of an
appropriate equation for the calculation, as described in E1.2
above. The inspectors also verified licensee documentation of
training on the revised procedure, which was recorded on a
training report dated May 30, 1997. The inspector plans to verify
correct application of the revised guidance during a future
inspection.
(3) Corrective Action: Perform a self-assessment of the MOV program by
September 29, 1997.
Inspectors' findings: The self-assessment had been satisfactorily
completed, as discussed in E7.1.
E8.2 (Open) Violation 50-261/96-12-06, Inadequate Evaluation of Test Results:
This violation identified that the licensee had not adequately evaluated
certain test data. The licensee had not adjusted valve opening thrust
measurements for test measurement error, failed to recognize that test
data indicated that the test conditions in some tests were not as
intended, and failed to resolve significant anomalies exhibited in some
test data. Section E1.2 describes the licensee's response letters and
schedule to resolve this violation. The inspectors found that the
related corrective actions identified by the licensee were specified in
CR 96-03178. The corrective actions and the inspectors' findings as to
the status of each were as follows:
(1) Corrective action: Revise procedures to require documentation of
the acceptability of test parameters, an MOV engineer's presence
23
and preliminary review of each dynamic test at the test location,
and consideration of extrapolation error.
Inspectors' findings: The inspectors verified the procedure
changes had been incorporated into Robinson Procedures TMM-032,
Revision 9 and TMM-035, Revision 11.
(2) Corrective action:
Review previous MOV test evaluations to verify
the accuracy of the test parameters used in reconciliation
calculations. Revise set-up and reconciliation calculations based
on the findings of the review.
Inspectors' findings: The inspectors verified that the licensee
had completed the review of previous MOV test evaluations and
documented the results in Technical Report No. 97111-TR-01,
Revision 1. Set-up and reconciliation calculation revisions were
still in process. Appropriate revisions to set-up and
reconciliation calculations will be verified during a future
inspection.
(3) Corrective action: Conduct classroom training of MOV personnel on
interpretation of test data by September 29, 1997. Perform
additional training during the Spring 1998 refueling outage.
Inspectors' findings: The inspectors verified records of
completion of the initial classroom training, including test
scores. These were documented in a letter to the licensee dated
September 23, 1997, from the contractor (Liberty Technologies)
that conducted the training.
(4) Corrective action: Revise MOV program controls to require
periodic program assessments utilizing industry experts. Conduct
a self-assessment of the MOV program by September 29, 1997.
Inspectors' findings: The inspectors verified that requirements
for periodic program assessments were included in Procedure TMM
032, Revision 9. The assessments were specified to be performed
annually by MOV program experts. The self-assessment had been
completed, as discussed in E7.1.
In addition to the above corrective actions, which specifically
addressed the Robinson plant, the licensee also provided changes to a
corporate procedure (EGR-NGGC-0203), as described in E1.2 of this
report.
E8.3 (Closed) URI 50-261/97-08-02, Review Root Cause of Increased Containment
Air Temperature: This URI involved the unexpected increase in
containment average air temperature experienced between June and August
of 1997. The licensee initially implemented several initiatives to
lower containment air temperature, including; starting an additional
service water cooling pump, starting additional control rod drive
mechanism cooling fans and containment air iodine removal exhaust fans.
24
These actions had marginal impact on reducing temperatures. On July 5,
1997, the volumetrically weighted average calculation of containment air
temperature indicated that the temperature had increased slightly above
the 120oF containment design basis limit, resulting in the licensee
reporting to the NRC, a condition outside design basis. Immediate
actions were implemented to perform a containment purge to reduce
containment temperature below 120 0F. Later, continuous containment dome
cooling was implemented, as well as a pump installed to transfer water
from a cooler portions of the lake directly to the suction of the
service water intake structure. As a result of these initiatives, the
licensee was able to continue operation at full power during the hot
period of the summer without exceeding 120oF.
The containment average air temperature is an initial condition used in
the licensee's design basis accident analyses that establishes the
containment environmental qualification and operating envelope for both
pressure and temperature. The 120oF design limit ensures that operation
is maintained within the assumptions used in the design basis analyses.
UFSAR Section 9.4.3.1 states, in part, that one of the design functions
of the Containment Building Ventilation System is to remove the normal
heat lost from all equipment and piping in the reactor containment
during plant operation and maintain a temperature of 120'F or less
inside the containment, with 950F cooling water and three-out-of-four
Containment Air Recirculation Cooling (CARC) fans operating.
The inspector reviewed modification package ESR 9500783, which was
implemented during the previous (RFO 17) refueling outage.
Additionally, the inspector reviewed CR 97-01471, which was initiated by
the licensee to investigate the increased containment temperature. The
modification changed the normal power operation alignment of the four
CARC fan units, HVH-1, 2, 3, and 4. Specifically, the normal air intake
dampers were secured in the closed position and the emergency intake
butterfly valve dampers were secured in the open position. Previously,
during normal power operations, the normal intake dampers were
positioned open and the emergency intake dampers were closed. Following
a Safety Injection (SI) actuation, prior to the modification, the normal
dampers would close and the emergency dampers would open. Following the
modification, the damper positions would not change following an SI,
(i.e., the normal dampers would stay closed and the emergency dampers
would stay open).
The inspector noted several inadequacies in the licensee's
implementation of modification ESR 9500783. On October 11, 1996,
partial turnover of the modification was performed following completion
of air flow measurements. The CARC system was declared operable at that
time. However, from review of modification test data results, on
July 15, 1997 the inspector noted that only 245,133 cubic feet per
minute (cfm) of total air flow was measured with all four CARC units
operating, while the test acceptance criteria specified a minimum of
255,000 cfm. At that time, a note was written next to these steps,
indicating that the lower flow was acceptable due to flow measurement
instrument tolerances. The inspector considered the documentation of
25
these steps to have been untimely, and the note was not adequate
documentation for resolving the failure to meet the test acceptance
criteria.
The design review also did not consider that four CARC units were
necessary to maintain temperatures below 120 0F in the hot periods of the
summer during normal operations, as opposed to three, as stated in UFSAR
Section 9.4.3.1. The modification should have recognized this and
appropriately updated the UFSAR section to accurately reflect the plant
configuration requiring four CARC units maintain temperatures below
120 0F. The aforementioned UFSAR section had existed prior to the
modification, and the modification presented an opportunity to address
this inconsistency, but was not recognized by the licensee.
Additionally, it was expected that the modification in damper alignment
at power would result in enhanced cooling performance from the CARC
units. The modification erroneously assumed that air flow through the
cooling coils would be more evenly distributed as compared to the
original configuration. The change actually had an opposite effect,
resulting in less cooling capability of the CARC units. The
modification package had indicated that appropriate testing would be
performed to assure that the air flow as well as cooling performance of
the system during normal power operation would be equal to the original
Design Basis. However, post modification testing consisted of only
measuring the air flow through the system with all four CARC units
operating concurrently, and not the cooling performance testing.
The
measurement of air flow alone was not sufficient to confirm the cooling
performance of the CARC units nor the assumption that the cooling
performance would be improved. Thus, the cause of the elevated
containment temperatures stemmed from the implementation of modification
ESR 9500783 during the previous refueling outage. The normal cooling
capability of the CARC system at power was adversely impacted due to
uneven air flow distribution across the cooling coils of the CARC units.
10 CFR 50, Appendix B, Criterion V, Instructions, Procedures, and
Drawings, requires in part that activities affecting quality shall be
prescribed by documented instructions and shall be accomplished in
accordance with these instructions. Further, the instructions shall
include quantitative or qualitative acceptance criteria for determining
that important activities have been satisfactorily accomplished.
The inspector determined that when post modification test acceptance
criterion was not met, an evaluation was not conducted prior to
declaring the CARC system operable. The failure to follow procedure as
required .by procedure EGR-NGGC-005 was a violation of 10 CFR 50,
Appendix B, Criterion V. This is identified as example 1 of violation
50-261/97-12-04: Failure to Accomplish Modification Related Activities
for the Containment Air Recirculation Cooling System. Further, the
modification package indicated that appropriate testing would be
performed to assure that the air flow as well as cooling performance of
the system would be equal to the original Design Basis. However, the
post modification test for ESR 9500783 did not evaluate the cooling
26
capability following the modification to confirm that the system was
capable of meeting its Design Basis during normal plant operations. The
failure to follow the modification package requirements is identified as
example 2 of violation 50-261/97-12-04.
E8.4 (Closed) URI 50-261/96-14-03, Review Aspects of Containment Spray
Additive Tank Eductor Line Sampling: This URI involved inspector
concerns with the adequacy of the licensee's 10 CFR 50.59 evaluation
associated with procedures for conducting quarterly sampling/flushing of
the sodium hydroxide eductor piping associated with the Containment
Spray System (CSS). In order to conduct the sampling/flushing
activities, the spray additive tank eductor piping was aligned in a
configuration that degraded its design function of automatically
providing sodium hydroxide to the suction of both CSS pumps. Instead of
declaring the sodium hydroxide addition system inoperable, credit was
taken for manual operator actions to restore the eductor piping to its
automatic alignment within six minutes in the event of a CSS actuation
signal.
The six minutes was based on the results of a licensee
calculation that indicated the effect of this delay in sodium hydroxide
addition to the containment atmosphere during a design basis accident
would not result in exceeding General Design Criteria (GDC) 19 or 10 CFR
100 dose limits.
In the previous inspection, the inspector questioned the adequacy of the
safety evaluations which supported these procedure changes, in that, the
evaluations appeared to involve an unreviewed safety question. The
procedures introduced two new failure modes for the iodine removal
system, which both, increased the possibility for malfunction of a
different type than any evaluated in the UFSAR. These new failure modes
involved the potential malfunction of manual valves that the operator's
were dedicated to open which were previously in an open position, and
the potential failure of the operators to perform the manual realignment
within the time allowed. The failure to return the system to its
automatic alignment due to either of the above reasons within six
minutes could have resulted in the failure of the iodine removal system
from performing its intended safety function. In addition, the
licensee's calculation to support the six minute delay indicated that
the GDC 19 dose limits would increase from 27.3 rem to 29.7 rem. This
was very close to the GDC 19 limit of 30 rem.
The licensee had previously indicated that they believed that guidance
in NRC Inspection Manual Chapter, Part 9900, Interim Guidance on 10 CFR
50.59, issued in April 1996, and Generic Letter 91-18, Information to
Licensees Regarding Two NRC Inspection Manual Sections on Resolution of
Degraded and Nonconforming Conditions and on Operability, allowed the
substitution of manual action for automatic action.
As a result of the inspector's concerns regarding the sampling/flushing
activities, the licensee initiated Engineering Service Request (ESR)
9700050 to evaluate whether the practice should be continued. The
results of this evaluation indicated that, since 1991, there had not
been any evidence of leakage past the Spray Additive Tank Isolation
27
Valves, SI-845A&B; therefore, it was recommended that the
sampling/flushing activity be discontinued. Based on the results of ESR
970050, the sampling/flushing activities were removed from operations
surveillance procedures (OST-352-1 and -2) associated with the CSS that
were conducted at power operation.
The inspector noted that the NRC has recognized that better regulatory
guidance is needed to ensure the consistent implementation of 10 CFR
50.59, including such areas as, the use of compensatory measures to
offset small potential increases in probabilities, as well as defining
when reductions in the margin of safety are evident. Based on
discussions with licensee engineering and licensing personnel,
additional training to personnel conducting 10 CFR 50.59 safety
evaluations had been provided since the time that the inspector
initially identified this problem. The inappropriate crediting of
operator action in lieu of automated actuations had been specifically
addressed during this training.
The licensee was in the process of enhancing their 10 CFR 50.59 program
by the addition of a Design Review Panel and PNSC Safety Evaluation
Subcommittee. The purpose of the panel and subcommittee was to review
all 10 CFR 50.59 evaluations and a sample of evaluation screens to
ensure adequacy and adherence with procedures and guidance.
Additionally, procedures related to the implementation of 10 CFR 50.59
were scheduled to be revised in December 1997 to provide better 10 CFR
50.59 guidance, especially in the area related to compensatory measures
and margin of safety. Based on the licensee's completed and planned
actions to enhance their 10 CFR 50.59 program, and the licensee's
revision of procedures deleting the sampling/flushing activities, this
URI is closed.
IV. Plant Support
R1
Radiological Protection and Chemistry Controls
R1.1 General Comments (71750)
The inspector periodically toured the Radiological Control Area (RCA)
during the inspection period. Radiological control practices were
observed and discussed with radiological control personnel including RCA
entry and exit, survey postings, locked high radiation areas, and
radiological area material conditions. The inspector concluded that
radiation control practices were proper.
28
S1
Conduct of Security and Safeguards Activities
S1.1 General Comments (71750)
During the period, the inspector toured the protected area and noted
that the perimeter fence was intact and not compromised by erosion nor
disrepair. Isolation zones were maintained on both sides of the barrier
and were free of objects which could shield or conceal an individual.
The inspector periodically observed personnel, packages, and vehicles
entering the protected area and verified that necessary searches,
visitor escorting, and special purpose detectors were used as applicable
prior to entry. Lighting of the perimeter and of the protected area was
acceptable and met illumination requirements.
V. Management Meetings
X1
Exit Meeting Summary
The inspectors presented the inspection results to members of licensee
management at the conclusion of the inspection on December 3, 1997. No
proprietary information was identified.
29
PARTIAL LIST OF PERSONS CONTACTED
Licensee
J. Boska, Manager, Operations
H. Chernoff, Supervisor, Licensing/Regulatory Programs
T. Cleary, Manager, Maintenance
J. Clements, Manager, Site Support Services
J. Keenan, Vice President, Robinson Nuclear Plant
R. Duncan, Manager, Robinson Engineering Support Services
R. Moore, Manager, Outage Management
J. Moyer, Manager, Robinson Plant
R. Warden, Manager, Nuclear Assessment Section
T. Wilkerson, Manager, Regulatory Affairs
D. Young, Director, Site Operations
NRC
B. Desai, Senior Resident Inspector
J. Zeiler, Resident Inspector
30
INSPECTION PROCEDURES USED
IP 37551:
Onsite Engineering
IP 40500:
Effectiveness of Licensee Controls in Identifying, Resolving, and
Preventing Problems
IP 61726:
Surveillance Observations
IP 62700:
Maintenance Implementation
IP 62707:
Maintenance Observation
IP 71707:
Plant Operations
IP 71714:
Cold Weather Preparations
IP 71750:
Plant Support Activities
IP 92901:
Followup - Operations
IP 92902:
Followup - Maintenance
IP 92903:
Followup - Engineering
IP 93702:
Prompt Onsite Response to Events at Operating Power Reactor
T12515/109: Inspection Requirements for Generic Letter 89-10, Safety-Related
Motor-Operated Valve Testing and Surveillance
ITEMS OPENED,
CLOSED,
AND DISCUSSED
Opened
Type
Item Number
Status
Description and Reference
50-261/97-12-01
Open
Failure to Log TS Surveillance Completion
in Accordance with OST-020 (Section 01.3)
50-261/97-12-02
Open
Failure to Verify Dose Equivalent 1-131 in
Accordance with GP-005 (Section 01.3)
IFI
50-261/97-12-03
Open
Review Safety Injection Discharge Piping
Configuration (Section E1.1)
50-261/97-12-04
Open
Failure to Accomplish Modification Related
Activities for the Containment Air
Recirculation Cooling System
(Section E8.3)
Closed
]Iype
Item Number
Status
Description and Reference
50-261/97-12-01
Closed
Failure to Log TS Surveillance Completion
in Accordance with OST-020 (Section 01.3)
50-261/97-12-02
Closed
Failure to Verify Dose Equivalent 1-131 in
Accordance with GP-005 (Section 01.3)
50-261/97-10-01
Closed
Complete Review of LPMS Testing and
Maintenance Activities (Section 08.1)
31
50-261/97-08-02
Closed
Review Root Cause of Increased Containment
Air Temperature (Section E8.3)
50-261/96-14-03
Closed
Review Aspects of Containment Spray
Additive Tank Eductor Line Sampling
(Section E8.4)
Discussed
13pe
Item Number
Status
Description and Reference
50-261/96-12-05
Open
Unjustified Design Assumptions and
Incorrect Stem Rejection Load (Section
E8.1)
50-261/96-12-06
Open
Inadequate Evaluation of Test Results
(Section E8.2)