ML20199D343
ML20199D343 | |
Person / Time | |
---|---|
Site: | Hope Creek |
Issue date: | 11/13/1997 |
From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
To: | |
Shared Package | |
ML20199D308 | List: |
References | |
50-354-97-07, 50-354-97-7, NUDOCS 9711200314 | |
Download: ML20199D343 (42) | |
See also: IR 05000354/1997007
Text
s - _- ._ _ - .
. ,
g"
U. S. NUCLEAR' REGULATORY COMMISSION
' REGION I
- Docket No: ' 50 354
License Nos:- NPF 57-
. Report No. - 50 354/97 07
Licensee: Public Service Electric and Gas Company -
- Facility: . Hope Creek Nuclear Generating mation.
Location:- P.O. Box 236
Hancocks Bridge, New Jersey 08038
Dates: August 24,1997- October 4,1997
Inspectors: S. A. Morris, Senior Resident inspector
J. D. Orr, Resident inspector
J. D. Noggle, Senior Radiation Specialist
A. L. Dclla Greca, Senior Reactor Engineer
K. Young, Reactor Engineer
J. T. Yerokun, Senior Reactor Engineer
\;
Approved by: James C. Linville, Chief, Projects Branch 3
Division of Reactor Projects
.'
r.
'
.,-
,
9711200314 971113
. gDR ADOCK 05000354
PDR ,
4 ,
h #,s - . ~ - -- * ,
.
.
EXECUTIVE SUMMARY
Hope Creek Generating Station
NRC Inspection Report 50 354/97-07
This integrated inspection included aspects of licensee operations, engineering,
maintenance, and plant support. The report covers a six week period of resident
inspection; in addition, it includes the results of announced inspections by three regional
inspectors. Region-based inspections involved reviews of the Hope Creek Inserv:ce
inspection (ISI) program, engineering support, and outage radiological controls.
Operations
Plant operators were frequently challenged by unexpected operational transients and
material conditions caused both by equipment failures and human errors. However,
immediate operator response to the events was typically good in that proper procedures
were used, technical specification action statement requirements were followed, and three-
way communications. (Section 04.1)
Station operators exhibited inconsistent performance with regard to attention-to-detail and
human error. Weaknesses in this area were highlighted by a violation involving a f ailure to
detect an inoperable station fire pump despite several opportunities for both operators and
fire protection technicians to identify the issue in a timely manner. (Section 04.2)
Operators demonstrated proper implementation and knowledge of all applicable technical
specification requirements during plant eperation, shutdown, and refueling. Additionally,
conservative decision making was evident during the course of various infrecuently
performed evolutions. lSection 04.3)
QA inspectors provided excellent oversight of plant operations, and routinely
communicated observed deficiencies to shift management. (Section 07.1)
Poor operations department internal communication was evident when chemistry
technicians did not inform plant operators of an out-of specification standby liquid control
tank concentration until two and one half hours after the discovery (Section 08.2)
Maintenance
Technical specification required surveillance testing was conducted effectively. Governing
procedures were of good quality and were properly followed. Pre-job briefings were
thorough and demonstrated appropriate coordination for test completion. (Section M1.1)
Plant housekeeping and cleanliness declined during the period. (Section M2.1)
The ISI program was appropriately implemented in accordance with an approved plan and
ASME Code Section XI. The ISI program manager exhibited a good understanding and
ownership of the ISI program. (Section M2.2)
l ii
l
l
!
.
.
The snubber surveillance program was implemented in accordance with technical
specifications by knowledgeable and technically competent individuals. The ongoing
efforts to upgrade all plant snubbers was technically sound and was judged to be a good
initiative to preclude problems previously experisnced with plant snubbers. (Section M2.3)
Immediate response to understand and resolve an IGSCC-induced through wallleak on the
"A" ; ore spray pipe nozzle was good, however, questions remained regarding the
adequacy of ultrasonic testing data enalysis during the previous refueling outage. (Section
M2.4)
In spite of proactive measures by PSE&G management to reinforce expectations regarding
maintenance procedure adherence and attention-to-detail, several issues involving non-
compliancer and poor work quality were identified. Supervisory oversight of contract
maintenance technicians was weak. (Section M4.1)
PSE&G personnel continued to document deficient conditions at the station with a low
threshold for initiation. Plant management demonstrated good reviews of each new issue
and required resolution during the refueling outage when appropriate. However, several
issues were not documented in a timely manner to ensure timely review and corrective
action. (Section M7.1)
Procedure changes incorporated following the most recent failure of the residual heat
removal shutdown cooling suction line snubber, which modified the method used to place
shutdown cooling in service, were effective and prevented recurrence. (Section M8.1)
Enaineerina
System engineering efforts in identifying and resolving emergent equipment f ailures were
prompt and effective. The large scope of design change work scheduled for complet;on
during the refueling outage demonstrated appropriate management focus on resolution of
long standing equipment deficiencies. However, several examples of repeat equipment
failures, extended system maintenance activities, and design change package deficiencies
highlighted weakness in the quality of engineering support. An internal review of an
extended reactor wate cleanup pump outage was thorough and self critical. (Section
E2.1)
The program for designing and installing configuration changes to plant systems was
acceptable. However, inadequate review of design documentation prior to implementation
of a reactor core isolation cooling system modification resulted in a violat..in of 10 CFR
50.59 requirements in that no written safety evaluation was performed for the change as
required. This failure was also an example of weak implementation of the peer review
process. (Section E1.1)
The licensee generally took acceptable actions to address nonroutine plant events. When a
root cause analysis was prepared, the analysis was thorough and detailed. Not all
conclusions were conservative, however, as in the case of Struthers-Dunn relays in a mild
environment. (Section E2.2)
lii
.
.
For Struthers Dunn relays in harsh environment, the failure to include five relays in the EQ
list and to provide a reasonable technical justification for accepting a less than required
relative humidity qualification resulted in a violation of 10 CFR 50.49. Also, the use of a
celculated relay qualified life without reconciling the difference with actual data indicated -
an excessive reliance on a theoreticallife extension method that is highly dependent on the
correct selection of independent variables. (Section E2.2)
The delay in initiating a Struthers-Dunn relay failure analysis (24 relay replacements in
three years) indicated a weakness in the program for monitoring the per'ccmance of safety-
related components in a mild environment. (Section E2.2)
Less than acceptable judgement was used in the selection of the coil temperature rise of
normally energized, safety related Telemechanique and Agastat relays in a mild
environment. Failure to verify the acceptability of the values used in life extension
calculations and tests resulted in a violation of 10 CFR 50, Appendix B, Criterion 3.
(Section E2.3)
The OA audit of Hope Creek engineering was good and provided an accurate assessment
of the engineering programs. (Section E7.1)
Acceptable actions were taken to address several previously-identified issues. An
inspection followup item was opened to evaluate resolution of medium voltage circuit
breaker f ailures experienced during a recent event. (Section E8.1)
Plant Suonort
Generally good radiological control practices were observed during the period, which
included both operational and shutdown conditions. Radiologically controlled crea access
controls were effective. (Section R1.1)
Refueling outage radiation work permits generally provided effective contamination control
requirements, however, exposure reduction plans were not specified as job requirements.
(Section R1.2)
PSE&G demonstrated good progress in developing and implementing an initial drywell
shielding plan during the refueling outage, however, some weaknesses were noted with
drywell postings and electronic dosimetry setpoints. (Section R1.2)
Some weaknesses in radiation protection controls were observed during the outage,
including refueling tool contamination control and inside torus air sampling practices.
(Section R1.2)
Radiation protection planning activities were not wellintegrated with outage work
management planning and scheduling which resulted in ten than effective ALARA
performance. (Section R1.3)
The radiation protection continuing training program has been weak as evidenced by poor
technician performance on a recent examination. (Section R5.1)
iv
I
. . _ . . . _ . _ . _ _. . __ . , _ _ _ ._ _ . . _ _ - _ _ . . _ _ _ ... . _ . _
.-
.,.
? Generally good implementation of Hope Creek emergency plan requirements was observed--
'- - during an unannounced drill. Appropriate procedures were used, good communications
were established, and a proper turnover from the senior nuclear shift supervisor 'was
completed. (Section P1.1)
>
a
i
!
4
s
9-
! V
l
l
l.
l
!
i
- . x. . . . a . z _,, . _ + - , .~.. . .,_. _ , .
.. - - ._ -- -
.
.-
TABLE OF CONTENTS
' EX ECUTIVE SUM M A RY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
TA BL E O F C O NT E NT S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
1. O p e r a t i o n s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
04 Operator Knowledge and Performance ' . . . . . . . . . . . . . . . . . . . . . . . . . 1
04.1 Event Re sponse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
04.2 Attention to-Detail and Human Performance ' . . . . . . . . . . . . . . . . 2
04.3 Technical Specification Compliance and Operator Decision Making . 4
07- Quality Assurance in Operations . . , . . . . . . . . . . . . . . . . . . . . . . . . . 5
07.1 Quality Assurance Oversight of Operations . . . . . . . . . . . . . . . . . 5
08 Miscellaneous Operations issue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
08.1 (Closed) DEV 50 354/9611-01: failure to revise TS bases . . . . . . 5
08.2 (Closed) LER 50 354/97-021: standby liquid control (SLC) system ,
tank concentration below technical specification limits ........ 5 ,
08.3 (Closed) LER 50 354/97-022: engineered safety feature actuation -
unplanned manual scram following a relay malfunction in the 'A"
- phase main generator step-up transformer ................. 6
ll M a i n t e n a n co . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
M1 Conduct of Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
M1.1 Technical Specification Surveillance Testing Observations .......S
M2 Maintenance and Material Condition of Facilities and Equipment . . . . . . . 7
M2.1 Station Housekeeping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
M 2.2 Inservice inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
M2.3 Snubber Surveillance Program . . . . . . . . . . . . . . . . . . . . . . . . . 10
M2.4 "A" Core Spray Nozzle (NSB) Safe End Weld Pin Hole Leaks . . . . 11
M4 Maintenance Staff Knowledge and Performar.ce . . . . . . . . . . . , , . . . . 12
M4.1 Maintenance Observations . . . . . . . . . . . . . . . . . . . . . . . . , . . 12
M7 Quality Assurance in Maintenance Activities . . . . . . . . . . . . . . . . . . . . 14
M7.1 Problem identification in Maintenance ...................14
M8 Miscellaneous Maintenance issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
M8.1 (Closed) VIO 50-354/E 96-014-01013: repeat failure of residual heat
removal system shutdown cooling common suction line mechanical
snubber ........................................15
M8.2 (Closed) VIO 50-354/97 02-01: failure to shut emergency diesel
generator (EDG) cylinder test cocks prior to engine operation . . . 15
M8.3 (Closed) LER 50-354/97-018: engineered safety feature actuations as
a result of reactor protection system (RPS) motor generator set output
bre a k e r t rip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
fil. Engine ering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....................16
.
E1.1 Plant Modifications - Configuration Changes . . . . . . . . . . . . . . . 16
E2 Engineering Sepport of Facilities and Equipment ................. 18
4 E2.1 Engineering Support of Plant Operations and Maintenance . . . . . 18
E2.2 Engineering Involvement in Site Activities . , . . . . . . . . . . . . . . . 19
vi
i
,
i
y
.
e-
_ ._ _ _ _ _ __ _ - . . __ __. . _
q. ,
.
.
E2.3 Service Life of Relays in Mild Environment ................21
E7 . Quality Assurance in Engineering Activities . . . . . . . . . . . . . . . . . . . . . 27
E7.1 Audits and Assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
EB- Miscellaneous Engineering issues ._ , . . ._ . . . . . . . . . _. . . . . . . . . . . _. . . 27
E8.1 Medium Voltage Circuit Breaker Failure ..................27
E8.2 (Closed) VIO 50 354/96-04-01: Failure to account for all Bailey solid
state logic module (SSLM) f ailures . . . . . . . . . . . . . . . . . . . . . . 28
E8.3 (Closed) URI 50 354/96-04-02: Control of the 8ailey solid state
- module automatic tester programs. . . . . . . . . . . . . . . . . . . . . . 29
E8.4 (Closed) IFl 50 354/96-04-03: Bailey solid state logic module
electromagnetic interference circuit testing . . . . . . . . . . . . . . . . 30 .
E8.5 (Closed) IFl 50-3 54/9 6-04-04 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 '
E8.6 ~ (Updated) VIO 50-354/96-09-03: Interaction between nonsafety and
safety related components in the standby diesel generator room
ve ntilation syst e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
E8.7 (Closed) URI 50 354/97-01-04: apparent creatior' of an unreviewed
safety question following installation of cross-tie ones between
residual heat removal subsystems . . . . . . . . . . . . . . . . . . . . . . 32
E8.8 '(Closed) VIO 50-354/97-01-05: reactor core isolation cooling turbine
overspeed trip due to governor valve stem binding . . . . . . . . . . . 33
E8.9 (Closed) LER 50-354/97-019: closure of SACS to TACS isolation
valve ..........................................33
E8.10 (Closed) LER 50-354/97-020: past inoperability of safety related
chillers due to operation with low safety auxiliaries cooling system
(S ACS) temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 3 3
E8.11 (Closed) LER 50 354/97-005: operation in a technical specification
prohibited condition due to failu;e to perform monthly flowpath
verification surveillance checks of residual heat removal system cross-
tie valves .......................................33
I V. Pl a nt S u p p o rt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4
R1 Radiological Protection and Chemistry (RP&C) Controls . . . . . . . . . . . , 34
R1.1 General Observations of Radiological Control Practices ....... 34
R1.2 Outage RP Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
R1.3 Outage ALARA Performance . . . . . . . . . . . . . . . . . . . . . . . . . . 36
R5 Staff Training and Qualification in RP&C . . . . . . . . . . . . . . . . . . . . . . . 37
R5.1 RP Technician Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
R7 Quality Assurance in RP&C Activities . . . . . . . . . . . . . . . . . . . . , , . . . 38
R8 Miscellaneous RP&C lssues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
R8.1 (Closed) URI 50-354/97-03-02: lack of timely completion of a design
change package for meteorological monitoring instrumentation . . 38
R8.2 (Open/ Closed) VIO 50-354/97-07-12: violation of the new
Department of Transportation (DOT) shipping paper requirements
..............................................38
P1 Conduct of EP Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9
P1.1 Unannounced Emergency Preparedness Drill . . . . . . . . . . . . . . . 39
P8 Miscellaneous EP lssues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9
P8.1 (Closed) URI 50-354/96-07-03:UFSAR discrepancies regarding
vii
, _ _ -
.
.
PSE&G emergency plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9 -
'-
V. M anagement M eeting s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9
X1 Exit Meeting Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9
,
Viii
. -__ . _ . . . _ -- .. _
.:
l
4 .
.Recort Details .I
1. Operations .
l
04 Operator Knowledge and Performance
04.1 Event Response
a. Inspection ScoatllflQ2)
- The inspectors caserved or reviewed operator actions during and following several;
unexpected tran sient events and conditions throughout the report period,
b. Observations and Findinos
.,
The ir spectors noted that a relatively high number of unanticipated operational
events' occurred during the six week report period, which challenged the plant
operators, including:
- "A" reactor re circulation pump trip due to failed transformer (9/2)
- turbine auxiliaries cooling system isolation due to failed fuse holder (9/4)
- offsite alert notification siren actuatito due to poor maintenance controls (9/8)
- manual reactor scram due to failed main transformer cooling controls (9/10) ,
e condensate pump breaker failure due to faulty trip coil (9/10)
- reactor water cleanup system isolations due to failed flow transmitter (9/11)
- remote shutdown panel channel " takeover" during relay replacement (9/12)
- safety auxiliaries cooling system pump start due to operator error (9/14)
- service water pump start during remote shutdown panel surveillance (9/16)
Other unexpected issues also required prompt vperator notifications and attention,
including the discovery that the technical specification (TS) minimum boron
concentration in the standby liquid control system storage tank was not maintained,
and the identification of a through-wallleak in the "A" core spray loop p! ping nozzle
to the reactor pressure vessel. Additionally, a failed refueling bridge main hoist
cable emergency brake solenoid resulted in an emergency stop/deenergization of the
bridge controls while moving spent fuel from the reactor core to the fuel storage
pool.
The inspectors judged the causal factors of the noted eveMs were an even mix of
both human performance and material condition issues. Human error induced
-
. events typically involved failures to "self-check" or to understand system design
and operation _ Other events primarily resulted from random equipment failures.
Several of the noted events are described in further detail in later sections of this
report.
In general, immediate operator response to the events was good; governing
abnormal procedures were used,- communications were clear and accurate, and
conservative decision making was evident. When required, events were properly
!? - reported to the NRC in accordance with 10 CFR 50.72 and 50.73. For example, cn
-
,
,_a. -. .
-
)
.
2
September 2,1997,the "A" reactor recirculation pump tripped causing a rapid
power reduction to 40E The inspectors witnessed operator respor.se to this event
and observed good follow up activities. Appropriate control rod insertions were
made to ensure that the reactor power / flow ratio was maintained outside to the
" instability" region, TS action statements were properly implemented for single
recirculation loop operation, and balance of plant equipment was appropriately
aligned to support the reduced power level. This event was caused by a f ailed
transformer in the recirculation pump motor generator field excitation circuit.
On September 10, operators received a main output step up transformer trouble
alarm in the control room ,nd promptly dispatched equipment operators (EO) to
investigate. Upon arrival, the EO's reported that the transformer cooling fans were
not running, there was an acrid smell, and there were unusual noises. Operators
acted conservatively by quickly reducing reactor power using recirculation flow and
manually scramming the plant. The inspectors observed proper use of the
emergency operating procedures, good three way communications, and detailed log
keeping. The main output transformer problem was later attributed to a failed relay
in E cooling system control circuit. (See also section 08.3)
On September 19, while in operating condition 5 with fuel in the reactor vcssel, EOs
identified leakage in the drywell which was promptly traced to a through-wallleak
from the "A" core spray reactor vessel nozzle. The inspectors independently
validated the leak source tocation. Operators properly deciated ll affected plant
systems inoperable, including "A' core spray, high pressure coolant injection, and
standby liquid control, the latter two systems which use the core spray penetration
as their vessel injection point. Because the "B" and "D" channels were inoperable
a. the time of the discovery for scheduled outage work, the TS specifying the
minimum operable trains of emergency core cooling could not be satisfied. Again,
operators promptly recognized this condition and took appropriato actions. A timely
and accurato non-emergency report was made to the NRC in accordance with 10
CFR 50.72 (See also section M2.4)
c. Conclusions
Plant operators were frequently challenged by unexpected operational transients and
material conditions caused both by equipment feitures and human errors. However,
immediate operator response to the events was typically good in that proper
procedures were used, technical specification action statement requirements were
followed, three-way communications were employed, and conservative decisions
were made.
04,2 A11#ntion-to-Detail anijduman Performance
!
a, inspection Scoce (71707)
The inspectors evaluated several events during the period wF 'ocus on individual
attention-to-detail and human error. Interviews, observatit e. u.id log reviews were
conducted while forming this assessment.
I
.
._. . _ .- . _ _ _ . . - - __ . . . . __ _ _ -_ _
~
.
,
4
3
b.- Observations and Findinas
. The inspectors observed inconsistent performance at alllevel' s of the operations
department with respect to attention to-datail and human error. Several examples
of quality operations department findings were noted which resulted from good
questioning, awareness, and research into current plant conditions. For example,
several "near-miss" tagging events were avoided by alert operators assigned to
remove blocking tags from equipment which was declared ready for restoration by
maintenance personnel. Also, during core alterations, refuel bridge operators
identified a double blada guide in the fuel storage pool that was improperly
configured. Lastly, a senior operator dettmined that 18-month relay testing on a
4160V vital bus was overdue and was not properly scheduled before expiration of
the TS allowed grace period. ,
However, the inspectors also noted several examples of poor operator performance,
improper tracking of surveillance activities for two trains of the filtration,
recirculation, and ventilation system nearly resulted in unplanned inoperability of the
system. A spent fuel pool cooling pump tripped after an operator failed to adjust an
automatic flow control setting before restoring a filter demineralhr valve to service.
The "D" station service water pump started unexpectedly during a surveillance
activity when a remote shutdown panel "emergene, takeover" logic transfer switch
was returned to its normal position because the pump was left in automatic control.
On September 15,1997, the operations shif t supervisor determined that the electric-
motor driven fire pump at Hope Creek had been inoperable for approximately 34
hours, likely the result of a bus swap which caused the motor supply breaker to
open. Though a fire protection system trouble alarm was generated and
acknowledged in the control room on September 14, no actions were taken to
determine the cause of the alarm, and fire pectection technicians were not notified
of the condition. Additionally, fire technicians missed two opportunities to ioentify
the inoperable pump during routine operator rounds. The inspectors judged that this
event highlighted weaknesses in degraded condition problem identification;
specificaily attention-to-detail, questioning attitude, and understanding fire
suppression system status. Because this issue resulted in an unknown degradation
of the fire suppression system for an extended period, and because of the number
of missed opportunities to identify and correct the issue, the inspectors determined
that it was a violation of 10 CFR 60 Appendix B, Criterion XVI in that the fire
suppression system degradation was not promptly identified and corrected. (VIO
50-354/97-07-01)
c. Conclusions
Station operators exhibited inconsistent performance with regard to attention-to-
detail and human error. Weaknesses in this area were highlighted by a violation
involving a f ailure to detect an inoperable station fire pump despite several
opportunities for both operators and fire protection technicians to identify the issue
in a timely manner.
l
l
l
.r, ,
.
.
4
Q4J Itchnical Soecification Comoliance and Operator Decision Makina
a. !nsoection Scone D1707)
Several plant operational condition changes occurred during the report period which
required a thorough review and understanding of TS requirements to ensure that
compliance was maintained. The inspectors focused en operator performance
during the changing conditions to ensure that all regulatory requirements were
satisfied,
b. Observations and Findinas
Several plant operational condition changes, both planned and unplanned, occurred
during the report period, including:
- sing!e reactor recirculation loop operation and recovery
- reactor scram from operating condition 1 to 3
- reactor plant cooldown to operating condition 4/5
- reactor vessel disassembly and core alterations
e operations with potential to drain the vessel (control rod drive replacemen's)
In every case, the inspectors observed that applicable TS requirements were
satisfied. Additionally, " tracking" action statements were logged to ensure that
i'1 operable equipment needed for an operating condition not currently applicable
were known before modo changes. Surveillance requirements for infrequently
performed evolutions were properly conducted at the correct frequencies.
Appropriate retests were performed following unexpected failures of the refueling
bridge.
The inspectors witnessed generally conservative plant operations and decision
making. For example, refuel bridge operators did not hesitate to delay core
alterations when reactor cavity lighting or water clarity degraded. Fuel handling
wa:, generally restricted to single dimension movements to enhance control of the
evolution. Lessons learned from previous loss of shutdown cooling events at the
station were reviewed during pre-evolutions briefings to ensure that reliable cooling
system operation was maintained. A detailed engineering analysis and operability
determination were performed and reviewed by the station operations review
committee to justify the acceptability of " flooding up" the reactor cavity with the
degraded core spray nozzle,
c. Conclusions
Ope stors demonstrated proper implementation and knowledge of all applicab:e
technical specification requirements during plant operation, shutdown, and
refueling. Additionally, conservative decision making was evident durir g the course
of various infrequently performed evolutions.
- _ _
,
..
5
07 Quality Assurance in Operations
.Q11 Quality Assurance Oversicht of Operationg.
The inspectors observed nearly continuous quality assurance (QA) department
presence in the Hope Creek control room just prior to ind during the refueling
outage. QA inspectors provided excellent oversight of plant operations, and-
routint.ly communicated observed deficiencies to shift management. Daily QA
observation reports were completed which documented findings. The inspectors
reviewed three week sample of QA observation reports and judged them to be
thorough and insightful. Several action requests were generated as a direct result
of QA . questioning and assessment. - The inspectors concluded that the frequent QA
oversight in the control room was a good initiative which enhanced safe plant i
'
operation.
08 Miscellaneous Operations issue
08.1 (Closed) DEV 50-354/96-11-01: failure to revise TS bases as committed in a PSE&G -
license amendment for specific conditions that must be satisfied to justify extended
allowed outage times for the "C" and "D" emergency diesel generators. The
inspectors reviewed PSE&G's March 21,1997 letter which acknowledged the
- deficiency, and judged the stated corrective actions to be reasonable. Actions
included submitting revised TS bases for NRC approval (approved in TS Amendment
101), enhancing the licensing department process for tracking commitments, and
verifying implementation of previous cornmittnents. The inspectors observed that
these actions were completed in a timely manner.
08.2 (Closed) LER 50-354/97-021: standby liquid control (SLC) system tank
concentration below technical specification limits. This issue was self identified
during routine monthly tank sampling. The sodium pentaborate concentration in the
tank was restored to acceptable levels within the TS action statement allowed
outage time. PSE&G performed a detailed root cause evaluation following this
event and concluded that the tank concentration was diluted by both the manner in
which quarterly SLC pump inservice testing was conducted as well as leaking
demineralized water supply valves connected to the system. The inspectors verified
- that PSE&G implemented the LER-stated corrective actions which included
increased frequency tank sampling and inservice test procedure enhancements.
During initial follow up to this event, the inspectors noted poor internal
communications in that chemistry technicians did not inform plant operators of the
.
out-of-specification tank concentration until two and one half hours after discovery.
This delayed operator awareness of the need for a four-hour 10 CFR 50.72 report
and entry into an eight hour allowed outage time for SLC TS action statement.
Boron concentration was properly restored before expiration of the allowed outage
time.
.
A
v , -w :
.
.
6
08,3 (Closed) LER 50-354/97-012.: engineered safety feature actuation - unplanned
manual scram following a relay malfunction in the "A" phase main generator step-up
transformer. This event was described in section 04.1 of this report. The
inspectors verified that the f aulty relay was replaced and that the transformer was
thoroughly inspected for damage following the event. No new issues were revealed
by this LER.
II. Maintenance
M1 Conduct of Maintenance
M 1.1 Technical Soecification SurveilMnce Testina Observations
a. Inspection Scone (61726,71707)
The inspectors observed the conduct of several TS required surveillance tests during
the report period, including:
- "B" emergency diesel generator 24-hour run with hot restart
- torus to-drywell vacuum decay test
, * high pressure coolant injection (HPCI) system inservice test
l * rod block monitor channel calibrat on for single recirculation loop operation
l * reactor mode switch functional test
- reactor core alteration surveillances (ono-rod-out interlock test, etc.)
Implementing procedures and FSAR system descriptions were resiewed in forming
this assessment,
b. Observations and Findinas
in general, the inspectors observed ef fective inter-departmental coordination and
communication during surveillance testing. Pre-jobs briefings, especially for the
torus-to-drywell vacuum decay test, were thorough and followed a scripted format
to ensure all contingencies were addressed. The quality of the various
implementing procedures was good in that all TS acceptance criteria were listed and
necessary plant condition prerequisites were established. Operators and technicians
demonstrated proficiency with the governing procedures.
l
Appropriately conservative decisiori making was noted with respect to the reactor
mode switch surveillance. Specifically, a computer-generahd alarm indication
associated with a reactor protection system channel f ailed to change state during
the test, which the procedure listed as one of the acceptance criteria. However, all
other indications were as expected and the noted alarm point was not required per
TS. In spite of the belief that the mode switch was in f act demonstrated to be
l operable, the shif t supervisor declared the switch inoperable (and implemented
appropriate actions) until the computer point discrepancy was resolved.
l
e
l
7
Good clanning and coordination were noted prior to toe HPCI inservice test, but test
completion was hampered by pour communications between contro: room and field
operators. Specifically, because of background noise in the HPCI room and poor
portable radio equipment performance, delays in satisf actory test completion were
experienced which required a system trip when suppression pool temperature
reached its TS maximum ailowable value,
c. Conclusions
Technical specification required surveillance testing was usually conducted
effectively. Governing procedures were of good quality and were properly followed.
Pre-job t,riefings were thorough and demonstrated appropriate coordination for test
completion.
M2 Maintenance and Material Condition of Facilities and Equipment
M2.1 Station Housekeeoina
The inspectors conducted numerous tours of the Hope Creek facility during the
report period and observed generally adequate plant housekeeping and cleanliness.
Detailed walkdowns in the torus, the drywell, and the main steam tunnel indicated
! acceptable material conditions in those areas. However, the inspectors identified
deficiencies as well. Specifically, several unsecured ladders were discovered
throughout the plant, as well as 55 gallon drums, tools boxes, and test carts. All of
these issues were resolved after station supervision was notified. Additionally,
i,cueral self-contained breathing apparatus used by fire protection personnel were
obserad lying on the floor near a 7 KV motor control center two days after an
event involving a railed condensate pump circuit breaker, in part because of the
increased number of station work activities due to the unit outage, the inspectors
concluded that the state of housekeeping and cleanliness had declined during the
period.
ML2 Inservice inspection
a. trLsoection Scone (73753)
This inspection was conducted to determine whether the inservice inspection (ISI)
of Class 1,2, and 3 pressure retaining components was being performed in
accordance with technical specifications (TS), and Section XI of the American
Society of Mechanical Engineers (ASME) Code,
b. Qbiervations and Findinas
Hope Creek Generating Station (HCGS) TS 4.0.5 prescribes the surveillance
requirements for ISI and testing of ASME Code Class 1,2, and 3 components as
required by 10 CFR 50.55a and ASME Code Section XI. 'ICGS was committed to
ASME Code Section XI,1983 edition, through Summer 1983 addenda, and to
inspection program B of subsection lWA-2400, inspection Intervals. The inspection
!
!
l
l. l
.
.
8
program required that the first inspection interval be completed ten years following
initial unit commercial operation. This inspection was conducted during the third
and final refueling outage in the third period of the first ten year interval.
The ISI Program requirements for HCGS were contained in a document titled, Long
Term Plan, First 10 yr. Interval, Revision 0, dated July 1,1987, developed by
Southwest Research Institute. Plant procedure NC.NA-AP.ZZ.0027(Q)- Rev. 2,
inservice Inspection Program, contained the requirements and responsibilities for the
control and implementation of the ISI Program. The procedure clearly delineated the
responsibility for the program implementation and established the contents of the ISI
long term plan. The procedure provided a broad scope for program implementation
and specified the various other procedures that contained the detailed information
on how to implement the various processes, it prescribed the processes for
addressing program submittal to the NRC, establishing frequency of examinations
and test activities, conducting augmented examinations, reporting requirements,
controlling work and maintaining records.
Review of procedures (73052)
The insnector reviewed the following procedures to ascertain whether they were in
compliance with TS, ASME Code, and Updated Pnr.1 Safety Analysis Report
(UFSAR) commitments:
- NC.NA AP.ZZ 0027(Q)- Rev. 2, inservice inspection Program.
This procedure identified the requirements for the control and implementation
of the ISI program.
- SH.RA-AP ZZ-0101(Q)- Rev. 7, Control and Coordination of NDE Activities.
This procedure contained the requirements for the control and coordination
of nondestructive examinations (NDE) of mechanical components and piping
at Hope Creek.
e HC.SS-IS.ZZ-0006(Q)- Rev. O, Visual VT-2 Examination of Nuclear Class 1
Systems.
This procedure provided general guidance on performing VT-2 of Class 1
systems during system leakcge tests,
o HC.RA-IS.ZZ-0007(Q)- Rev. 2, Visual VT-2 Examination of Nuclear Class 2
and 3 Systems.
This procedure provided the instructions necessary to accomplish VT-2
examination of Nuclear Class 2 and 3 systems during system inservice or
functional tests.
.
..
9
The procedures implemented ASME requirements and were up to date for all
currently approved code relief requests. By Safety Evaluation Report (SER) dated
March 17,1995, the NRC approved Hope Creek's requast to use the provisions of
Code Casa N-4981, " Alternative Rules for 10-year System Hydrostatic Testing for
Class 1,2,and 3 Systems." This allowed the licensee to conduct a system
pressure / leak test instead of a hydrostatic test at the end of the outage. Based on
this, the licensee's use of the ibove referenced leak test procedures would be
appropriate following the outage. However, the procedure for conducting VT-2
examination of class 2 and 3 systems would have to be revised when relief request
dated July 15,1997 is epproved. This request asked for relief from the required i-
hour pressure hold time for insulated systems during system pressure tests. The
licensee was aware of this and indicated that upon receiving the NRC approval of
the request, the necessary changes would be made to the procedure.
Observation of Non-Destuctive Examination (NDE) Activities
The inspector observed ongoing NDE activities. During each observation, activities
were performed by qualified individ:;31s using approved procedures. Portions of the
following activities were observed:
- Liquid Penetrant Test (PT) of core spray pump plate to pipe weld, CP 206-
CSPW4.
This surface examination was conducted with an approved work order and in
accordance with GE Procedure, PT-HPK-100V1, Rev. O, Procedure for Liquid
Penetrant Examination. The Authorized Nuclear inspector (ANI)
demonstration / process qualification for the process was accomplished prior
to its use in the field as required. The technicians showed good questioning
attitude when, as a result of minor indications observed, they opted to re-
perform the examination.
- Ultrasonic Test (UT) of welds on the reactor water cleanup pipe (lines 1-BG-
6DBA-001 and i dG-6DBC-002).
This volumetric examination was conducted with an approved work order.
Prior to conducting the UT in the field, the technicians completed the block
calibration of the instruments as required.
The inspector noted no discrepancies with the portions of field activities observed.
Most field activities were conducted by level ll technicians with all data subject to
review and approval by level lli examiner. The inspector reviewed the qualifications
of some of the NDE technicians to ascertain whether NDE activities were being
conducted by qualified individuals in accordance with the ASME code. There were
four qualified NDE Level 111 examiners at Hope Creek. TNy were supported by NDE
level 111 contractors (General Electric) during the refueling outage. The inspecto?
reviewed the NDE certification for the four HCGS level lli examiners. The
certifications were current and reflected that the individuals were trained and
qualified to perform level lli NDE activities.
.
.
10
c. Conclusions
The ISI Frogram was being implemented with an approved plan and in accordance
with ASME Code Section XI. Code cases used had been approved for use as part
of the plan and re'iefs from code requirements that were used had been approved
by the NRC. The ISI engineer who manages the program showed a good
understanding ar.d ownership of the ISI program. Proceduren were implemented
properly by knowiecgeable and technically competent individuals.
M2.3 Snubber Surveillance Proaram
a. inspection Scone (50090. 61726)
The inspector reviewed the implernentation of the snubber surveillance program to
ascertain whether it was being implemented in accordance with the technical
specifications.
b. Observations and Findinas
TS 4.7.5 prescribes the surveillance requirements to demonstrate that snubbers are
operable. The inspector observed portions of the program implementation with
focus on the ongoing snubber replacement project.
As a corrective action to previous industry and plant specific problems (such as the ,
repeated failures of the common RHR shutdown cooling suction line snubbers), the
licensee was in the process of replacing all "PSA" and "E Systems" brand snubbers
(approximately 640) with "Lisega" brand snubbers,
The !nspector reviewed a portion of the design package (4EO-3507, Rev. 0) that
was being used for the snubber replacement effort. The package was thorough and
properly addressed snubber design specifications such that there was no reduction
in the capability of snubbers to perform their design functions. The replacement
snubbers met all previous design specifications (e.g. ASME Code, Section 111 design
specification; functions; load capability; and environmental qualification). The 10
CFR 50.59 safety evaluation used was a previously documented Equivalent
Replacement and Document Update Generic Evaluation." The licensee had
determined that this modification qualified as an equivalent replacement and as such
was encompassed by the previously documented safety evaluation. The inspector
reviewed the safety evaluation and discussed various aspects of it with licensee
personnel. No deficiencies were noted.
The inspector reviewed the applicable UFSAR section and determined that following
the complete replacement of the snubbers, section 3.9.3.4.6.1, Snubbers, would
need to be revised. The section specified that mechanical type snubbers (PSAs)
were used in seismic category I systems inside and outside the primary
containment. At the end of the project, there would not be any mechanical
snubbers in use. The licensee was already aware of this and indicated that the
l
1
.
.
11
necessary UFSAR changes would be processed as part of the requirements for
implementing design changes. No other discrepancy was noted.
The _ inspector observed a bench test of a PSA 1/4 snubbar (OP-RC-204 SH001)
which had just been removed from service. The test was performed as part of the
required TS 10% sampling. TS 4.7.5.e.1 required that at least 10% of the total of
each type of snubber shall be functionally tested either in-place or on a bench. The
test was conducted per Work Order 970915029 and procedure SH.RA ST 22-
0105(O), Rev. O, Snubber Examination and Testing. The test instrument used was
a "Wyle 150" snubber test stand. The certification of calibration of the test stand
was up to date and traceable to the NationalInstitute of Standards and Technology.
The test was conducted satisf actorily and the snubber functioned as designed,
c. Conclusions
.
The licensee implemented the snubber srveillance program in accordance with
technical specifications. The ongoing efforts to upgrade all plant snubbers appeared
technically sound and was aimed at precluding problems praviously experienced
with snubbers. The individuals involved with the testing of the snubber appeared
very knowledgehble and technically competent.
M2,4 "A" Core Sorav Nozzle (NSB) Safe End Weld Pin Hole Leaks
a. Insoection Scone (73753)
The inspector reviewed licensee activities involving the pin-hole leaks identified in
the "A" core spray nozzle (N58) safe end weld.
b. Qbservations and Findinas
On September 19,1997, the licensee identified three pin hole leaks in the " A" core
spray nozzle. The leaks were from the top of the weld connecting the nozzle to the
safe end. The licensee notified the NRC of this situation as required by 10 CFR
50.72 (b) (2) (1) for a degraded /unanalyzed condition (Event Number 32962). The
plant had been shutdown for a refueling nutage when the leaks were identified.
The nozzle / safe end configuration consists of the followLg: the nozzle from the
reactor vessel welded to the safe end; the safe end (inconel SB 166 material)
welded to the safe end extension (Low Alloy SA 508 Cl 1), which is welded to the
injection piping. A thermal sleeve lining in the nozzle is welded to the safe end.
The nozzle / safe end and safe end/ safe end extension welds consisted of inconel
182 (Ni-Cr-Fe) material. The SB 166 safe ends were installed in 1982, as a
rerlacement for stainless steel material.
Initial PSE&G efforts to obtain a clear definition of the extent and nature of the flaw
were unsuccessful. However, using a " Smart 2000" ultrasonic testing (UT) system,
l
the licensee was able to identify a volumetric crack. There were three pinholes
l within 1.5 inches of each other, with the largest being about 1/16 inch.
_
_ _ . _ _ _ . - . _ _ _ _ ..
[-
m-
-12
The nozzle was examined as part of the ISI program last refueling outage (1995)
and was not scheduled for inspection this outage. When the crack appeared to be
intra Granular Stress Corresion Cracking (IGSCC), the licensee retrieved the 1995
examination data of the weld area for further and independent reviews. In 1995,
the examination method was UT and, at that time, the licensee did not identify any
flaws in this particular area of the nozzle. Hcwever, a re review of that data
revealed that there was an indication that should have been identified. A pattern
that could have been indicative of a flaw was found to be present. The licensee
organized two teams to follow up on this issue. One team was tasked with
conducting a root cause investigation and the other was tasked with determining
the most appropriate repair methodology. The licensee conducted a review of
previous examination data for all other nozzles in an attempt to determine the
potential extent of the problem.--
.
At the end of this inspection, the licensee had not reached a conclusion as to the
nature of the flaw causing the through wallleak, however, their preliminary
determination was that tlie crack was lGSCC-based on the following: industry
experience with IGSCC in this particuiar location and environment, previous NDE
data, previous repairs of the nozzle safe ends, and the type of material (182)in the
nozzle weld. PSE&G continued to conduct more detailed reviews using independent
contractors. For repairs, the licensee was pursuing the " Weld Overlay" method.
This procesu would require NRC approval prior to implementation. This item
remains open pending completion of NRC's review of the circumstances that caused
the flaw to remain undetected until it resulted in a through wall leak from reactor
coolant system pressure boundary. (URI 50 354/97-07 02).
"
c. Conclusions
The licensee demonstrated good efforts at addressing the core spray nozzle through
wall cracking problem once it was identified. Good program management was
demonstrated by the formation of two independert teams to concurrently conduct
root cause investigation and determine the appropriate repair method. Howevet,
questions remain regarding the apparently inadequate ultrasonic test data analysis
during the last refueling outage which allowed the flaw to remain undetected until it
became self-disclosing.
M4 Maintenance Staff Knowledge and Performance
. M4.1 Maintenance Observationg
7- a. insoection Scoce (62707,92902)
The inspectors directly observed numerous work activities at the station during the
report period, including:
- Struthers-Dunn safety-related relay replacements
-
4- * 10 A-402 4160V Vital Bus undervoltage relay checks
- "B" and "D" station service water corrective maintenance
- * "B" reactor protection system motor generator work
,~
.
.
13
Additionally, several work orders and maintenroce procedures were independently
reviewed to assass their quality and completion. Action requests involving
maintenance activities were evaluated to determine whether adverse trends were
evident in work qua!ity, procedure compliance, and supervisory oversight. The
inspectors conducted frequent interviews with maintenance supervision and QA
inspectors in formulating this assessment.
b. Observations and Firgraga.
In recognition of recent poor work performance, PSE&G management conducted
several site wide " stand down" meetings with maintenance personnel just prior to
the refueling outage to reinforce expectations with regard to procedure comoliance,
attention to detail, and self checking. Outage contractor technicians were present
at these meetings. However, the inspectors judged that these sessions achieved
only limited success in meeting their objectives. Specifically, while the inspectors
did observe some maintenance activities during which good practices were evident
(e.g. Struthers Dunn relay work, rod block monitor channel calibrations), a large
number of inspector and QA identified issues were raised which indicated either
poor workmanship or lack of ,;rocedure adherence. Issues involving inadequate
oversight of contract workers were apparent as well.
C' September 18,1997, the inspectors observed relay department technicians
orforming undervoltage devu calibrations on the 10 A-402 vital bus in
accordance with procedu e HC.MD-ST.PB-0010(O). This procedure requires
verbatim step-by-step compliance. However, while performing the testing as
writteri, the technicians observed that a relay under test did not respond as
expected, and decided to deviate from the established sequence of steps in the
procedure to complete the activity. The inspectors questioned the technician's
ections; the worke> s did not consider their actions contrary to procedural
requirensents. During subsequent inspector discussions with the workers'
supervi. , that individual recognized the workers' actions to be contrary to
procedure compilance expectations. Corrective actions for this issue involved
additional "standdown" sessions for relay departroent technicians and individual
disciplinary actions.
On October 4,1997, the inspectors reviewed a completed work order package,
including applicable procedures, for reactor feed flow transmitter calibration checks.
The inspectors discovered several discrepancies .n the attached documentation,
including missed signature approvals for several activities, in one case, technicians
determined that a flow trant,mitter required re-calibration. According the work order
and the attached instrument calibration data card, the calibration was completed
successfully. However, the sections of the procedure (HC.lC-DC.ZZ-0030 (01)
necessary to conduct the calibration were marked as "not applicable," indicating
that they had not been used. Several other portions of the procedure were also not
compieted, including a listing of the test equipment used, transmitter reassembly
instructions, and supervisory review and approvals. The inspectors determined that
this issue, along velth the relay technician compliance issue described above, were
two examples of f ailures to implement required maintenance procedures, which was
a violation of technical specification 6.8.1.a. fVIO 50 354/97 07 03)
!
.
.
14
Several other instances of poor maintenance quality and procedure compliance,
largely involving contracted wrak, were also self or QA identified. For example,
while in the process of removing blocking tags, equipment operators determined
that control rod hydraulic control unit maintenance performed by contractor
personnel was not yet complete even though the associated work orders had been
signed off QA inspectors noted that verification hold points in a service water
piping repair procedure and a reactor vessel disassembly procedure were missed.
Contract technicians bent a source range nuclear instrument guide tube during an
under vessel control rod drive mechanism replacement. An underwater camera was
inadvertently drawn into a reactor jet pump inlet during in vesselinspections. The
"A" reactor water cleanup pump was inoperable for more than four weeks while
maintenance technicians attempted to correct pump sealleakage and motar
vibration probit,ms. This latter issue resulted in increased radiation expostres and
extended operation with degraded reactor Gemistry.
The inspectors discussed each of these events with station management, and
determined that they were fully cognizant of the issues involved and were devising
and implementing corrective measures to address the immediate concerns as well
as minimize the potential for future occurrences. Maintenance management
recognized the need for increased oversight of contract workers, and planned to
devote more supervisory resources toward this effort.
c. Conclusions
In spite of proactive measures by PSE&G management to reinforce expectations
regarding maintenance procedure adherence and attention-to-detail, severalissues
involving viniations and poor work quality were identified. Supervisory o.orsight of
contract maintenance technic!ans was weak.
M7 Quality Assurance in Maintenance Activities
MZd Problem Identification in Maintenance
a. insoection Scope (62707)
The inspectors evaluated the effectiveness of PSE&G's controle in identifying and
resolving problems in maintenance by reviewing corrective action program
performance indicators, action requests, and root cause analyses, and by
interviewing various plant staff and supervision,
b. Observations and Findinag
PSE&G personnel nearly doubled the initiation rate of action requests since the
beginnirg of the refueling outage, which began three weeks into the report period.
The inspectors attributed this increase to the greater volume of work activities being
conducted during the outage, coupled with continued good focus on documenting
" low threshold" concerns. Every new action request was collectively reviewed by
the Hope Creek management team at a daily meeting. Root cause analyses were
_ _ . _ _ _ _ . _ _ . . _ ._. __.-_.__ _ _ ___ _ _ _ - - _ _
i
U
i
!
.. !
15 ;
!
required and completed when necessary. The inspectors observed good critical ,
reviews of each issue, and noted that identified deficient conditions were added to !
the outage work scope as appropriate. Most issues involving poor or inadequate l
maintenance practices were documented by maintenance department personnel, j
though several maintenance related action requests were initiated by quality
'
assurance, operations, and engineering staff.
in spite of the generally good performance, the inspectors noted delays in initiating l
several action requests involving conditions adverse to quality. For example, the
NRC identified issue involving feed flow transmitter procedure adherence (see .
'
section M4.1) was not documented until five days af ter the discrepancy was raised,
-
and only then because of inspector prompting. An issue involving foreign material
dropped into the reactor cavity was not documented until nearly a week after the !
event occurred. A discovery by the maintenance manager conducting field - !
obsorvations that technicians had f ailed to properly " log on" to a work order was !
liot documented in an action request. The inspectors judged that in these and other
vvents station personnel failed to meet PSE&G management expectations for
documenting problems dentified at the station. ;
i
c. Conclusions
PSE&G personnel continued to document deficient conditions at the station with a
-
low threshold for initiation. Plant management demonstrated good reviews of each
new issue and required resolution during the refueling outage when appropriate. !
'
However, severalissues were not documented in a timely manner to ensure timely
, review and corrective action.
M8 ' Miscellaneous Maintenance issues !
MHJ (Closed) VIO 50 354/E 96-014 01013: repeat f ailure of residual heat removal
system shutdown cooling common suction line mechanical snubber. The inspectors ;
reviewed PSE&G's response to this violation and Ndged the stated corrective
actions to be reasonable and complete. The inspectors independently verified that
-increased testing conducted on the subject snubber was completed appropriately
and that no further f ailures were experienced. As such, the inspectors judged that
procedure changes incorporated following the most recent failure, which modified
the method used to place shutdown cooling in service, were effective and
'
prevented recurrence. Additionally, the inspectors noted that this particular snubber
was replaced during the current refueling outag'., with an improved design hydraulic
M8,2 (Closed) VIO 50 354/97-02-01: failure to shut emergency diesel generator (EDG)
cylinder test cocks prior to_ engine operation. - The in:per tors reviewed PSE&G's
June 20,1997 letter which responded to this violation, and judged the stated root
+
cause and corrective actions to be reasonable. PSE AG attributed the cause of the
event to human error, and implemented disciplinary actions as necessary.
Additionally, the inspectors verified that the surveillance test and operating i
> procedures were revised to require independent verification of test cock closure
_p rlor
. to future EDG operation.
!
( ,
l
+
u , -.-. _ _ _ __ ._ _ _ _ ____ _ . . _ - _ _ _ _ . -
.
.
16
MD_3 (Closed) LER 50 354/97-018: engineered safety feature actuations as a result of 1
1
reactor protection system (RPS) motor ger:erator set output breaker trip. PSE&G
determined that the August 3,1997 trip was caused by an age ' elated component
f ailure on the associated breaker underfrequency monitor, which is not safety-
related. This event resulted in the unplanned loss of *B" RPS power, which in turn
caused automatic isolations of the reactor water cleanup system, reactor water
sample system, and steam line drain headers, as well as a "heff" scram. The
inspectors verified that the underfrequency monitor was replaced and appropriately
re tested. Additionally, PSE&G developed a recurring task to periodically perform
functional testing on this device to ensure future reliable operation,
llkl0.gheerina
E1 Conduct of Engineering :
EL1 Plant Modifications - Confinuration Channes
a. Insoection Scope (37550)
The inspectors reviewed four design change packages (DCPs) focusing un design
change program implementation and the licensee's review of the plant configuration
changes to ensure that the criteria delineated in 10 CFR 50.59 anc' 10 CFR 50,
Appendix B, " Design Control," were met. On a sampling basis, specific DCP ltems
reviewed included: scope of design change, basis for the design change,10 CFR
50.59 applicability review and safety evaluation, applicable calculations, drawing
and plant procedure revision, post modification testing, and acceptance criteria.
Some field walkdowns were also conducted,
b. Observations and Findinas
The inspectors review of DCPs 4HE-0245, Revision 0, " Elimination of Flow Signal
Noise on Dead Leg Transmitter"; 4HE 359, Revision 0, " Main Steam Line
Continuous Drain Flow Orifice Resizing for MW Improvement"; 4EC 3644,
Revision 0, " Standby Diesel Generator /CO2 System Modification"; and 4HE-0170,
Revision 0, " Reactor Vessel Wide Range RCIC Level 8 Channels Relay
Replacement"; determined that, in general, engineering had done an acceptable job
to review the issu] and develop a design changs to resolve the issue.
The inspectors found that the DCPs included sufficient documentation to permit
evaluation of the effect of the changes on the design and licensing basis. Where
applicable, the engineer had prepared appropriate calculations to justify acceptability
of the design. No concerns were identified with three calculations reviewed.
Complete review of one of the packages determined that it included necessary
procedure and drawing changes, acceptable installation instructions, and
appropriate retest instructions. Walkdown of tv>o design changes confirmed
acceptability of the installation. Except as noted 'oelow, acceptablo 10 CFR 50.59
applicability reviews had been prepared for the DCPs and proper peer review and
safety operation review committee (SORC) approval, as Jirected by the procedures,
. - .
.
.
17
had been obtained. The inspectors further confirmed that the UFSAR and design
basis documents had been updated af ter the design chenges.
Enaineerina Channe Authorization 4HE-0173
r
This design modification 'nvolved the replacement of Agastat GP series relays in the
reactor core isolation cooling (RCIC) turbine trip logic with Agastat TR Series time
delay relays. The purpose of the time delay was to prevent a RCIC level 8 trip due
to instrument " ringing" caused by oscillatory transmitter signals during reactor
transients. The signals to the timers were derived from the reactor vessel wide
range level 8 channelinstrumentation. PSE&G later evaluated the time delay and
found it to be acceptablo.
To verify that the added time did not impact the system function and other
emergency actuation signals, the inspectors reviewed applicable sections of the
UFSAR. They determined that the signals to other functions initiated by the same
levelinstrumentation were unaffected by *he change. The inspectoru also observed
that the 10 CFR 50.59 applicability review form, prepared by the licensee to
determine whether a 10 CFR 50.59 safety evaluation applied, did not include a
review of Section 7.4.1.1 of the UFSAR which describes the functions and
operation of the RCIC system. This section also includes a flow control diagram
that should have been revised to reflect the design change.
Because a review of the above UFSAR section was not done, PSE&G engineering
improperly answered "no" to the question of whether the modification changed the
facility as described in the FSAR and failed to perform a safety evaluation to ensere
that the change did not involve an unreviewed safety question. A peer and
approver review of the applicability review form also f ailed to recognize this
deficiency. This is a violation of 10 CFR 50.59 requirements. (VIO 50 354/97 07-
04)
c. Conclusions
The program for designing and installing configuration changes to plant systems
was acceptable. Howaver, inadequate review of design documentation prior to
implementation of a reat.or core isolation cooling system modifica! ion resulted in a
violation of 10 CFR 50.59 requirements in that no written safety evah'ation was
performed for the change as required. This failure was also an example d*weak
implementation of the peer review process.
.. . _ .
.
l
6
18
-
3
'
T2 Engineer;ng Support of Facilities and Equipment
'
W Enaineerina Suncort of Plant Operations and Maintenanca
-
gs a. ln}pfstion Scope (37551,62707,71707)
"* The inspectors evaluated engineering department technical support following plant
events and during equipment maintenance activities. Reviews of design change
package implementation during the refueling outage were conducted. Root cause
analyses performed by engineering personnelin support of identified degraded
conditions were assessed. Additionally, the inspectors compared UFSAR system
descriptions and design bases information to actual plant conditions.
b. Observc1lqns and Findinas
The inspectors observed good system enginoering involvement in troubleshooting
activities following plant events and other adverse conditions. For example, system
engineers actively participated in the resolution of the September 2,1997 "A"
reactor recirct,lation pump trip event and the condition evaluation of the "A" main
output step up transformer following the September 10,1997 manual rea:: tor
scram. The inspectors also noted good follow up and evaluation of increased
unidentified drywellleakage rates during the last ten days of the operating cycle.
Engineering department response following the identification of the through wall
leak on the "A" core spray reactor pressure vessel nozzle was appropriate.
PSE&G experienced several repeat equipment f ailures during the report period,
which indicated that past engineering resolution to degraded conditions were less
than fully effective. The rod sequence control system was declared inoperable
following a repeat f ailure of the self test logic on September 9,1997. A f ailed
"inop/ inhibit" switch on an intermediate range neutron monitor (IRM) channel
caused an unexpected half scram signal, nearly identical to previous occurrence on
a different IRM channel earlier in 1997. The scope of corrective actions for the
earlier event did not include inop/ inhibit switches on the redundant IRM channels.
The inspectors reviewed a root cause evaluation stemming from an action request
which documented problems experienced during maintenance on the "A" reactor
water cleanup pump. This planned five day on-line maintenance activity required
more than one month to complete, and had an observable negative impact on
reactor water chemistry. The root cause evaluation was very critical of both
engincoring and maintenance department performance, and highlighted several
problems with inter departmental coordination, data collection and evaluation,
cuntingency planning, and troubleshooting. The inspectors noted that eight
separate corrective actions were recommended, which addressed both specific and
generic concerns raised in the evaluation.
PSE&G engineering developed approximately thirty design changes for
implementation during the refueling outage, a large percentage of which were
intended to eliminate long-standing equipment deficiencies, including temporary
- .-- -- . - - _ - - . .. . . - - - _ - - - .
. i
e
19
modifications, oprability determinations, and operator workarounds. Examples
included reactor recirculation pump seal upgrades, safety related relay
replacements, rod sequence control and rod worth minimizer modifications, and
service water pump discharge strainer improvements. However, several of the
change packages were of poor quality, as evidenced by a high number of in process
package modifications. Based on inspector discussions with engineering and t
maintenance personnel, many of these self identified package deficiencies could ,
4
have been prevented had more detailed reviews been conducted during :
development and had more time been allotted for pre implementation walkdowns. -
Several packages, like the emergency diesel generator (EDG) relay replacement
work, did not specify adequate retests. This package required only that the
monthly EDG surveillance run be conducted, in spite of the fact that many of the
relay functions affected by the modification would not be verified by this test.
,
c. .C.pnclusions
System engineering ef_ forts in identifying and resolving emergent equipment f ailures ,
'
were prompt and effective. The large scope of design change work scheduled for
completion during the refueling outage demonstrated appropriate management focus
on resolution of long standing equipment deficiencies. However, several examples
of repeat equipment f ailures, extended system maintenance activities, and design
change package deficiencies highlighted weakness in the quality of engineering
support. An internal review of an extended reactor water cleanup pump outage ,
was thorough and self critical.
E2J Enaineerina Involvement in Site Activitin
a. Inspection Scoce f37550)
,
The inspectors e faluated the effectiveness of the engineering staff in supporting
plant needs through a review of licensee event reports (LERs) and root cause
analysis reports in the area of engineering, interviews of responsible engineering
personnel, and an assessment of the quality of the analyses performed to resolve
the reviewed issues,
b. Observations and Findinas
Thermal Dearadation of Struthers Dunn Relavs
On April 7,1997, PSE&G notified the NRC via LER 50-354/97 07of the results of
an evaluation they had performed regarding an increase in Struthers Dunn relay
failures: they had experienced approximately 40 relay failures between 1988 and
1996 and 24 in the last three years. PSE&G concluded that the relay f ailures were
due to thermally induced aging of a magnetic vinyl plastic used as bearing pad
material. As determined by licensee engineering personnel, the failure of the
bearing pad affects the alignment between the armature and the ac relay coil and
causes rapid oscillatory motion of the relay armature and contacts. This rapid
motion eventually results in relay failure. The observed degradation involved
normally-energized relays.
_ __ _ _ - - _ _ __ _ _ _ _ _ _ _ _
.
.
20
To determine the extent of relay degradation, the licensee conducted walkdowns of
51 Class 1E panels containing Struthers Dunn 219NE series relays. They identified
a total of 48 degraded relays. PSE&G considered a relay to be degraded if pieces of
bearing pad material were visible at the bottom of the relay case. All degraded
relays, except three, were in a mild environment; almost all of the mild environment
relays were mounted in the remote shutdown panel.
The root cause analysis performed by the engineering team was thorough with
many background details, including vendor assessments. A number of corrective
actions resulted and, as of the end of the inspection, all degraded relays had been
replaced Tha inspectors, however, loontified several concerns:
- The walkdown considered degraded only those relays having visible debris in
the relay case. The basis for this was an understanding that a degraded
relay would fitst begin to chatter and thrt several days or weeks of
continuous chottering would be required before the relay f ailed dao to
contact sealing (arcing) or spring f atigue.
The inspectors judged this reasoning to be less than acceptable, because it
did not consider the effect of a seismic event on those relays that showed
no evidence of " degradation" and, therefore, it did not evaluate their ability
to perform their safety functions. While the relays in question were in a mild
environment, the number of f ailures also in panels considered to opercto at a
lower temperature indicated that aging was pervasive. Therefore, the
decision to conduct weekly panel walkdowns and inspect for buzzmg was
similarly less than acceptable. For these relays, the adequacy of the
walkdown and the service life of the relays is unresolved pending appropriate
evaluation by the licensee ano review by the NRC (URI 50 354/97 07 05)
- For five Struthen, Dunn relays identified during the plant walkdown that were
located in a harsh environment, calculation No. 6TRDUN ARRH-001, dated
April 30,1937, determined that the normally onergized qualified life was
15.4 years. For this calculation, engineering took coil and pad material
temperature readings; then, they used these measured temperatures to
calculate the relay qualified life using the Arrhenius method of extrapolation.
The inspectors did not identify any errors in the calculation itself and had no
comments regarding the calculation method. However, considering that
three of the five relays in question were " degraded" af ter only eleven years
and without taking into consideration the post-accident operability period,
the inspectors concluded that the inputs (activation energy and/or operating
temperature) to the equation had to be incorrect, if the temperature-aging
relationship postulated by Arrhenius had to hold true. The calculat;on results
were not adjusted to reflect observed conditions and no explanation for the
difforence between calculated and actual qualified life was provided. Four of
the five relays in harsh environment were replaced. An appropriate
justification had been prepared for not replacing the fif th relay,
<
I
.
21
- The equipment evaluation summary sheet (EESS), revised to add the five
harsh environment relays, established the qualified life of the same relays at
twelve years. For other Struthers Dunn relays in the same panels, the EESS
specified the qualified life to be 5.04 years. The EESS provided no
explanation for the difference in qualified life.
The f ailure to include the noted five harsh environment relays in the environmental
qualification master list and to replace them af ter their qualified life constitutes a
vio'ation of 10CFR 50.49 requirements. (VIO 50 354/97 07 06)
'
While reviewing the EESS for the Struthers Dunn relays, the inspectors observed
that the postulated post accident relative humidity (RH) environment was 100%,
but that the relays were qualified for only 90 98% RH. PSE&G accepted the RH
qualification " based on analysis of test instrument accuracy" (Note 5 of the EESS).
Because instrument accuracy can also be nonconservative, the inspector questioned
the adequacy of the justification. Following the inspection, the licensee provided
additional documentation that demonstrated qualification of the relays in a
postulated 100% RH environment.
Additional observations regarding relays la both harsh and mild environments are
included in Section E2.3 below,
c. Conclusi2DA
The inspector concluded that the licensee took generally acceptable actions to
address nonroutine plant events. When a root cause analysis was prepared, the
analysis was thorough and detailed. Not all conclusions were conservative,
however, as in the casa of the Struthers Dunn relays in a mild environment.
For Struthers-Dunn relays in harsh environment, the f ailure to include five relays in
the EQ list and to replace them at the end of the qualified life resulted in a violation
of 10 CFR 50.49. Also, the use of a calculated relay qualified life without
reconciling the difference with actual data indicated an excessive reliance on a
theoretical life extension method that is highly dependent on the correct selection of
independent variables.
The delay in initiating a relay f ailure analysis (24 relay replacements in three years)
indicated a weakness in the program for monitoring the performance of safety-
related components in a mild environment.
EL3 Service Life of Relavs in Mild Environment
a. Lningetion Scone (37550)
Information Notice (IN) No. 84 20, " Service Life of Relays in Safety Related
Systems," advised licensees that the service life of all relays in the normally
energized state is significantly shorter than when used in a cycled or normally
doenergized application. The notice also advised that preventative maintenance
4
.
22
programs should recognize the application-dependent service life of the relays and
that it may be prudent to increase the frequency of surveillance activities of those
systems where tne surveillance interval was not smallin comparison with the
service life of the relays in those systems. Because LER 50-354/97 071ndicated an
increase in f ailures of normally energized Struthers Dunn relays, the purpose of this
portion of the inspection was to evaluate the actions taken by the licensee to
address the service life of normally energized relays other than Struthers Dunn.
b. Observations and Findinog
Hope Creek was still under construction when IN 84 20 was issued. However,in a
letter to PSE&G, dated May 13,1986, the Architect Engineer (AE) provided a table
of the normally energized relays used at Hope Creek and their service life. A review
of this letter and attached table showed that, for some relays, the service life was
very short. For instance, the service life of Potter & Brumfield,24 Vdc MDR relays
and for Telemechanique J13 and J14 relays the service lives were stated to be 2.2
and 3.77 years, respectively. The letter also stated that General Electric (GE) had
extended the life of the Agastat FGP and EGP normally energized relays from the
4.5 years specified in the IN to 5.9 and 6.6 years, depending on the ambient
temperature inside the relay panels.
Apparently, because of " human error and organizational / programmatic problems,"
as stated in Action Request (AR) 00970218207,the AE's recommendations were
not put into action. The AR also stated that the relay changes were to be
accomplished "through the Mild EQ program which was canceled. No group inside
or outside of HC took ownership of the issue."
Potter & Brumfield Relay.g
For the Potter & Brumfield relays, the licensee had neither recalculated the service
life nor taken measures to replace them after the expiration of the service life that
had been calculated by GE. However, when the inspectors identified the concern,
they secalculated the service life of the relays and determined it to be 12.18 years.
The revised service life was based on: (1) the relays being energized 25% of the
time: (2) an operating temperature of 85" F; and 3) a coil temperature rise (while
the relay is energized) of 60* C.
The inspectors reviewed the preliminary calculation provided by the licensee and
considered the results acceptable, primarily based on the conservative value of
energization f actor (25%). Data provided by the licensee indicated this to be less
than 20% since reactor criticality in April 1986. However, the inspectors made the
following observations:
- Since the relays are energized wher: the reactor is shutdown, the licensee
should estimate the energization period prior to commercial operation. This
could greatly impact the replacement date.
'
.
.
23
- The operating temperature of the relay was assumed to be 85* F. The .
licensee should confirm the reasonableness of this value and justify the
difference between the values used in the calculation and those used by
General Electric in their qualification report.
- Since the temperature rise (calculated by resistance measurement) changes
with the operating temperature, its value should be recalculated more
accurately.
Telemechanlaue J10 and J13 Relavs
The licensee stated that Telemechanique (also known as Gould and ITE) J10 and
J13 relays are used only in the emergency diesel generator panels. These panels
contain 42 normally energized relays each.
On April 3,1992, the NRC lasued IN 92 27 to inform the industry about thermally-
Induced aging f ailures of ITE/Gould J10 relays at Seabrook and Millstone. At
Millstone the relays had been in operation for seven years. On March 21,1997, the
NRC issued a supplement to this IN describing similar failures at the Beaver Valley
- plant. At Beaver Valley, the relays had been in operation for nine years.
in an internal memorandum dated January 24,1996, PSE&G engineering stated
they had submitted to seismic tests two naturally aged relays selected among those
showing most degradation (discoloration). These tests showed that the relays, in '
their current aging condition, would have been capable of performing their safety
function during and following a seismic event, had they been required to do so, The
memorandum also stated that the relays would be submitted to accelerated aging
equivalent to one plant operational cycle and retested seismically to prove their
ability to perform their function through the 1997 refueling shutdown. This was
done and because of the successful results, the ilcensee eventually decided to
further extend the service life of these relays for one additional operational cycle.
Nuclear Environmental Qualification Report No. 96030.1, Revision 0, prepared by
Farwell and Hendricks (F&H), shows that the relays were subjected to accelerated
aging for 182 hours0.00211 days <br />0.0506 hours <br />3.009259e-4 weeks <br />6.9251e-5 months <br /> at 140" C and that the aging time was calculated, using the i
'
Arrhenius equation, assuming an additional required service life of 21 months at an
ambient temperature of 50* C. The F&H calculation also assumed a coil
temperature rise of 28* C. However, this value was incorrect because in the above
referenced memorandum the licensee stated that they had taken temperature rise
measurements and, with one relay energized in the center of three relays ganged
together, they liad measured the temperature rise to be 92.3* F, equivalent to -
51.28* C -
To be conservative, the licensee should have measured the temperature rise of the
center relay with all three relays energized, unless an analysis demonstrated that the
physical configuration reflected the tested configuration. However, as a minimum
the F&H calculation should have used the meast, red temperature. If a temperature
rise of 51.28" C was used in lieu of the 28* C in the Arrhenius equation, the 182
1
(
.
.
24
hours accelerated aging used by F&H would have yielded only 97 days of operating
,
time, not 21 months. The licensee did not address this discrepancy. Because the
relays are in a mild environment and underwent complete seismic testing (five
operating basis and one shutdown earthquakes) three times, the inspectors had no
immedSte safety concern with these relays. However, the use of the lower
temperature rise in the life extension calculations was inappropriate.
Anastat Relavs in a Mild Environment
Performance improvement Request (PIR) 951120103lndicated that there were
1197 Agastat relays at Hope Creek. Ninety one of these wure in harsh environment
and covered by the plant equipment qualification program. The remainder, of which
approximately 330 were normally energized, were located in mild environment
zones.
As stated previously, IN 84 2Olnformed licensees that the service life of normally
energized GP/EGP type relays is 4.5 years based on GE test data. For GE supplied
Agastat relays, GE extended the service life to approximately 6.5 years. This was
done through either an evaluation of the ambient temperature to which they are
exposed, as stated in the May 1986, AE letter to the licensee, or through a material
analysis of relays that had been in service for four years, as related in corrective
action no. 2 of PIR 970218207. PSE&G stated that, for those relays, replacement
occurred at 6.5 year intervals and that recurring tasks (to replace the relsys at the
stated interval) existed. For non-GE supplied normally energized relays, PSE&G set
the service life at 10 years, apparently based on the service life that had been
established by the AE for the EGP relays in 125 Vdc and 250 Vdc battery chargers.
No recurring tasks existed to replace most of these relays,
in November 1995, approximately six months prior to the expiration of the relays
servicc life, the licenseo found that, for certain Agastat relays, no recurring task
existed. The ensuing PIR (No. 9511200103)tosulted in a detailed review by the
licensee of the Agastat relay service life issue. Originally, the service life was
calculated to be 11 years, placing the end of the service life on April 16,1997. No
formal calculation was performed, however, for the 11 years. In an attempt to
extend the service life further, the licensee decided to conduct life extension tests.-
For this purpose, the licensee se!ected two E7000 and two EGP type relays that
had been naturally aged in tb plant and sent them to a laboratory for testing. It
was not clear whether these relays were representative of the lot, also considering
that one of the EGP relayrs later was found to be new.
The inspectors' review of the test report determined that the E7000 relays passed
the seismic test after an equivalent of 27 months of aging. The aged EGP relay,
however, with an equivalent of only seven months aging, f ailsd the seismic test.
Because the design basis earthquake is an event during and after which equipment
in mild environment must remain functional, its f ailure after a relatively small
amount of additional aging time was significant because it indicated the possibility
that the relay might have not passed the same test, even in its preaged state.
Because the relay f ailed to withstand the seismic test, all normolly energized relays
were replaced after eleven years.
1
i
_ ._ _ _ _ . _ _ _ ._ _ _ _ _ . ___ _ _ _ _ .. __
e
!
,
'
25
i
in determining the amount of accelerated aging required to extend the service life of
the EGP relays specified by the licensee, F&H used a temperature rise of only 32' F
(17.7* C). This same value was also used by the licensee in service life calcelation 3
i
Agestat.ARRH 002. However, the accuracy of this value is doubtful because it is
consistent neither with temperature rise data availablo to the licensee for other
-
relays, not with measurements taken by GE and others for the same relay. For
instance, (1) the temperature rise measured by the licensee for an energized J10
relay was 85' F; (2) qualification re. cords provided by the licensee for the 115 Vac l
ant.i 24 Vdc MDR relay showed that, at normal ambient temperature, the
temperature rises were 72 and 108" F, respectively; and (3) licensee performed
service life calculations for Struthers DL.nn relays ue,ed a temperature rise of 105' F.
Measurements taken by GE and, more recently, by another licensee, show that the
enargized Agastat EGP relay coil temperature rise is approximately 100* F. The use !
of a realistic temperature rise in the aging calculation greatly reduces the life !
expectancy of the energized relays.
Regarding the E7000 series relays, F&H concluded that the service life could be
extended by 27 months. In the accelerated aging time calculation, F&H used a
temperature rise value provided by PSE&G._ This value,34' F (18.9* C), appeared to
be similarly unrealistic and not in conformity with expected values. If the licensee
had used a value 75' F, for instance, which is in the lower range of the heat rise
measurements for J10, MDR and Struthers Dunn relays, the amount of service life
extension they would have achieved through the F&H accelerated aging would have
been less than three months, not 27 months. -
The inspector judged that the f ailure to employ appropriate values in relay service
life calculations was a violation of 10 CFR 50, Appendix B, Criterion Ill, in that
appropriate decign control measures were not established. (VIO 50 354/97 07 07)
Agastat Relavs in Harsh Environment
In conjunction with this review, the inspectors determined that ninety-one E/000 ,
and ETR Agastat timing relays are used in a harsh environment. The ETR relays are
equivalent to the Agastat EGP relays. The applicable EESSs stated that the relays
are required to operate 100 days in a post accident environment with a maximum .
'
teraperature of 148* F, and 100% RH. The EESSs also stated that the relays are
qualified for this environment and that they have a service life of 4.95 years. The >
inspectors review of the EESS identified several concerns:
- The original calculation, Agastat ARRH 001, Revision 0, that evaluated the
qualified life of these relays, apparently no longer exists. Therefore, the
accuracy and bases for the stated qualified life could not be verified. New
calculations showed the service life to be longer than the stated 4.95 years.
However, these calculations assumed a coil temperature rise of 34' F.
Therefore, the resu!ts may not be accurate. 1
- The EESSs stated qualification to greater than 148' F. Howaver, for relays
- that remain energized during and following the postulated accident,
'
qualification may not be demonstrated by the manuf acturer's tests,
t
f
, __,, _ . _ .- , -
- . - , . _- . _ . _ . _ - - ., , _ _ , _ . - - -,
-- _ - - _ _ _ __ . -_ _. _. __. _ _ _ _ _ _
.
.
26
,
According to the Amerace test reports, the relays were aged (in the
' '
deenergized state) at 212' F. In addition, they were subjected to a " hostile"
environment for two hours at 95% RH and at each of nine temperatures
ranging between 40 and 172' F. During the " hostile" environment test, the
'
relays were energized only during functional tests. The relays were
functionally tested at their minimum and maximum design voltages ten times
(five times at each voltage level) during each two hour period.
At an ambient temperature of 148' F, when margin (15' F) and panel and
coil heat rise are considered, the operating temperature of the energized ETR
relays is expected to be approximately 250 255' F,i.e., well above the aging
and " hostile" testing temperatures. For the energized E7000 relay,-
qualification to the specified post accident temperature would be
demonstrated only if the coil heat rise remained below approximately 40' F. ;
i
- Note 3 of the EESSs stated that, although the relays had been quallfled for
only 95% RH, qualification could be extended to 100% RH because " heat of
the current carrying components in the panels will reise their internal
temperature above the surrounding ambient temperature effectively reducing '
the R.H." Because, following an accident, the panel internal temperature is
suddenly increased by approximately 50' F, it was not immediately evident
that the RH within the cabinets would not also be at 100%.
Following the inspection, the licensee determined that the ETR relays had on
energized duty cycle of 10%. In addition, they provided an analysis showing that a
2* F difference between the internal and external temperatures would reduce the
panelinternal RH to 95%. The issue concerning the qualified service life of the
Agastat relays in a harsh environment is unresolved pending appropriate revisions to
the aging calculation. (URI 50 354/97 07 08)
c. Conclusions
,
Less than occeptable judgement was used in the selection of the coil temperature
rise of normally energized, safety related Telemechanique and Agastat relays in a
mild enviranment. As a result, their calculated service life was longer than what
industry axperience (NRC information noticos) supported, Because of the incorrect
temperature rise, life extension tests were similarly unsupported.
Acceptable justification was provided to demonstrate quah'ication of the ETR relays
to the specified post accident environment. However, the qualification of ;he
normally energized E7000 series relays to the post accident environment is
unresolved pending the licensee's confirmation that the coil heat rise is less than
, approximately 40' F.
The licensee acknowledged the findings presented, but disagreed with the
inrpectors' conclusinns regarding the relays located in a mild environment, stating
that an adequate relays monitoring program existed.
_- -
, . _ . _ _ _- __ _
_ _ _ _
. . _ ._ _. -_ _ _ _ _._ _ . . _
v
!
,
.
t
27
E7 Quality Assurance in Engineering Activities
EZd Audits and Assessments
a, inspection Scone (37550)
The inspectors reviewed audits and assesstnents to determine the effectiveness of
the licensee's self assessment program in engineering. ,
b. Observations and Findinas .
s
The inspectors reviewed QA audit 97100, the final report for the SSWS (July 18,
1997), and several Hope Creek quality assessment surveillance reports. The review
indicated a good audit program with good plans in place. The inspectors found that ;
the assessments were broad in scope and presented substantial findings and i
observat!ans including recommendations for those engineering programs or
processes requiring additional management attention. Engineering areas reviewed
included design and control, configuration control, corrective action program, self-
assessment and previous OA audit findings, reactor engineering / fuels, engineering -
cultural assessment, and training and qualification. The inspectors noted that the
findings were clearly stated, directed to the appropriate personnel, and assigned
tracking numbers. In addition, the inspectorc verified appropriate resolution of
selected findings,
c. Conclusion
The inspectors concluded that the OA audit of Hope Creek engineering was good
and provided a good assessment of the engineering programs in place.
E8 Miscellaneous Engineering issues
Egd M,edium Voltaae Circuit Breaker Failure
On September 10,1997, PSE&G notified the NRC Mat the plant had been manually
scrammed due to an inoperable 'A' main phase transformer. The notification also
stated that, during plant restoration following the scram, the 'A' secondary
condensate pump [ circuit breaker] f ailed to trip. Although the secondary
condensate pump and the supply circuit breaker are not safety related components,
the purpose of the inspection followup was to determine the causes of the breaker
f ailure to open on demand and to evaluate potentially generic implications.
The inspectors' walkdown and review of this issue determined the following:
o The failed circuit breaker was manufactured by ITE/Gould and used in a
7.2 kV nonsafety related application,
o The medium voltage (4.16 kV) breakers used in safety-related applications
were also manufactured by ITE/Gould.
L
L .- . .
-
e
.
28
- PSE&G was conducting a root cause analysis of this failure and, although
had not reached definite conclucions, was attributing the breaker f ailure to a
f ailed trip coil.
- In a separato incident, but during the same event, with the bus do energized
a turbine building chiller supply breaker closed on a dead bus and cycled six
times before it finally opened. This breaker f ailuro was being reviewed by
engineering, but apparently it was viewed more as a control circuit issue.
- The proventive maintenance program for safety and nonsafety related
medium voltage circuit breakers is similar and involves primarily the electrical
(high voltage) portion of the breakers, but not the operating mechanism.
Proventivo maintenance was scheduled every 36 months.
- Twenty nine circuit breakers underwent a complete overhaul during the
previous refueling outage and an additional nineteen were scheduled for the
current outage. Some circuit breakers (including the f ailed ones) were never
overhauled since their shipment,i.e. prior to initial plant startup.
- At the time of the inspection no information was available regarding vendor
recommended overhauling scope and schedule.
Because the licensee's analysis was incomplete, the inspectors did not pursue this
issue further. They did, however, express a concern regarding breaker
maintenanco This issue remains open pending completion of the analysis by the
licensee and review by the NRC. (URI 50 354/97 07 09)
E2 (Closed) VIO 50 354/96 04 01: Failure to account for all Bailey solid state logic
module (SSLM) f ailures. The establishment of a reliability program to monitor the
performance of the Bailey 862 SSLMs is a Hope Creek license condition (2.C.5)
requirement. To implement this program, PSE&G issued procedure HC.lC DD.ZZ-
0017(Z), " Bailey Module Reliability Program," which required that work sheats for
all rework, repair, replacements, and/or testing of any type of modules be sent to
the system engineer for review and failure characteristic analysis. During the April
1996 inspection, the NRC determined that PSE&G personnel had not been
consistently complying with the procedural requirements.
To address this issue, PSE&G initiated a root cause analysis. They determined that,
during the first quarter of 1996, work sheets had either not been completed or
delivered to the system engineer on eight of the ten cases. Additional examples of
procedure noncompliancos were identified in 1995 (5 of 19 instances) and 1994 (3
of 21 instances). They concluded that the root cause of their failure to comply with
the procedure requirements was inadequate program monito.ing by maintenanco,
engineering, and quality assurance personnel,
in their response ?o the notice of violation (letter LR N96180, dated July 5,1996),
PSE&G indicated that they would revise the applicable procedures to require system
engineering signaturo prior to closure of the work order, condact appropriate
.
.
29
training, counsel personnelinvolved, and monitor the effectiveness of the corrective
actions.
During the current review, the inspectors examined the results cf root cause
analysis, evaluated the adequacy of the corrective actions, and confirmed that such
actions had been completed. Based on this review, the inspector concluded that
acceptable actions had been teen to address the violation and the item is closed.
The licensee's internal closure of this issue included a reportability review. PSE&G
concluded that the evant was not reportable (as required by license condition 2.F)
because they had established a program to comply with the license condition, but
just f ailed to comply with the procodures that implemented the program. The
inspectors disagreed with PSE&G's conclusions because the effectiveness of the
program relied solely on the personnel complying with the procedural requirements.
PSE&G's review of their past performance in this area identified procedural
noncompliances in each of the thrne years reviewed, especially in 1996 when they
failed to comply with the applicable procedures eight out of ten times. Therefore, a
successful monitoring program and, hence, compliance with the intent of the license
condition could not be claimed.
Because the licensoo originally identified the procedural noncompliance, a notice of
violation was already issued for such noncomM!ance, and acceptable steps were
taken to address the monitoring concern, this minor violstion of the licensee's
reporting requirements is not being cited in accordance with section IV of the NRC
Enforcement Policy. (NC.V 50 354/97 07 10)
EfL;) (Closed) URI 50-354/90 04 02: Control of the Bailey solid state module automatic
tester programs. Testing of Bailey cards is primarily accomplished through the use
of an automatic tester in accordanew with procedure HC.lC-GP.ZZ 0075(O). During
the 1990 audit, the NRC inspectors determined that the tester stored approximately
200 different programs and that the revision of these programs was being
controlled by the plant design change process. The inspectors also determined that
the programs were being backed-up overy six months and stored on site. However,
they were unable to determine the storage location, or the method of storage and
access control of the backed up media.
Following the inspection, the licensee evaluated the NRC concerns regarding
storage of the backed up media and determined that they did not have a procedure
to control the sof tware of the logic module tester. The root cause analysis team
assembled to address this issue examined its significance based en past practices.
They determined that, in the past, the program always had been updated by the
vendor who was responsible for verifying the accuracy of the changes and
maintaining control of the program configuration.
The team also determined that the database, also required by the tester, was
controlled through the plant modification process. Design changes that impacted
the Bailey logic modules were routed to the responsible engineer who would modify
the database. The databare changes, password protected, underwent appropriate
l
l
. . _ _ _ _ _ __ _ __ . _ _ _
.
.
30
design verification. Tests followed hardware and software changes. Periodic tapes
of the tester program were made by the vendor. The database was copied on the ;
tape at the same time.
Based on their review of past practices, tester performance, and test results, the
team concluded that the issue was not a safety concern. They realized, however,
the need for better software configuration control. Toward this end they initiated -
several actions, including: (1) issuance of program and database software as
controlled documents; (2) revision of procedure for programming new chips;
(3) preparation of procedures for changing the tester database and for performing
backup and restoration to and from tapes; and (4) issuance of the Bailey Card ,
Tester Manual as a controlled manual. In addition, they required that all
departments verify that safety related and critical software complied with the
sof tware configuration requirements of procedure NC.NA AP.ZZ 0064, " Software ;
'
and Micro Processor Based Systems."
Based on their review of the root cause analysis and verification that the required
actions had been completed by responsible groups, the inspector concluded that
acceptable actions had been taken to address the NRC concern regarding control of
the tester software. This item is closed.
E32 (ClosedUfl 50 354/96-04-03: Bailey colid state logic module electromagnetic
interference circuit testing. The Bailey cards experienced a high rate of f ailures -
during plant startup that was attributed to electro-mannetic interference (EMI) and
high humidity. To address these issues, the cards were modified to provide a larger
gap between the circuit input traces and to add EMI input buffer circuits. It was not
clear, however, that the modifications were in place during subsequent temperature
'
and seismic' tests.
During the current review, the inspectors reviewed Bailey Controls Report number
OR-3101 A-E93 75, dated October 7,1985, and PSE&G drawing PJ2000-2994,
dated January 17,1985. Based on this review and discussions with licensee
personnel, the inspectors concluded that the SSLM modifications had been installed
prior to the seismic and temperature qualification tests. This followup item is
closed.
ERJi [Qlgjigd) IFl 50 354/96-04-04: UFSAR inconsistencies. While reviewing UFSAR
sections related to the inspection of the Bailey solid state logic modules, the NRC
inspectors observed some inconsistencies between the UFSAR wording and plant
practices, procedures and/or observed parameters. Specific UFSAR sections cited
in the inspection report included Sections 7.1.2.9.2,7.3.1.1.9, and 7.3.1.1.10, and
-
Tables 7.1-2 and 7.1-3.
In their response to the NRC observations (Letter No. LR-N96180, dated July 5,
1996), the licensee confirmed that the UFSAR sections were misleading regarding
methodology of system testing and stated that the sections would be revised
appropriately. Regarding Tables 7.12 and 7.13, PSE&G ctated that they had
reviewed the testability guidance contained n flegulatory Guide (RG) 1.22 and the
4
- , . - - , . , , - - . , - ,-
.. - - . - _
.
. .
.
31
Hope Creek Safety Evaluation Report (SER) and determined that they complied with
Regulatory Position D.d.a, b, c. and d of RG 1.22 and that their compliance with
this regulatory position was documented in Section 7.3.2.5 of Supplement 6 to the
SER.
The insNetors confirmed that the applicable UFSAR sections had been revised and
verified that testing practices conformed with the guidance of the applicable
sections of the RG and SER. This item is closed.
fifLD lypdated) VIO 50 354/96-09-03: Interaction between nonsafety and safety related
components in the standby diesel generator room ventilation system. NRC
inspection report 50 354/96 03 described a concern regarding potentialinteraction
between the nonaafety related fire suppression system and the aafety related
standby diesel generator room ventilation system. Subsequent review of this issue
resulted in the issuance of a notice of violation on December 5,1996.
In their response to the notice of violation, letter No. LR N96436, dated January 9,
1997, PSE&G stated that a temporary modification had been implemented to
disconnect four rionsafety related relays in the ventilation system and that various
options were under consideration for permanent resolution of the circuit interf ace
concern. The NhC found the response insufficient to address the inadvertent
actuation concein and in a letter, dated June 10,1997, requested that the
permanent resolution address the NRC concerns.
As of the end of the inspection PSE&G had not responded to the June 10,1997,
letter. The inspectors, however, learned that the licensee had designed a
modification to be installed during the current refueling outage. This modification,
DCP No. 4EC 3644, entailed: (1) the removal of the fire dampers between the
standby diesel generator and recirculation system vent rooms; (2) the removal of
the fice damper actuation circuitry; (3) the disabling of a portion of the recirculation
f an circuitry associated with the fire suppression system; and (4) the
addition / removal of appropriate appurtenances. The modification,in essence,
expanded the protection zone of the fire suppression system to include the
recirculation system vent room, rendering the isolation of the standby diesel
generator room in the event of a fise unnecessary.
The inspectors reviewed the modification package, including the safety evaluation
and the calculation for required carbon dioxide (CO,) concentration and concluded
that the planned modification was acceptable and would resolva the i..teraction
concerns. A walkdown of the affected areas also indicated that no combustibles
existed in the recirculation vent room to increase the probability of a fire in the
standby diesel generator room.
Prior to the installation of the modification the licensee decided to perform a CO,
discharge test to ensure that the diesel rooms would achieve the required CO,
concentration. Discussions with licensee engineering personnel indicated that a first
test conducted in accordance with special test procedure GH.PI-AP.ZZ-0012(0),
" Total Flooding CO, System Discharge Test," did not achieve the required
. _ . . - - - - - - - . .- - .- - _. - _ _ .
I
ii
,
32
concentration (34%) because, shortly af ter the discharge test was initiated, seals in .;
the CO, header isolation valve failed and began to leak CO,into the upper chamber :
of the valvo. As a result, the valve began to close. The maximum concentration
achieved during this test was 14%. A new Isolat'on valve was installed and plans
were made for a second test.
- A second test, performed on September 10,1997, was considered successful by
the licensee because a maximum CO, concentration of 38% was achieved. .
'
. However, during the test, the increase in diesel room pressure caused the access.
door to blow open. Recordings of pressure versus time showed that the pressure :
bad reached approximstely 15.1 psia when the door blew open, at which time the !
gressure dropped instantly to its normallevel of 14.7 psig.
Discurdons with responsible engineers regardina both events indicated thet limited ,
i
preventive maintenance had been done on the CO, system since the initial plant
startup ami that the diccharge test performed at that time might ha're been done in
r, cold f6om. The 'leensee was investigating the latter event primarily and
a-Notetood some of the implications of the blown open door, for example the
irt. pact of the pressure spike on structuret components, and seals; the ability to
prevent a fira from spreading through vents, drains and f ailed structures; and the
maWmom attainable pressure in a room with limited air leakage.
Bec. use both discharge tests quostioned the licensee's ability to maintain the
rege red CO, concentration in the standby diesel generator room, the inspectors also
questioned: (1) the ability to suppress a fire in the diesel generator rooms with a
'siled discharge valve or failed access door: (2) t..e use of CO,in other fire zones;
(3) the use of other fire suppression agents; and (4) the impact of incomplete CO,
system ma!ntenance on the safety systems protected. During the inspection, the
CO, system was considered inoperable and compensatory measures were in place
to address eventual fire suppression needs. According to licensee responsible
engineering personnel, these measures will remain in place until the CO, system
operability in restored.
This item remains open pending completion of appropriate analyses by the licensee
and resolution of issues resulting from such analyses.
112 (Closed) URI 50-354/97-0104: apparent creation of an urseviewed safety question
following installation of cress tie lines between resHual host removal subsystems.
This issue was also the subject of LER 50 354/97 05(see section E8.4 below). A
pre decisional enforcement conference with PSE&G management was held on
August 12,1997, to discuss this issue. As a result of previous inspector reviews
of this issue and the information gained at the conference, the NRC concluded that
this matter involved a violation of 10 CFR 50.59. A Notice of Violation NIO E 97-
160-01013)was issued under separate correspondence to PSE&G dated October
20,1997.
- - ~ -- - . ,
. ,- - - - - - - -_ __ -. - - - - . - - - -. - - . _
'
.
!
.
l
33 l
- HJ (Closed) VIO 50 354/97 01 05: reactor core isolation cooling turbine overspeed trip :
!
due to Governor valve stem binding. This violation involved a repeat f ailure of the
governor valve stem due to corrosirm, a well-documented industry issue (PSE&G
issued LER 50 354/97 03which described the details of this event). At the time e
this violation was issued, all planned corrective actions listed in the noted LER had l
been implemented and were judged to be acceptable. As such, no response to the
violation was required,
i
DJ (Closed) LER 50 354/07-Q19: closure of SACS to TACS isolation valve. This LER '
described the inadvertent automatic actuation, on August 7,1997, of the loop "C"
Safety Auxiliaries Cooling System (SACS) in response to a low flow signal from the
Turbine Auxiliaries Cooling System (TACS). The signe! resulted from the closure of
the TACS supply valve. A similar ever.t occurred Inter on September 4,1997. In !
this second event the initiating signal was from a low low low SAC expansion tank l
level alarm. - !
. ,
.
Af ter the latter event the licensee traced the problem to a loose fuse clip in the
Class 1E Analog Bailey Cabinet that provides inputs to the Digital Logic Control
System. Although the licensee did not originally identify the initiating cause of the
inadvertent engineered safety feature actuation, they were effective in identifying it
af ter the second event, even though the symptoms were different. A root cause
analysis was planned to address these failures. Bared on their review of PSE&G's <
corrective actions and confirmation that the licensee had checked for other loose ,
connections in the Bailey cabinets, the inspectors concluded that acceptable actions
had been taken to address these issues.
E8<10 (Closed) LER 50 354/97 020: past inoperability of safety related chillers due to
operation with low safety auxiliaries cooling system (SACS) temperature. This issue
was described in detailin NRC Inspection Report 50 354/97 05 and was determined 3
to involve a Non Cited Violation of 10 CFR 50.59. This issue involved a recent self-
identified discovery that safety-related chillers would not perform their design
, function with a loss of instrument air and SACS temperatures below 55 degrees F.
Corrective actions described in this LER were judged to be appropriate, in that they ,
'ocused on the implementation of a hardware design change to eliminate the 4
deficiency. The inspectors learned subsequent to issuance of this LER that PSE&G
had developed a modification to add dedicated instrument air accumulators for the
safety related chiller controls such that the impact of a loss of air would be
tolerable.
E8.11 1 Closed) LER 50 354/97 0Q5: operation in a technical specification prohibited -
condition due to failure to perform monthly flowpath verification surveillance checks
of residual heat removal system cross tie valves. This issue is also described in
section E8.1 above. An additional deficiency identified in this LER involved the
discovery that the cross tie valves had not been previously included in a monthly
flowpath verification procedure. The inspectors verified that the associated
procedure was revised to include the subject valves. Additionally, PSE&G
submitted a TS amendment request to the NRC which would add an additional
mor*'ily surveillance requirement to verify that these specific cross-tie valves are
locked closed.
, . _ _ - - _ _ __ ._ _ ___ _ _ _ _ _ . _
. _ . _ _ _ _ _ _ . _ _ _ . . _ - _ _ _ _ - _ _ _ _. - . . ._.
.
.
34
t
IV. Plant Suonort ;
'R1 Radiological Protection and Chemistry (RP&C) Controls
El. General Observations of Radioloolcal Control Practices
The inspectors observed generally good radiological control practices during the
period, which included both operational and shutdown conditions. Radiologically
controlled area (RCA) access controls were judged to be effective, largely due to
corrective actions instituted following past events. Specifically, ALNOR activated >
RCA entry point turnstiles eliminated a past problem involving personnel entering
the RCA without proper electronic dosimetry. Radiological Work Permit (RWP)
briefings were comprehensive. Technicians frequently questioned workers entering
the RCA regarding RWP information to ensure expected dose rates and exposure l
reduction techniques were understood. The inspectors observed that a thorough ;
pre job " flush" of the reactor water cleanup system significantly reduced general
area dose rates in the vicinity of associated purnp work. This effort resulted in the
overall exposure received during the activity to be well below job estimatcs, in spite
of the fact that the work continued more than three weeks past the original
completion date.
E2 Outaae RP Performance
a. Insoection Scone (83750)
The inspectors reviewed the licensee's radiological protection program
implementation during the Fall 1997 refueling outage. This review consisted of in-
plant work observations, interviews with licensee personnel, and review of
applicable documents,
b. Observations and Find)DQ1 t
!
The RP organization expanded staffing with the addition of 63 contractor RP
technicians with local control points estabilshed for the drywell, refuel floor, torus
and the turbine operating floor. The inspectar observations of work in progress
determined that there were sufficient RP resources available to cover the outage
work activities.
On the refueling floor, the inspector observed two sets of racks and a sealant
container containing fuel handling poles with approximately 10 wrapped poles that
had been opened at the ends exposing the contaminated components. The licensee
resealed the wrapped poles,
inside the torus, diving operations were conducted to clean the underwater surf aces
in preparation for emergency core cooling system suction strainer modifications.
Surveys indicated 10 30 mrad /hr smearable contamination levels on exposed
surfaces next to the catwalk work areas. The diving operation involved equipment
contamination at similar levels with RP technicians wiping down the contaminated
~
. - . - - _, , , . - , - - . - - . . . - - .
, - .- =_ - - - ,_ . . _ _ - _ _ - - . . - . - - - ._ . -. -
I
!
l
35
wet diving equipment as it was removed from the water. Drying out of this
equipment and the above water contaminated surf aces provided a potential for !
airborne radioactivity. The licensee maintained one stationary low volume air
sampler at the entry hatch to the torus, which was mor6 than 100 feet from the
observed diving and equipment laydown areas. No personal air samples had been '
taken for at least the previous nine days. This was considered poor air sampling
practice. Due to the inherent wet work of the diving operation that tends to reduce
airborne radioactivity, and the lack of significant personnel contamination incidents, l
the safety consequence of the poor air sampling practice was considered low and
was therefore viewed as a non cited violation of 10 CFR 20.1501 pursuant to
20.1204 (NCV 50 354/97 0711).
>
Refueling floor activities were regulatad by one RWP. This RWP specified electronic
dosimeter setpolnis of 50 mrem and 100 mrem /hr. During refueling floor evolutions l
with the cavity flooded, dose rates ranged from < 2 5 mrem /hr, with generally less (
than 10 mrem expected per entry. The drywellinservice inspection weld exam
HWP specified electrunic dosimeter setpoints of 220 mrem and 500 mrem /hr for all
workers. A variety of drywell work area dose rates existed depending on the piping
system being inspected. General finld dose rates of 20 300 mrem /hr could be '
expected with similarly varied entry doses.
The licensee made detailed evaluations of snubber locations and weld examination
locations to support dose estimates for the outage. These individuallocation dose
estimates were utilized as input to shielding considerations, but were not
determined as shielding requirements for the applicable RWPs. There was the
potential for work to commence prior to shielding installation.
The inspector's evaluation of the drywell shielding effectiveness indicated a good
dose reduction in the basement areas of approximately one-half. The entry level
and first level above general radiation fields were not appreciably influenced by the
shielding provided, although dose rates were reduced somewhat in the immediate
vicinity of the recirculation pump discharge nozzles. Significant dose rate gradients ,
existed in most of the drywell areas.
The inspector observed a weld insp etion work station setup in a high dose rate
area of the 130 foot elevation of the drywell(70 mrem /hr with adjacent areas
15 mrem /hr) that was not posted.
The 130 foot elevation of the drywellinvolved safety relief va;ve replacement
activities. The SRVs were in general dose rate fields of between 15 80 mrem /hr
within a three foot span. These areas were not posted to allow workers to orient
themselves in the lower dose rate fields and to maintain their doses as low as
reasonably achievable (ALARA).
The outage RWPs/ALARA reviews did not specify that the planned flushing of
reactor vessel nozzles or piping systems or the planned shielding installations were
a prerequisite for the applicable work,
t
l
.- , , . - . - - . --- - . . - - . _
- . - . . .. - - _--- . ._. . - --- .
.
.
.
36 ;
f
c. Conclusions ,
During the Fall 1997 refueling outage, RWPs generally provided effective ,
contamination control requirements, however, the exposure reduction plans '
(shielding or pipe flushing) were not specified as job requirements.
!
Progress in developing and implementing an initial drywell shielding plan was
fulfilled during the refueling outage. Some limited drywell shield thicknesses
resulted !n significant dose rate gradients in many drywell work areas and it was
determined that drywell postirigs were not effective to enable workers to reduce
their exposures while working in these high dose gradient areas. With generally
high alarm setpoints, th s use of electronic dosimeters for exposure control was not
optimized.
i
Some weaknesses in RP controls were observed during the refueling outage that
included refueling tool contamination control, and inside torus air sampling
practices.
E3 Outaae ALARA Performance
a. Insoection Scone (83750)
The inspector reviewed the ALARA planning, exposure estimate methods, and
interf aces with work management planning, as well as evaluated the results of
exposure reduction techniques planned and implemented during the Fall 1997
refueling outage. This review consisted of various documentation,in plant surveys,
and interviews with plant personnel,
b. Observations and Findir,as
The ALARA group had planned to " hydrolase" the scram discharge volume headers -
of the control rod drive systsm to reduce exposures to outage work in these areas.
The inspector noted that this exposure reduction work was schedu!ed to occur af ter
the work on hydraulic control units (HCU) (the principal work affected by the
hydrolasing activities) was completed. An interview with the responsible work
management planner determined that typical outage windows for system availability
delayed the hydrolasing activity until after the HCU work was completed. Detailed
review of control rod drive system activities to allow earlier sequencing of the
hydrolasing activity had not been adequately performed and some exposure
reduction opportunity was lost as a result.
The work order computer database (MMIS - Maintenance Managsment Information
System) contained the preliminary dose estimates. Periodically, the work order
information was electronically down-loaded into the scheduling software and
allowed the ALARA planning group to determine when the work activity doses were
scheduled. For the Fall 1997 outage, the work order dose estimates had been
completed, however, the scheduling software indicated almost no dose for the
outage tasks up to September 23,1997 (the outage began on September 10,1997)
_
.-
.- - -. -. -.
.
.
37
with indications that the needed MMIS information had not been down loaded for
many weeks. Therefore, the schedule of work activities could not provide a dose
estimate schedule as planned. The ALARA group used the outage schedule and
manually developed a dose schedule for each of the eight weeks of the outage.
This outage dose schedule assumed the original work schedule would occur as
planned. Due to emergent work and material delays, the outage schedule was
changed significantly and the RP department could not determine if outage exposure
performance was on track or not.
c. ,Q,or:clusioD1
ALARA/RP planning activities were not wellintegrated with work management
planning and scheauling and the interf ace was not always effective. Scheduling of
scram discharge volume header flushing activities after the HCU work activities .'
were completed and lack of integration of the outage dose schedule with the outage
schedule resulted in less than effective ALARA performance.
R5 Staff Training and Qualification in RP&C
fLQd RP Technician Training
a. Impection SconeE3760)
In a previous instoction report (No. 50 354/96 09)it was noted that PSE&G's
continuing training program for Radiation Protection (RP) technicians did not provide
periodic review of RP fundamentals,
b. Observations and Findinns
Subsequent to the October 1996 inspection, the licensee has evaluated RP
technician level of knowledge of RP fundamentals by administering the Mid Atlantic
Nuclear Training Group neneric examination. For the five subject areas,77% of the
RP technicians received a grade below 80% in one or more subject areas. One or-
one remediation training was conducted to recover the revealed deficient areas.
Due to the large percentage of RP technicians that had RP knowledge weaknesses,
the RP Services Group was evaluating if other areas of RP performance should be
evaluated that are not covered in continuing training, and were determining plans
for incorporating RP fundamentals into the continuing training curriculum,
c. Concimign
The licensee's RP technician continuing training program has been weak as
evidenced by poor RP technician performance on an examination given in the Spring
of 1997.
.
.
38
R7 Quality Assurance in RP&C Activities
a. [r.socction Scone
The inspector reviewed radiological occurrence reports (RORs) initiated since the
beginning of the outage,
b. Observations and Findinas
During the first 19 days of the Fall 1997 refueling outage,27 RORs were generated
and reviewed by the inspector. They consisted of 18 personnel contamination
events, four lost or wrong thermolumitiescent dosimeter incidents, three incorrect
RCA entries, and two other minor radiological discrepancies. All reports were still
open at the time of the inspection and they were under active investigation during
the outage,
c. Conclusions
The inspector determined that during the Fall 1997 refueling outage, the RP
correcti se action program was being actively implemented, with good low threshold
and high volume of discrepancies being reported.
R8 Miscellaneous RP&C lssues
f11 (Closed) URI 50-354/97-03-02: lack of timely completion of a design change
package for meteorological monitoring instrumentation. This issue, which
questioned the process for maintaining configuration conteol of plant equipment,
was identified f'uring an NRC inspection of the radiological effluents monitoring
program. The inspectors subsequently concluded that this issue involved an
isolated f ailure to implement the requirements of PSE&G's design change process.
Additionally, because the meteorological monitoring system is not within the sco,9e
of 10 CFR 50 Appendix B regarding quality assurance criteria, th, lack of timely
implementation of this change package did not constitute a violation of regulatory
requirements,
1102 IQpen/ Closed) VIO 50 354/97 0712:victation of the new Department of
Transportation (DOT) shipping paper requirements, in a previous inspection report
(No. 60 354/97 04),a violation of the new DOT shipping paper requirements was
identified. Specifically, radioactive laundry shipping papers did not indicate the
appropriate low specific activity (LSA) group notation from April 1,1996 through
June 9,1997 as required by 49 CFR 172.203(d!!11). Hope Creek and Salem
stations ship laundry separate from each other and, therefore, a violation was
issued to each station.
During this inspection, the inspector reviewed procedure, NC.RP-RW ZZ-0906(Q),
Rev. 2, " Shipment of Radioactive Material", and verified that the LSA group
designations were specified in the procedure. The inspector also verif'ed that Hope
Creek radioactive laundry shipment no. 97 27 was shipped as LSA Il as required.
_ _-- ,
.. ._ .
t
.
!
.
39
4
- P1 Conduct of EP Activities ,
>
P_11 Unannounced Emeroenev Preparedness Drill ;
,
PSE&G conducted an unanreounced off-hours drill of all emergency response l
'
1 facilities on August 27,1997. The inspectors observed the drill from the technical
1
support center (TSC), and noted generally good implementation of Hope Creek ;
emergency plan requirements. Appropriate ocedures were used, good i
communications were established, and a proper turnover from the senior nuclear :
shift supervisor was completed. One notable issue involved the f ailure to staff the !
"
TSC in a timely manner (i.e. 29 minut*s late). The post drill critique wa', thorough
and captured this issue and all other i ievant concerns. 4
. PS . Miscellaneous EP lesues ;
I
P_lL1 (fd9 ped) URI 50 354/96 07 03:UFS AR discrepancies regarding PSE&G emergency
plan. NRC inspectors identified twa issues during a 1996 program review: (1) no
radiological instrumentat;on was availab!e in the training center laboratory for use as
i a bsckup as stated in the emergency plan, and (2) PSE&G was not conducting an
annual program to provide emergency response information to the media and public i
as described in the emergency plan. The inspectors verified that both of these
discrepancies have since been resolved, in the first case, PSE&G eliminated the
statements describing backup instrumentation from the emergency plan because the
training center laboratory has been dismantled, in the latter case, the emergency
plan was revised to clarify how the emergency response information could be
disseminated. PSE&G can now take credit for media participation in annual
emergency preparedness dr:11s or exercises.
.
V. W.anaaement Meetinga
-:
'
X1 Exit Meeting Summary ,
A recent discovery of a licensee operating their f acility in a manner contrary to the UFSAR
description highlighted the need for a special focused review that compares plant practices,
procedures dr d/or parameters to the UFSAR descriptions. While performing the
inspections discussed in this report, the inspectors reviewed the applicable portions of the
UFSAR that related to the areas inspeM. The inspectors verified that the UFSAR
wording was consistent with the observea Slant practices, procedures and/or parameters.
The inspectors presented the inspection U suits to members of licensee management on s
October 10,1997. Licensee personnel acknowledged the findings presented. Howeve',
4
- on September 30,1997, where the engineering inspection findings were presented, the
- licensee disagreca with the inspectors' conclusions regarding the relays located in a mild
environment, stating that an adequate relay monitoring program existed.
- The insrectors asked the licensee whether any materials examined during the inspection
should be considered proprietary. No proprietary information was identified,
i
?
t
. - . , ,w., , . . - ._ , , , , , , . . ._m,-.. , , , , -- - - - - - .,_..m...,_,
. - . .- - . - . _- - . _ . -
o- !
.
,
I
r
40
!
INSPECTION PROCEDURES USED
t
IP 37550: Engineering ,
IP '57551: Onsite Engineering
IP 50090: Spent Fuel Storag6 Racks ,
IP 61726: Surveillance Observations
IP 62707: Maintenance Observations
IP 71707: Plant Operations
IP 73052: Inservice inspection - Review of Procedures .
!
IP 73753: Inservice inspection
IP 83750: Occupational Radiation Exposure '
IP 90712: Event Report Review
IP 92902: Followup Maintenance
,
ITEMS OPENED, CLOSED, AND DISCUSSED
'
Cooned
50-354/97 07 01 VIO lack of timely identification of inoperable electric fira
water pump.
50 354/97 07-02 URI degraded reactor coolant system pressure boundary.
50 354/97 07 03 VIO f ailure to adhere to maintenance procedures.
50 354/97 07 04 VIO failure to perform a safety evaluation per 10 CFR 50.59
50 354/97-07-05 URI service life of mild environment Struthors Dunn relays -
50 354/97 07 06 VIO f ailure to include five relays in 10 CFR 50.49 program
50-354/97-07 07 VIO f ailure to perform adequate relay service life
calculations
50 354/97-07 08 URI service life of Agastat relays in harsh environments
50 354/97-07 09 URI circuit breaker failure analysis
Ooened/ Closed
50 354/97-07-10 NCV f ailure to report violation of license condition
50 354/97-07 11 NCV poor air sampling practice
50 354/97 07 12 VIO f ailure to include current LSA group specification (LSA-
1,II,or ll) on laundry shipping papers since April 1,1996.
Closed
50 354/96-04 01 VIO f ailure to account for Bailey SSLM f ailures
50 354/96-04-02 URI control of Bailey logic tester programs
50 354/96-04-03 IFl Bailey SSLM EMI circuit testing
50 354/96-04 04 IFl UFSAR discrepancies
50 354/96 07-03 URI . UFSAR discrepancies regarding PSE&G emergency p!an
50 354/96 11 01 DEV f ailure to revise TS bases as committed in a PSE&G
license amendment
- .- . . . . - _ . -__- - -. .-
._ _ . . _ _ . . _ . . - _ _ _ __ . . . _ . - . _ _ _ - . . . _ . _ .._ ._. _ - __
- .-
. ,
3
i 41
4
- 50-354/EA 96-014-01013 VIO repeat failure of RHR system shutdown cooling common _
suction line mechanical snubber; i
50 354/97-01-04 URI = apparent creation of a USQ following lnstallation of _
.
_
l
, .
..
cross tie lines between RHR subsystems
50 354/97 01-05 VIO - RCIC turbine overspeed trip due to governor valve stem ,
binding
50 354/97 02 01' VIO ' f ailure to shut EDG cylinder test cocl s prior to engine
operation
'
. _
a
50 354/97 03-02 URI lack of timely completion of a design change package ,
for meteorolegical monitoring instrumentation -
. 50 354/97-005 LER vperation in a TS prohibited coadition ,
50 354/97-018 LER ESF actuations as a result of RPS motor generator set
- . output breaker trip i
50-354/97-019 -LER SACS to TACS isolation valve closure
50 354/97-020 LtR- past inoperability of the safety related chillers due to _ l
operation with low SACS
50-354/97-021 LER standby liquid ' control system tank concentration below
TS limits
50 354/97-022 LER engineered refety feature actuation - unplanned manual scram
Discussed:
50-354/96-09-03 VIO interaction of safety and non-safety related components
in the EDG room ventilation system
1
a
f
$
4
l
l
l
, , - - , - ,- -
-
.- . . . . . . ._. .- ..- . . --- . . . - .
o
d
e
42
LIST OF ACRONYMS USED
AE Architect Engineer
ALARA As Low As is Reasonably Achievable
ANI - Authorizea Nuclear inspector
ASME . American Society of Mechanical Engineers
DOT U.S. Department of Transportation
EDG Emergency Diesel Generator
EESS Equipment Evaluation Summary Sheet
EO Equipment Operator !
EMI Electro-Magnetic laterference !
-F&H Farwell & Hendricks
GE General Electric
-HCGS Hope Creek Generating Station ,
HCU Hydraulic Control Unit >
HPCI High Pressure Coolant injection
-lGSCC Intragranular Stress Corrosion Cracking
IRM Intermediate Range Neutron Monitor
ISI - Inservice Inspection
LSA Low Specific Activity
MMIS Maintenance M inagement Information System
NDE Nondestructive Examinationi
NRC Nuclear Regulatory Commission
PDR Public Document Room
.PSE&G Public Service Electric and Gas
PT Penetrant Test
GA Quality Assurance -
RCA Radiologically Controlled Area
RCIC Reactor Core Isolation Cooling
RORs Radiological Occurrence Reports
RP Radiation protection
RP&C Radiological Protection & Chemistry Controls ,
RWPs Radiation work permits
SACS Safety Auxiliaries Cooling System
SER Safety Evaluation Report
SORC Station Operations Review Committee
SSLM Solid State Logic Module
'TS Technical Specifications
UFSAR Updated Final Safety Analysis Report
UT Ultrasonic Test
_