ML18102A780
| ML18102A780 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 01/21/1997 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18102A779 | List: |
| References | |
| 50-311-96-81, NUDOCS 9701270185 | |
| Download: ML18102A780 (74) | |
See also: IR 05000311/1996081
Text
Docket No:
License No:
Report No:
Licensee:
Facility:
Location:
Dates:
Inspectors:
Approved by:
9701270185 970121
ADOCK 0500031l
G
.
. . ..~.,*---*:--/ :..::.: -*.
.. -.. ~ . -.. :-
.' . ." *-:.
U.S. NUCLEAR REGULATORY COMMISSION
50-311
50-311196-81
REGION I
Public Service Electric & Gas Company
Salem Nuclear Generating Station, Units 2 .
Hancocks Bridge, New Jersey 08038
December 2-13, 1996
J. Trapp, Team Leader, DRS
S. Klein, Reactor Engineer, DRS
G. Morris, Reactor Engineer, DRS
L. Prividy, Sr. Reactor Engineer, DRS
W. Sherbin, Contractor
S. Stewart, Sr. Resident Inspector, DRP
James T. Wiggins, -Director
Division of Reactor Safety, Region I
- .. ,: '-*
TABLE OF CONTENT$
EXECUTIVE: SUMMARY ...... ** . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
03
Operations Procedures and Documentation . . . . . . . . . . * * * . * . . . . . . . . . . . 1
03. 1 Emergency Operations Procedures and Single Failures * * . * . . . . . . . . . 1
03.2 Operating and Abnormal Procedures * * . . . . * * * . * * . . * . * * . . * . * . * * 4
05
Operator Training and Qualification . * . . * . * . * * . * * * * * * * * * * * * * * * * * * * * * . 5
05.1 .. Training Material and Simulator Fidelity *.***** ** * * * * * * * * * * * * * * * * 5
__ , -.*
M 1
Conduct of Maintenance . . . . . . . . . . . . . . . . . . . .. . . . . . . . * . . . . . . . . . . . 6
M 1. 1 System Flow Balance Test * . . * * * * * * * * * * * * * * * * * * :. ; * * * * * * * * * ** * 6
M1 .2 Pump and Valve Testing ................... * ...*
- ~~-~~~~~:-........ -~-~~~-~::9-
M1 .3 Maintenance and Testing of Heat Exchangers **** -** O..~'. * * * . * . . . . 12 *
M1 .4 -Testing of Instrumentation and Controls (l&C) *.. ~ : * * * * * * * * * . * . * 15
- *M2
- Maintenance and Material Condition of Facilities and Equipment * . * * * * * * * * 16
M2.1 >CC Radiation Monitors .. ~ .....* *. ~ **....*... * * ~~- ... _* .*... * ~ .. : *. .
1*6
M2.2 Root Cause Evaluations and Corrective Actions for System Failures * * * 17
-
.
M3
Maintenance Procedures and Documentation ....***. ~ . . * * * * * . * * * . . . * 18
M3.1 Ventilation System Testing and Documentation ****** -* * * . * * * * * . . 18
-... M3.2 Test Procedure Acceptance Criteria . . * . . * * * * * * * . * * * * * * * * * * * . * 19
E1
Conduct of Engineering . . * . . * * . * . . . . * * . . . . * * * * * . * * * * * . . . . . . . . * . 22
E1 ;-1' Component Cooling Pump Runout and NPSH * * * . * * * . * * * . * * * * . . . 22
E 1*.2
CC Pump Room Ventilation * . * . . . . * * * . * * * * * . . * * * * * . * * * * * * * . 26
E1 .3 Pump Seal Water Cooling . * *. * . . * * * * . * * * . * * . * * * * * * * * . . * . * * * 27
E1 .4 Electrical Protective Devices .......***. * * . . .. . . * * * * * * * * . . . . . . 28
E1 .5 Setpoint Control . * * * . * . * * . . . . . * . . . . * * * . * . . . . * * * . * . . . . . . 33
E1 .6 Equipment Power Supplies * . * . . . . . * . . . . . * . . * * * . . * * * . * . . . * * 34
E3
Engineering Procedures and Documentation * . * . . . * * . * * * * * * * * * * . . . * . * 36
E3.1
.Electrical Calculations * * . * * * * . * . . . * * * * * * * * * * * * * * * * * * . * . * * 36
E3.2
Technical Standards . * . . * * * * . . * * . . . * * * * . * * * * * * * * * . * . . . . * 38
E3.3
Drawing Control ...................... ~ ... -. . . . . . . . . . . . . 40
E3.4 Configuration Baseline Document * . . * . . . * . * * . * * * * * * * . . . . . * * * 41
ii
TABLE OF CONTENTS (CONT'D)
Miscellaneous Engineering Issues * * * * * * . * * * * . * * * * * * * * * * * * * * * * * * * * 42
ES. 1
Post Accident Sampling System Heat Exchangers
. . * * * * * * * * * * * * * 42
E8.2
Licensing Basis Verification * * * * * . * * * * . . * . * * * * . . . * . . * * * * * * * * 43
E8.3
Licensing Basis Updates . * * * * * * * . * * * * * . * . * * * * * * * * * * * * * * * * * 45 *
E8.4
Probabilistic Safety Assessment * * * * * * * * . * . . * * * * * * * * * * * * * * * * 46
. *;_-;_
- X 1
Exit Meeting Summary * * * * * * .* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 4 7
--. ..-
iii
Report Details
The objective of this inspection was to conduct an independent inspection to determine if
the Salem Unit 2 component cooling* (CC) system would perform its intended safety
function. The scope of the inspection included verifying that the system had a technically
sound design and licensing basis, system components were tested to demonstrate design
requirements, and operating practices and procedures were consistent with the design.
The team used the guidance provided in NRC Inspection Manual Chapter 93801, Safety
System Functional Inspection (SSFI), to conduct this inspection activity.
The team noted that significant improvements were made to the CC system during the
current outage. These improvements included the completion of a system flow balance,
resolving instrument calibration errors, and the completion a significant number of
corrective and preventive maintenance activities. The team found that the licensing basis
CC description was, with a few exceptions, consistent with the actual CC system's design
and operation. However, the team did identify two design issues regarding the CC pump
room ventilation and the maximum acceptable flow limit for the CC pump, where operating
practices were inconsistent with system design information. The resolution of these issues
was ongoing at the conclusion of this inspection. The team concluded that contingent
upon the satisfactory resolution of the SSFI findings and the licensee's already identified
CC system restart issues, the Salem Unit 2 CC system would perform its intended safety
function.
I. Operations
03
Operations Procedures and Documentation
03.1
Emergency Operations Procedures and Single Failures
a.
Inspection Scope
The team reviewed the PSE&G contingencies for emergency operations to verify
that plans and emergency operating procedures were consistent with the design
bases for the CC system.
b.
Findings and Observations
CC Pump Operability Requirements
The Salem Updated Final Safety Analysis Report (UFSAR), Section 9.2.2.3, states
that "In the event of a_ loss-of-coolant accident (LOCA), one pump and one heat
exchanger are capable of fulfilling system requirements." Salem Unit 2 Technical
Specification 3. 7 .3 requires two independent CC loops to be operable in Modes 1,
2, 3, and 4 *
2
In early 1995, PSE&G identified that certain emergency scenarios could place a
single running CC pump in a runout (excessive flow) condition. Specifically, if a
standby CC pump was out-of-service for any reason and a LOCA occurred
coincident with a loss of offsite power and failure of a vital bus (which fails a
second CC pump), two residual heat removal (RHR) heat exchangers would
automatically be placed in service with only one operable component cooling pump. *
This alignment resulted in a single CC pump providing flow to two RHR heat
exchangers. The licensee determined that this condition could be resolved by
operating two CC pumps when two RHR heat exchangers were inservice. As an
interim measure, in March 1995, PSE&G operations established an operating policy
wherein if any one of the three CC pumps were not available, the technical ': :- : * *
specification action statement for an inoperable CC loop would be entered. The
operating policy provided added assurance that two CC pumps would always be
available.
- :
The EOPs could have been revised to include contingencies for single CC pump
operation; however, to reduce decisional steps and to allow for simplification of the
EOPs for the LOCA, the initial availability of two CC pumps was assumed.
Minimizing operator decision steps in the EOPs assured completion of the
realignment of valves from the refueling water storage tank (RWST) injection phase
to containment sump recirculation phase within an established time limit that
assured adequate core cooling. Appropriate contingencies for loss of a vital bus
and accompanying equipment were included in the EOPs.
CC Pumo Room Ventilation (For additional information see Section E1 .2)
The team reviewed the EOPs and the need for two operable CC pumps in the
accident mitigation strategy. The EOPs were found consistent with the PSE&G
operating policy. However, the team questioned PSE&G on the need for the CC
pump room ventilation and whether the ventilation system controls were adequate
to support three CC pump availability under normal and accident conditions.
The 21 CC pump room ventilation equipment (2VHE-33) and the 23 CC pump are
provided power from electrical Train C. The 22/23 pump room ventilation
equipment (2VHE-34) and the 22 CC pump are provided electrical power from
Train B. The 22/23 CC pump room ventilation equipment (2VHE-34) is common for
both the 22 and 23 CC pumps. Electrical Train A provided power to the 21 CC
pump. A failure of Train C electrical power would prevent operation of the 21 CC
pump ventilation equipment (2VHE-33) and the 23 CC pump. A failure of the 22/23
room ventilation equipment (2VHE-34) or electrical Train B would result in the
failure of room cooling for both the 22 and 23 CC pumps. The failure of the room
ventilation* equipment may fail the associated CC pump. The team identified that,
during certain accident scenarios and conditions, the EOPs require at least 2
operable CC pumps. If less than 2 CC pumps are operable, then the EOP
instructions cannot be completed to ensure adequate CC will be available .
3
PSE&G reviewed the team's concerns and determined that plant operations prior to
the decision to administratively require three CC pumps for technical specification
operability could have resulted in operations outside of the plant's design bases.
Specifically, for periods prior to 1995, when the 21 CC pump was out-of-service,
the postulated single failure of the 22/23 CC pump room cooler could have resulted
in a condition where no CC pumps would be available for accident mitiga~ion. The
postulated single failure of the 22/23 CC room cooler was a condition that had not
been evaluated by PSE&G and was a condition that alone could have prevented the
fulfillment of the safety function of the CC system. PSE&G made a notification of
their determination to the NRC in accordance with 10 CFR 50. 72(b)(2)(iii),- on
November 25, 1996.
.*.~-.. * : * *.
The team was informed that PSE&G was considering a licensing bases change to .
revise the requirement for CC pump operability to capture the need to have three.
pumps operable during plant operations. Resolution of ventilation concerns would
be required to support the change. PSE&G made resolution of the ventilation issue
a Mode 6 (Refueling) prerequisite.
CC Pump Runout (For additional information see .Section E1 .1).
The team observed that, even if standby electrical power w:re
0
availabl:~ {~:~he .three
CC pumps and to the room ventilation equipment, an EOP directed alignment allows
one CC pump to be placed in an apparent runout condition for a short time during a
postulated LOCA event coincident with a loss of offsite electric power. In the
Salem EOPs, one CC pump is started in the initial steps following the reactor trip at
the onset of the postulated event. During the change from the injection to the
recirculation lineups, in Salem procedure EOP-LOCA-3, the CC valves from the two
RHR heat exchangers are automatically opened prior to start of a second CC pump.
The second pump was manually started following the valve alignment change. The
entire change from injection lineup to the recirculation lineup is designed to be
completed in less than 12.5 minutes following receipt of the RWST low level alarm
at 15.2 feet. PSE&G operations had demonstrated that the second pump would be
started in less than eight minutes from the onset of the runout condition. PSE&G
personnel stated that single pump operations for a short period of time had been
evaluated; however, the team questioned the flow rate limit that had been used in
the evaluation.
The team was concerned that a single CC pump operating with both RHR heat
exchangers in service would result in CC pump flows in excess of that previously
evaluated and could result in pump damage. The team was also concerned that the
overall electric loadin9 of the emergency diesel generator supplying power to the CC
pump, when in the runout alignment, may be higher than previously evaluated. In
response to these issues, PSE&G initiated an action request (AR) and initiated both
an engineering review of the issue and a root cause evaluation .
... :_ ..... _.
'-.*-.
4
c.
Conclusions
PSE&G had identified that emergency operations for some postulated events could
result in operation beyond the design of the CC pumps. As an interim measure, in
1995, PSE&G operations initiated a policy to ensure that at least two CC pumps
were operable during plant operation. The policy to ensure that at least 2 CC
pumps would always be available failed to appropriately account for a single failure
of CC pump room ventilation. The failure of the 22/23 CC pump room cooler, ,vvhen
the 21 CC pump was out of service, could have resulted in no available CC puinps
during some postulated accident conditions. The failure to have considered. thisj .. _.~*
design deficiency remains unresolved pending NRC review of this issue for potential
enforcement action (URI 50-311/96-81-01).
-
The team identified that the operating policy of having three operable CC pti111ps .. ,.
failed to properly consider the affect of a loss of cc pump room ventilation orl'pump
operability. The failure of the EOPs to account for a* single failure of CC pump room
ventilation is an NRC unresolved item pending the completion of the licensee's
corrective actions and the review of this issue by the NRC for potential enforcement
action (URI 50-311/96-81-02).
- , *-"
The team identified that the EOPs allowed a CC pump to operate at flow rates
beyond its documented design limits for a short period of time. The failure to
provide a technically sound basis for operating a CC pump in this manner remains
unresolved pending the completion of the licensee's corrective action and the
review of this issue by the NRC for potential enforcement action (URI 50-311/96-
81-03).
03.2 Operating and Abnormal Procedures
a.
Inspection Scope
The team reviewed the CC system operating and abnormal procedures to verify the
procedures instructions properly reflected the system design.
b.
Observations and Findings
The team found that current revisions of operating procedures were in place to
support CC system operations. The procedures had been recently revised and
included enhancements such as basis sections for each procedure to describe
commitments and reasons for various steps, and detailed sections for contingency
actions when abnormal procedures were used *
--
------
5
The team identified a procedure discrepancy concerning when to trip rea~t~r coolant
pumps (RCPs) on loss of all component cooling. The discrepancy involved whether
to immediately trip all RCPs on loss of CC, as stated in the CC abnormal procedure,
or to allow five minutes for CC restoration as specified in the RCP abnormal
procedure. PSE&G personnel stated that 10 minutes of RCP operations without CC
was allowed by the technical manual for the pumps, however a previous
commitment to the NRC stated that the pumps would be conservatively tripped
immediately on total component cooling water loss. PSE&G prepared an action
request to resolve the discrepancy.
c.
Conclusions
Operating procedures related to normal and abnormal operations of the component
cooling water system had been recently upgraded by PSE&G, and the ~ea.m _
- _~,>, - *
considered the improvements to be a good initiative. The team conCluded that the
CC procedures were generally of good quality and appropriately reflected the CC
system design and licensing basis. However a discrepancy was identified on when
to trip reactor coolant pumps on total loss of CC.
05
Operator Training and Qualification
05.1 Training Material and Simulator Fidelity
a.
Inspection Scope
The team reviewed operator training for the component cooling system to verify
that appropriate design information was provided to the operators. The review
included an assessment of the technical completeness and accuracy of appropriate
training materials and examination tools. The fidelity of the plant simulator
regarding the CC system was also reviewed.
b.
Findings and Observations
PSE&G performed a crew readiness assessment in January 1996 to evaluate the
preparedness of the reactor operators for resumption of plant operations. The
examination was conducted in three parts, a written examination, simulator.
scenarios, and a plant walkthrough evaluation. After a detailed evaluation of the
examination results, PSE&G concluded that no specific weaknesses existed in
operator knowledge regarding CC system operations and design. Therefore, no CC
specific training was required before restart.
A CC lesson plan had* been prepared and was awaiting final supervisory review* and
approval. The team reviewed the training plan and found that design information
had been appropriately included. At the time of this inspection, the lesson plan had
not been used. The requalification examination bank for written evaluation and job
performance measures was reviewed by the team, and CC design and operations
information was appropriately included .
.::*
6
The team reviewed simulator fidelity for full power CC operations, abnormal
operations, and response to the loss of offsite power and loss of coolant combined
with loss of offsite power events. In each case, the simulator appropriately
modeled CC system performance and provided effective training for these events.
The team was informed that extensive training on recently revised EOPs had been
conducted and had included CC operations. Simulation of remote shutdo~n panel
operations was not provided, and training on these operations was done by in-plant
discussions and walkthroughs.
The team found that the operating, abnormal, and emergency operating procedures
in use at the training facility were current and of high quality. Instructors were**
knowledgeable of CC specific operations and training provided to evaluattitthe_
operations.
_ . .
- * * 'f/i*:
c.
Conclusions
M1
The team found that PSE&G had translated appropriate CC design information into
materials used for training and evaluating licensed operators. The simulator
provided an effective tool for training and evaluating CC operations during normal
~nd accident conditions. The procedures and lesson plans used for CC training *
were of high quality and appropriately complete for evaluation of operator -
knowledge and abilities on the CC system *
II. Maintenance
Conduct of Maintenance
M1.1 System Flow Balance Test
a.
Inspection Scope
b.
The team reviewed a special flow balance test that was conducted to verify that the
CC system would perform consistent with assumptions in the accident analysis.
The test procedure was reviewed to verify that proper acceptance criteria were
established for assuring adequate CC flow to safety-related equipment during
postulated accident conditions.
Observations and Findings
On July 18, 1996, PSE&G initiated performance improvement request (PIR)
960709221 to identi~y and resolve a concern regarding whether the required
component cooling flow through the residual heat removal heat exchanger (RHRHX)
during a LOCA alignment would be obtained. PSE&G concluded that t.he best
alternative to resolve the concern was to perform a special test in accordance with
Procedure No. TS2.SE-SU.CC-0001 (0), CC System Flow Balance. in a 10 CFR
50.59 evaluation of this procedure, which was conducted in October 1996, PSE&G
described the limiting LOCA alignment for the CC system as follows: one CC pump,
one CC heat exchanger, one RHR heat exchanger, all the emergency core cooling
-.
--..;:
... *.~*.:
..
. 7
system pumps, and the non-isolated, non-safety loads. The non-isolated, non-
safety loads for Unit 2 include: the positive displacement charging pump, the seal
water heat exchanger, the waste gas compressors, and radiation monitor system
(RMS) heat exchangers.
The two main purposes of the flow balance test were:
To ensure that the CC system safety-related components would receive
design flow in the limiting LOCA alignment with one train of the CC system
available.
To benchmark a CC system hydraulic flow model that had been developed
by a contractor and was being reviewed by PSE&G.
The CC system manager and the *cognizant mechanical design engineer presented
the results of the flow balance test to the Salem Unit 2 test review board during a
November 21, 1996 meeting which was observed by the team. PSE&G identified
the m_ajor test deficiencies. as follows:
The 21 and 22 RHR control room CC flow indicators read approximately .
1000 gpm higher than the temporarily installed ultrasonic flow measuring
instruments.
The CC flows for the 21 and 22 centrifugal charging pumps mechanical seal
heat exchangers were 9.5 and 10.8 gpm respectively compared to the
required test acceptance value of 11.5 gpm. The team also noted that
UFSAR Table 9.2-3 specified the CC flow to these components as 14 gpm
(max). The cognizant mechanical design engineer indicated that the pump
manufacturer had confirmed the minimum CC flow requirement to be 6 gpm.
On this basis the test results were considered acceptable. PSE&G stated
that UFSAR Table 9.2-3 and applicable plant procedures would be updated
accordingly.
During the latter portion of the flow balance test, lower than required flow
readings were observed using the local flow indicator (2FIC643A) for the 21
safety injection (SI) pump seal water heat exchanger. PSE&G attributed the
problem to a malfunction of 2FIC643A since initial flow readings from this
instrument for a comparable system alignment resulted in flows greater than
the required flow of 11.5 gpm. Work order (WO) 961031116 was issued to
troubleshoot and repair 2FIC643A.
During the initial flow balance of the system in accordance with Sectio~ 5~ 1
of Procedure No. TS2.SE-SU.CC-0001 (0), the CC flow to the 24 reactor
coolant pump thermal barrier as read on local flow indicator (2Fl620) was 38
gpm versus the required flow of 40-42 gpm. The cognizant design engineer
stated that 38 gpm was acceptable with the pump manufacturer.
8
During discussions between the team and the cognizant mechanical design
engineer, it was apparent that allowances for instrument error and CC pump
degradation had been considered to establish reasonable acceptance criteria for the
flow balance test. For example, the required CC flow to support operability of the
21 SI pump was 10 gpm (UFSAR Table 9.2-3). The cognizant mechanical design
engineer indicated that 5% of full instrument range ( +/- 1 gpm) was required to
account for instrument error and a 3-4% flow allowance was needed to account for
a 10% degradation in pump head. Therefore, the acceptance criteria for the 21 SI
pump CC flow was established as 11.5 gpm in the flow balance test procedure.
PSE&G had discussed this general approach for establishing CC flow acceptance
criteria in Action Request 961003083 which was issued to resolve the minimum
required CC flow to the 21 and 22 centrifugal charging pumps mechanical seal heat
exchangers. However, there were no documented calculations to support the CC*
flow acceptance criteria used in the flow balance test procedure. Pending PSE&G's
completion and documentation of the calculations. su.pporting the test acceptance
criteria and review by the NRC for potential enforcement action, this issue is
unresolved (URI 50-311196-81-04).
The team also noted that documented calculations had not been completed for the
required CC flow values to be incorporated into CC system procedures, such as the
CC pump surveillance test procedures, S2.0P-ST.CC-0001(Q), -0002(0), and -
0003(0), The team noted that an allowance for repeatability between surveillance
tests should be considered in establishing the acceptance criteria, This
consideration was illustrated by the following anomaly which could not be explained
by the CC system manager or the cognizant mechanical design engineer. CC flow
provided to the 21 RHR pump seals met the test acceptance criteria during the
performance of the flow balance test (11.5-13.0 gpm measured in October 1996
with 21 CC pump) while the flow recorded during a later troubleshooting procedure
did not (10.5 gpm measured in December 1996 with 23 CC pump). The team
noted that this anomaly was exacerbated by the fact that the 23 CC pump was the
strongest (i.e., highest developed head for a given flow) of the 3 CC pumps.
Therefore, the 23 CC pump should have provided more flow than the 21 CC pump
for the comparable CC system alignment.
c.
Conclusions
It was apparent that PSE&G had attempted to include reasonable acceptance criteria
into the flow balance test. However, final conclusions regarding the results of the
flow balance test could not be made pending completion of the calculations required
to properly document the CC flow acceptance criteria specified in Procedure No.
TS2.SE-SU.CC-0001_(Q).
9
M1 .2 Pump and Valve Testing
a.
Inspection Scope
Testing was reviewed to verify adequate pump and valve performance to support
system operability. The review included PSE&G's implementation of the inservice
test (IST) program and corrective actions taken to resolve deficiencies found during
IST program Audit 95-0125.
b. *
Observations and Findings
Pumps
The team reviewed PSE&G's actions regarding periodic testing of CC pumps. **Based
on a review of test records, the team noted that PSE&G was adequately testing the
CC pumps in accordance with the Salem IST program requirements. * The team also
noted that PSE&G had made a number of changes to improve pump testing in
response to the findings of Audit 95-0125. For example, AR 9507221196 had .
been issued to resolve a significant pump testing problem. The CC pump test
procedure had been written such that flow was not closely controlled which. *
resulted in questionable repeatability of test results for trending pump performance.
The team verified that the CC pump test procedures were corrected to ensure
adequate test repeatability.
Notwithstanding the changes made to improve the pump testing at Salem, the team
noted that several substantial CC pump testing activities, which the licensee had
identified as needing to be addressed, were not done at the time of this inspection.
The first activity involved elevated vibration readings for the three CC pumps which
have recently been observed. PSE&G has issued a troubleshooting procedure per
WO 960928055 to evaluate this problem. Also, after the documented pump
acceptance criteria has been developed based on the flow balance test, the pump
test procedures would need to be revised and pump baseline testing needed to be
reperformed.
Relief Valves
The team reviewed PSE&G's program for testing relief valves in the CC system.
The review included PSE&G's preliminary response to Generic Letter (GL) 96-06,
Item 3, regarding potential overpressurization of piping caused by thermal expansion
of trapped fluid between closed valves.
In the Salem IST prog.ram basis data sheets, PSE&G documented the basis for all
CC relief valves concerning their inclusion or exclusion from the American Society
of Mechanical Engineers (ASME)Section XI IST program. The team had the
following comments upon reviewing these data sheets:
10
PSE&G concluded that the CC excess letdown heat exchanger outlet relief
valve (2CC112) did not provide a safety function. PSE&G considered that
this thermal relief valve was not required to be in the scope of the IST
program because inadvertent opening of this relief valve combined with
subsequent failure to reclose would not prevent the excess letdown heat
exchanger from performing its function. However, the team determined that
2CC112 does perform a safety function since it is relied upon to provide
thermal relief protection for CC piping between containment isolation valves
2CC113 and 2CC115. Hence the relief valve should be periodically tested to
provide ongoing assurance regarding its operational readiness. The team
noted that the piping between 2CC113 and 2CC115 was identified by
PSE&G in their preliminary response to GL 96-06 as requiring relief
protection. PSE&G indicated that, even though relief valve. 2CC112 was not
currently in the IST program, it had been tested satisfactorily on
September 27, 1996. PSE&G stated that 2CC112 would be included in a
periodic surveillance test program.
PSE&G had appropriately included in the IST program the relief valves (21,
22, 23, and 24CC129) which protect the CC piping associated with thermal
barrier cooling for the reactor* coolant pumps, and the vacuum breaker
(2CC148) and relief valve (2CC147) for the CC surge tank. The team also
verified satisfactory testing of these valves.
Manual Valves
The Salem EOPs included steps to isolate both the boric acid evaporator and the
spent fuel pool heat exchangers from component cooling following certain
postulated accidents. These steps were intended to prevent runout of the single CC
pump started at the onset of the event. The team indicated that the manual
isolation valves specified by the EOPs were not included in the inservice testing
program and, therefore, could not be credited as operational. The team considered
that operation of the valves could not be assured if periodic testing and monitoring
was not accomplished. Further, for some interim period prior to shutting the valves
but after a CC pump was started, the pump could be in a runout condition. PSE&G
responded to the concern by demonstrating that in some* scenarios, runout of the
operating CC pump would result in the CC low header pressure alarm which could
cause the reactor operator to isolate the non-safeguards CC header using valves
CC-30 and CC-31, which were included in the IST program. This action could not
be expected to be accomplished until some time after the CC pump had been
operating in a runout condition. PSE&G had not evaluated this condition.
11
The team observed that no CC system manual valves were included in the scope of
the Salem IST progra*m. However, as stated above, the Salem EOPs included * .*
specific operating instructions regarding manual valves to isolate both the boric acid
evaporator and the spent fuel pool heat exchangers from component cooling.
PSE&G issued AR 961202179 to address the apparent inconsistency between the
EOPs and the IST program regarding manual valves. PSE&G committed to resolve
this inconsistency by establishing a periodic surveillance test for the applicable
manual valves.
Power Ooerated Valves
The team reviewed testing of several power operated valves in the Salem IST
program including MOVs and the CC surge tank vent which is a solenoid, air >
operated globe valve. The team also reviewed testing of the service water system
solenoid air operated valves (21 and 22SW129) that supply cooling water to the CC
pump room coolers. PSE&G had appropriately included power operated valves for
the CC system in the IST program.
Valves 21 and 22SW129 were being adequately stroke time tested in accordance
with Procedure S2.0P-ST.SW.0014(Q), Rev.3, lnservice Testing, Room Cooler*
Valves. In reviewing the testing of MOVs, the team noted that PSE&G recently
reviewed the thrust limits for the RHR heat exchanger outlet valves (21 and
22CC16) to be consistent with their response to GL 95-07, Pressure Locking and
Thermal Binding of Gate Valves. The team verified that the maximum thrust limits
for these MOVs would be limited to minimize valve closure thrusts and thus prevent
thermal binding upon opening. Maximum closure thrust values (28,611 lbs for
21CC16 and 26,566 lbs for 22CC16) not to be exceeded during testing were
established in the Managed Maintenance Information System. Diagnostic test
procedures required adherence to these limits while testing these MOVs.
The team reviewed the acceptance criteria that was included in Procedure S2.0P-
ST .CC-0001 (Q), 21 CC Pump lnservice Testing, for determining the acceptability of
the backflow check function of the CC pump discharge check valve 21 CC1. The
procedure requires 2 of 3 of the following indications for an acceptable test: (1) an
audible "clapping shut" when stopping the 21 CC pump; (2) decreased pressure
observed at the pump discharge pressure gage; and (3) no reverse flow observed at
the 21 CC pump. The team considered these criteria to be appropriate and noted
no test failures for the CC pump check valves.
,* .. -
12
c.
Conclusions
Although the licensee had identified that several substantial punip testing activities
remained to be completed, the team concluded that PSE&G was adequately
implementing pump and valve testing for CC as required by the IST program. The
team identified instances where controls were not in place for periodically testing
certain manual valves used in the EOP valves. PSE&G agreed to include these
valves in a periodic surveillance test program.
M1 .3 Maintenance and Testing of Heat Exchangers
a.
Inspection Scope
b.
The team reviewed CC heat exchanger maintenance and testing actions being-taken*.
regarding GL 89-13, Service Water System Problems Affecting Safety-Related
Equipment, as described in PSE&G correspondence NLR-N90021, dated
January 26, 1990, which was later revised in NLR-N90165, dated August 31,
1990.
PSE&G uses thermal performance testing to confirm that the component cooling
heat exchangers (CCHX) can transfer required heat loads to the ultimate heat sink
during a postulated accident. The results of this testing are used as input to a
computerized model of the CCHX to confirm that the heat exchanger
manufacturer's design data sheet performance can be achieved. The team sampled
the results of CCHX thermal performance testing ancl the related computer model
input and output.
Observations and Findings
CC Heat Exchangers
The CC heat exchangers are performance tested periodically in accordance with
Procedures S2.0P-PT.SW-0026(Q) and -0027(0), Revision 5, 21 and 22
Component Cooling Heat Exchanger Heat Transfer Performance Data Collection.
Testing is performed while shutdown in Mode 4 with the last test having been
performed in October 1994 for both CCHX. The Component Performance group is
responsible for reviewing the test data, calculating the heat transfer capability of the
heat exchanger, and providing the results to the system manager for review and_
approval.
The team reviewed the heat exchanger thermal performance test results obtained in
October 1994. PSE&G calculated fouling factors based on test data to predict heat
removal at design conditions. The predicted heat removal was multiplied by 0.95 to
account for uncertainty (i.e., a 5% instrument measurement uncertainty value was
assumed) and then compared to the heat removal requirements specified under
service conditions in the heat exchanger vendor data sheet.
13
Based on the review of the thermal performance test results, PSE&G concluded that
the CC heat exchangers would remove the required heat load under accident
conditions. However, the team questioned the technical basis for the 5%
assumption used to account for instrument measurement uncertainty. PSE&G
stated that the 5% measurement uncertainty value was based on "equivalent"
instrumentation, which was not identical to that used for the test. Hence, there
was no specific documented technical basis for the assumed measurement
uncertainty. The team determined that if the measurement uncertainty was greater
than about 11 %, the heat exchangers may not meet their required thermal
performance under accident conditions.
Also, in light of clogged tubes reported from recent heat exchanger inspections, the
team questioned how PSE&G calculated essentially clean heat exchangers with zero
fouling factors on both tube and shell sides from the 1994 thermal performance
testing results. For example, a review of the inspection and cleaning in March 1996
of the 21 CC heat exchanger performed under WO 960608023 indicated that "10%
of the inlet tubes and 75% of the return tubes were clogged." Assuming similar
heat exchanger condition during the 1994 performance test, it is not clear how the
thermal performance tests for heat exchangers in service, prior to cleaning, would
have zero fouling factors, when tube clogging was actually observed.
CCHX Thermal Performance Computer Model
The team found cases where computer model predicted near zero or negative
fouling for the CC heat exchanger. For these cases, the heat exchanger
performance calculated by the model was better than that predicted by the
manufacturer's data sheet for a clean heat exchanger.
The team questioned the potential for non-conservative prediction of heat exchanger
performance by the model. A non-conservative heat exc_hanger performance model
would indicate less tube fouling than may actually* be present in the heat exchanger.
Therefore, an unacceptably fouled heat exchanger, that would be unable to perform
its design function, may not be detected by conducting the performance test. In
response to the team's questions, the licensee stated that the computer program
was established to provide the most realistic assessment of cooler cleanliness and
intentionally removes inherent conservatism in the theoretical calculations used in
the model.
The licensee also ran an additional case using the model in which the
manufacturer's design data sheet performance factors (temperatures, flows, and
fouling) were input. Results showed that the model predicted approximately 12 %
better performance (overall heat transfer coefficient, U) than the data sheet
performance. Although the model only predicts a 5% better heat load capability,
the 12% over prediction in "U" is significant because this is the value used by
Westinghouse in their accident analyses .
c.
14
In response to the team's questions, the licensee contacted the manufacturer to
obtain further information on whether the model's performance predictions were
consistent with and applicable to the Salem Unit 2 CCHX.
CC Pump and Heat Exchanger Room Coolers
In a revised response to GL 89-13, dated August 31, 1990, PSE&G committed to
periodically inspect and clean the two CC pump room coolers. PSE&G viewed this
revised commitment as an acceptable alternative to the thermal performance testing
option of GL 89-13. As a result of this revised commitment, PSE&G no longer
committed to "trending important system parameters".
Based on a review of preventive maintenance WO 960528046 for inspection and
cleaning of the 22 CC pump room cooler performed in April 1996, the team was
concerned that the present method of inspecting and cleaning the room coolers may
not demonstrate adequate thermal performance of the CC room coolers. The cooler
inspection was performed in accordance with Procedure SC.MD.PM.SW-0006(0),
Revision 4, Service Water Room Coolers Internal Inspection. The specific concerns
were:
Comments in WO 960528046 indicated that silt, waterbox debris, and failed
lining were present in the heat exchanger waterbox. However, acceptance
criteria were not included for determining the as-found acceptability of the
cooler regarding fouling factors and service water flow rates. This was
inconsistent with PSE&G's response to GL 89'." 13 which stated in part that
"Procedures will include acceptance criteria and recommended actions for
acceptable results." Also, no technical justification for the inspection
frequency of the room coolers existed.
PSE&G did not periodically verify adequate service water or air flow through
the room coolers. This concerned the team since silting was known to exist
in service water lines and room cooler waterboxes and the manual damper
(2-VHE-747) supplying air to the 22 CC pump room was found closed during
a plant walkdown.
Conclusions
The team concluded that the computerized model used to predict CCHX
performance, based on test data, may not be conservative. PSE&G is in contact
with the heat exchanger manufacturer to resolve this issue. The team concluded
that the lack of a specific documented technical basis for the 5 % instrument
measurement uncertainty assumption used in CC heat exchanger performance
calculations was an unresolved item pending further evaluation of this issue by
PSE&G and review by the NRC for potential enforcement action (URI 50-311/96-81-
05). The lack of acceptance criteria for assessing the as-found condition of the
room coolers and for establishing adequate service water and air flow rates in CC
room cooler maintenance procedures was considered to be an unresolved item
pending PSE&G's evaluation and review by the NRC for potential enforcement
action (URI 50-311196-81-06).
15
M1 .4 Testing of Instrumentation and Controls (l&C)
a.
Inspection Scope
The team witnessed testing that was performed to resolve a discrepancy with
residual heat removal (RHR) heat exchanger CC flow measurement found during the
system flow balance test.
b.
Observations and Findings
CC RHR Heat Exchanger Outlet Flow Indication '(2Fl-601 A and Bl and Flow Element
(2FE-601A and Bl
CC flow is measured through each RHR heat exchanger by a transmitter which
senses the differential pressure (DP) between the inner and outer radius t_aps _th~t
are* located* on* a 12-inch, 90°* elbow installed in the* CC piping. The flow element
was original plant equipment specified by Westinghouse to develop a DP of 137 .
inches of water which would correspond to a full scale flow of 10,000 gpm on the
Control Room console indicators (2Fl-601 A and B). * During the performance of the
flow balance test, a discrepancy of CC flow to the RHR heat exchangers was
observed. The Control Room console indicators read about 1000 gpm higher than
temporarily installed ultrasonic flow meters (USFMs) which were used during the
test to more accurately measure flow through both RHR heat exchangers. PSE&G
reported this discrepancy to the NRC on Novembe~ 29, 1996, in Licensee Event
Report 96-028.
PSE&G performed a troubleshooting procedure authorized by W0961112091 to
accomplish an insitu calibration of the elbow flow meters for the 21 and 22 trains.
This procedure confirmed the problem was associated with the lack of initial field
calibration of the elbow flow meters. The team witnessed portions of the
troubleshooting procedure and verified the following:
The test equipment for flow (Panametrics Transport PT868 Flowmeter) and
DP had been calibrated to kf!OWn standards prior to the test.
Each transducer with mounting hardware was installed in accordance with *
the USFM vendor instruction manual and located at optimum locations (i.e.,
in straight run of pipe with more than 10 pipe diameters upstream and 5 pipe
diameters downstream).
The team confirmed with the cognizant l&C engineer that Calculation No. SC-
CC002-01, Revision 1, 1 /2 Component Cooling RHR Outlet Flow Indication and
Alarms, was the calculation of record and would be revised to include the correct
design inputs for incorporation into the channel calibration procedures for CC flow
through each RHR heat exchanger.
16
c.
Conclusions
The team concluded that appropriate measures had been taken to control the
collection of test data for the insitu calibration of the elbow flow meters. PSE&G
was taking appropriate actions to correct the elbow flow meter design inputs and
revise the CC RHR heat exchanger outlet flow indication and alarm channel
calibration procedures.
M2
Maintenance and Material Condition of Facilities and Equipment
M2.1 CC Radiation Monitors
a.
Inspection Scope
The team reviewed design and operations for containing radioactive contamination
that could enter the CC system.* *
- ** * *
- . ** *
- * * * * * * *
- *
b.
Observations and Findings
The Salem UFSAR, Section 9.2.2.3, states, "Since heat is transferred from the
component cooling water to the service water, the component cooling system
serves as an intermediate system between the reactor coolant and the service water
systems and insures that any leakage of radioactive fluid from the components
being cooled is contained within the plant." Further, in Section 11.4.2.2, the
UFSAR states that "Component cooling liquid monitors (2R-17A,B), continuously
monitor the component cooling water for radiation."
During system walkdowns, the team was informed by PSE&G that the CC radiation
monitors had been out-of-service since some time in 1995 due to problems with the
radiation monitoring computer. The CC radiation monitors are non safety-related
and are not used to determine offsite radioactive releases. PSE&G returned the
monitors to service on December 9, 1996. During the period that the monitors
were out-of-service, the reactor had been defueled leaving the spent fuel pool heat
exchanger as the only active CC load. No significant leakage of component cooling
either into or out of the system had been suggested by surge tank level changes.
Routine, weekly samples of component cooling water had been completed and had
included evaluation of gross activity. During two periods when component cooling
pumps were secured for maintenance, low levels of activity were detected in the
CC system. The likely source of the leakage was the spent fuel pool heat
exchanger. PSE&G informed the team that inspection of the heat exchanger did not
identify leaking tubes and no activity was detected when the CC pumps were
running. Also, no leakage of chromates into the spent fuel pool had been detected
by chemistry sampling of the pool.
17
c.
Conclusions
The team identified a discrepancy between the UFSAR requirement for CC radiation
monitors to continuously monitor CC for radiation and station practice wherein the
monitors had been out of service for over one year. The team was concerned that
the licensee appeared to not repair these radiation monitors in a timely manner given
the length of time the monitors were out-of-service and the function they provide
(i.e. prompt identification of leakage into the system from radioactive systems
served). These radiation monitor issues are unresolved pending further NRC review
for potential enforcement action (URI 50-311/96-81-07).
M2.2 Root Cause Evaluations and Corrective Actions for System Failures
a.
Inspection Scope
The teani reviewed corrective actions and root cause evaluations *for CC equipment" **
problems that had been identified by PSE&G.
b.
Observations and Findings
The team reviewed 1994, 1995, and 1996 corrective maintenance work orders
(WO) concerning pumps, various check valves, and several power operated valves
in the CC system. Based on this review, the team observed the following:
Increasing vibration levels had been noted for the CC pumps during inservice
testing. PSE&G was taking appropriate actions to address this problem even
though the vibration levels had not reached the alert level. WO 960928055
had been recently issued to perform a troubleshooting procedure for
gathering and analyzing pump vibration levels while operating different
pumps in paraliel.
Several check valves (2CC186 and 2CC119) had frequent local leak rate test
failures. However, PSE&G took appropriate action to correct these failures.
These valves were included in a comprehensive root cause analysis report
issued by the check valve performance group in April 1996 in response to
repeated failures of containment isolation valves identified in CR
96031914 7. Also, the CC pump discharge check valves had performed well
with no corrective maintenan.ce required during the last 3 years.
WO 950924133 had been issued to correct a bent spring rod pipe support
(2P-CCH-332) near the excess letdown heat exchanger containment isolation
valve. The team verified that the cause of this problem was adequately
assessed and not attributed to any hydraulic disturbance such as water
hammer. PSE&G had determined the cause of the damaged support to be
from an associated maintenance activity where the rod was bumped by a
heavy load.
18
High valve factors were determined for MOV 2CC117 during initial testing
that was performed in response to GL 89-10. PSE&G attributed the problem
to carbon steel wedge shoes without hard facing on the valve internal wear
surfaces causing high frictional forces. PSE&G implemented design change
2E0-2340 to provide replacement wedge shoes with hard facing that
improved the valve performance.
c.
Conclusions
The team concluded that PSE&G was taking good corrective actions to identified
CC equipment problems.
M3
Maintenance Procedures and Documentation
M3.1 Ventilation System Testing and Documentation
a.
Inspection Scope
The team reviewed testing and documentation aspects of the ventilation system to
determine its capability and readiness in supporting the operability of the CC
system.
b.
Observations and Findings
Separate room coolers are located in the motor driveri auxiliary feedwater pump
area for providing cooling for each of the two CC pump rooms. Air is cooled by a
fan coil unit_ and independently ducted to each CC pump room. Return air from the
pump rooms passes through a louvered fire damper mounted in the fire door of each
CC pump room. The return air discharges through this louver to a hallway on the
84 foot elevation of the Auxiliary Building. When in standby, the room coolers are
started on high temperature by individual room thermostats.
The team observed the following design and configuration deficiencies during plant
walkdowns:
The louvered fire damper in the fire door (Door C8-2) for 21 CC pump room
was closed. The team noted that an analysis had not been performed
regarding the impact on return air flow and room temperature with this
louver closed. PSE&G also informed the team that two design information
items concerning the ~oom coolers were not available: (1) PSE&G could not
determine design information concerning the louvered fire damper, such as
free area with the louver open; and (2) no air flow calculation existed for the
room cooler to determine if the louver in the fire door was adequately sized
to pass design air flow. The team concluded that the closed louvered fire
damper may have prevented the CC room coolers from performing their
design basis function .
---
-
19
In addition to the closed fire damper, the manual damper designated 2-VHE-
747 located in the supply air duct of the room cooler designated 2VHE33
was closed. Also, manual damper designated 2-VHE-749 located in the
supply air duct of room cooler designated 2VHE34 was closed. The team
noted that these dampers and associated duct were shown on auxiliary
building ventilation drawing 205337-A-8763-20 as supplying 600 cfm of air
to the 22 CC pump room and 1350 cfm of air to the auxiliary feed pump
area, respectively.
PSE&G could not account for the position of the dampers described above .
Therefore, the team questioned if PSE&G had administrative controls
regarding ventilation damper positions to support equipment operability in
safety-related systems. PSE&G indicated that there were no existing
controls. AR 961121961121204 was issued for the Operations Manager to
determine what controls should be implemented for ventilation system
equipment to assure that the plant is o"pe*rated c"o"r'1sistent with desigh
assumptions.
c.
Conclusions
The team concluded that these deficiencies demonstrated inadequate configuration
control of ventilation equipment needed to support CC system operability. This
issue remains unresolved pending further NRC review for potential enforcement (URI
50-311/96-81-08).
M3.2 Test Procedure Acceptance Criteria
a.
Inspection Scope
The team reviewed the maintenance and surveillance test procedures and results of
selected electrical equipment required to support the CC system. The 125 Volt
batteries are required to support control of the CC system pumps, the on-site power
supply and power the pilot solenoid valves. The emergency diesel generators
(EDGs) supply power for the CC system pumps and motor operated valves.
b.
Observations and Findings
1.
Salem UFSAR, Section 8.3.2.1 states that three 125 Volt batteries are
provided for the control power for the vital buses and power for the 125 Volt
de distribution cabinets. Technical Specification 4.8.2.3.2g describes the
requirements for a battery capacity discharge test to demonstrate at least
80% of the manufacturer's rating every 60 months. Technical Specification
4.8.2.3.2h further states that the test frequency shall be increased to 12
months if the battery shows signs of degradation including a drop of more
than 10% capacity from the previous test .
20
Salem Nuclear Generating Station implements the requirements of these
technical specifications by Procedure SC.MD-FT.125-0002(0), Rev. 4, dated
November 15, 1995, 125 Volt station Batteries Performance Discharge Test.
The team reviewed the latest battery capacity test data for the safety-related
125 Volt batteries. The team noted that procedure SC.MD-FT.125*.0001 (0),
Rev. 0, dated November 5, 1992, 125 Volt Station Batteries Performance
Discharge Test, was the controlling procedure at the time of those tests~
The team confirmed both procedures referenced Institute of Electrical and
Electronics Engineers (IEEE) Standard 450-1987, Recommended Practice for
Maintenance, Testing, and Replacement of Large Lead Storage Batteries.
Attachment 11 to procedure SC.MD-FT.125.0001 (0), required that the
capacity of the battery be calculated at the completion of the test in
accordance with procedure steps 5.4.37 or 5.5.18. The team noted that the
required calculation of battery capacity was defined in the body of the
- procedure *{steps 5.4.37 and 5.5.18) as the ratio of the time to reach" low
battery voltage to two hours (the time used to establish the discharge rate.)
The team confirmed this method of calculating battery capacity was in
agreement with the method contained in IEEE:..450-1987, Section 6.5, *
Determining Battery Capacity. The team found that the tests for batteries
2A and 2B (performed in May 1993 and April 1993 respectively) were
stopped at two hours instead of proceeding to the low battery voltage point
as implied by procedure steps 5.4.37 and 5.5.18. The licensee indicated the
tests were stopped at two hours because that duration was required by
other steps in the test procedure (5.4.33 and 5.5.17 .)
The team noted that
since the test was not properly completed, no calculation could be performed
in accordance with procedure steps 5.4.37 or 5.5.18. However, the data
sheets indicated the batteries had 100% capacity.
The team confirmed the batteries had at least 100% capacity as stated by
reviewing the recorded test data. However, the team was concerned that
there was no documented true battery capacity established in the 1993
tests, there was no value to compare with the next battery capacity test for
degradation.
In response, the licensee had the battery manufacturer estimate the probable
capacity from the 1993 test data. The results showed the battery capacities
were 11 5 % for battery 2A e1nd 112.5 % for battery 2B.
The team noted the latest revision to the battery test procedure (SC.MD-
FT.125.0002(0), Rev. 4, dated November 15, 1995), was changed, in part,
to incorporate a change to the Technical Specification Surveillance
Requirement 4.8.2.3.2 (approved by the NRC on September 19, 1995.) Part
of the Technical Specification change incorporated a requirement to increase
the frequency of the battery performance tests from 60 months to 12
months if degradation in battery capacity of more than 10% was found from
the previous test. The team found that the revised procedure failed to
incorporate an acceptance criteria for battery degradation. The team also
21
found that the discussion in the body of the procedure on degradation in the
form of notes (Procedure pages 21 and 48) indicated a frequency change to
every 18 months was appropriate for a greater than 10% capacity drop from
the previous test.
In response to these concerns, the licensee issued AR 961206169 to review
this item and revise the procedure.
2.
The team reviewed the battery charger maintenance procedure S2.MD-
ST.125-0001 (0), Rev. 0, dated July 27, 1996, 125 Volt Battery Chargers.
The 125 Volt battery chargers had been replaced under change 2EC-3332/1
because of maintenance problems with the original chargers. The original
chargers were rated for 250 Amps and the new chargers were rated for 300
Amps. Because of ampacity concerns of the ac power input cables, both the
original and the new chargers required the current limit to be set at a
maximum of 210 Amps. The team.found that the current* battery charger *
current limit as-found test contained a caution and required the electrician to
. adjust the controls to ensure the current drawn by the charger would not
3.
exceed 210 Amps. This instruction would inhibit recording the true as-found
current limit if it had drifted above 210 Amps because the test would not
permit loading the battery charger above 210 Amps. In response to the
team's observation, the licensee initiated an AR to review this item .
The team reviewed the Diesel Generator Speed/Load Control System
Alignment procedure, SC.MD-CM.DG-0006 (Q), Rev. 8, dated
June 12, 1996, to determine the setpoint for the EOG governor motor
operated potentiometer (MOP) allowable frequency range. This setpoint was
critical because the EOG loading calculation (ES-9.0002) is based on the EOG.
frequency not increasing above 60.5 Hertz. This restriction was further
emphasized in a recent request to change the Technical Specification
4.8.1.1.2 from 61.2 to 60.5 Hertz max. License Change Request S95-36
was sent to the NRC on November 25, 1996. However this was a re-write
of license change request (LCR) 94-40 which was initiated in response to
Incident Report 94-301, dated October 13, 1994, which first documented
the potential for EOG overloading because EOG frequency in excess of 60.5
Hertz conditions. The team found that even though the new procedure had
an acceptable as-left criteria of 59.80-60.20, there were no acceptance
criteria for the as-found condition (procedure step 5.5.1.) The team
observed that the as found condition of EOG 2C was recorded at 60.22 Hz.
when tested on May 11, 1996, following the installation of a new MOP.
The setting was returned to 60.03 Hz. The team confirmed the settings for
EDGs 2A and 2B were found and left within the allowable as-left tolerance
during their last preventive maintenance.
The licensee indicated that an AR would be initiated to revise that section of
the procedure addressing as-found frequency of installed MOPs .
c.
22
Conclusions
The team concluded that the failure to incorporate the latest technical specification
surveillance criteria in the battery surveillance performance test procedure was a
procedure weakness. The licensee stated that the procedure would be revised to
properly reflect the technical specification requirement.
The team concluded that the licensee had failed to follow their battery performance
test procedure for calculating the capacity of batteries 2A and 28 in 1993 because
of an inadequate test procedure.
The team also concluded that the errors found in the battery charger test procedure
and the EOG speed control alignment procedures had minor safety significance.
The licensee has initiated actions to correct these procedure deficiencies.
The* above issu*es are unresolved pending the completion* of the licensee's corrective
action and the NRC review of this issue for potential enforcement action (URI 50-
311196-81-09).
Ill Engineering
E1
Co.nduct of Engineering
E1 .1
Component Cooling Pump Runout and NPSH
a.
Inspection Scope
The team reviewed several design calculations, engineering evaluations, and other
design documents to assess the design basis and supporting analysis for the CC
system.
b.
Observations and Findings
Calculations
Calculation S-C-CC-MDC-0879, Revision 1, dated June 28, 1992, Maximum and
Minimum CC Pump Flow Requirements and NPSH, established the CC pump
minimum permissible flow, runout flow, and evaluated required and available NPSH
during runout conditions. The team reviewed this calculation in detail and identified
the following weaknesses:
1 .
The calculation (Sheet 1 l refers to an attached pump curve to establish the
recommended minimum flow rate of 1000 gpm for the CC pumps. However,
the pump curve is a generic Goulds pump "catalog cut" without any specific
customer designation. It was not clear that this curve is applicable to the
Salem CC pumps .
--
-
~
23
2.
Using the accident alignment for the CC system provided in TS2.SE-SU.CC-
0001 (0), Revision 0, CC System Flow Balance, with the minimum flow
requirements provided in UFSAR Table 9.2-3, and assuming isolation of CC
flow to the spent fuel pool heat exchanger and boric acid evaporator early in
the injection phase (in accordance with the EOPs) of a postulated large break -
LOCA, the team determined that total CC pump flow may be as low as 771
gpm. Consequently, the 1100 gpm minimum CC pump flow requirement
established in the calculation may not be satisfied during the injection phase
of a postulated LOCA.
3.
The calculation assumes (Sheet 2) that "CC pump runout flow is 20% higher
than the flow corresponding to its best efficiency point (BEP)." The
calculation states that this assumption is consistent with "normal industry
practice." In addition, the required pump NPSH at runout (5700 gpm) is
extrapolated (Sheet 7) from the manufacturer's design pump curve.
However, the p*ump curve* only shows retju"ired NPSH for Hows* up to 5000 *
gpm. It is not obvious from the curve that required NPSH could be limited to
the 23 feet value determined in the calculation.
4.
No documented basis was provided for the minimum surge tank water level
(El. 126'-0") used in the calculation to determine NPSH available to the
pump. Although a setpoint calculation exists for the surge tank levels,
PSE&G was unable to correlate the levels in the setpoint calculation (SC-
CC003-01, Revision 0) to the low level elevation specified in the calculation.
Licensee Identified Pump Runout Issue
The calculation states (Sheet 1) that "None of the operating modes of the CC
system require flow in excess of the design flow," and " ... there is no expected
scenario that would lead to pump runout condition ... " However, in 1994 PSE&G
identified the potential for CC pump runout during a postulated LOCA and failure of
a vital bus (PR 940805141). Consequently, PSE&G contracted an external
engineering organization to perform an additional analysis of the CC pump capability
to operate under runout conditions (Evaluation of Component Cooling Pump at
Runout Condition, dated November 7, 1994). That analysis also used
extrapolations of the manufacturer's pump curves to conclude that the pump could
operate "reliably at runout flow conditions for an indefinitely long period." The
report also stated that the CC pump will operate at runout until another CC pump is
restored to service or until an RHRHX can be isolated. Subsequently, PSE&G
developed administrative guidance to require three CC pumps to be operable
assuring the availability of two CC pumps assuming the single active failure of one
CC pump.
24
SSFI Identified Pump Runout Issue
The team identified other cases where the potential exists for the CC pump to
operate at or near runout conditions which had not been adequately evaluated to
assure the pump manufacturer's NPSH requirements would be satisfied. For
example:
In the event of a postulated LOCA, (using current EOPs) one CC pump is
started during the injection phase, and the RHRHX outlet valves
(21CC16&22CC16) automatically open during recirculation on low level in
the RWST, running out the pump for approximately 10 minutes.
When a CC pump is started in EOP-TRIP-1, the spent fuel pool heat
exchanger and boric acid evaporator may still be aligned for service. Recent
flow balance test results indicate that CC pump flow may be near runout
(5*500-5'600 gpm) until flow to these loads is.isolated.
A Westinghouse analysis (PSEB0-96-040, Revision 1, dated
September 3, 1996, Single Train Cooldown Analysis Report) indicates that
CC flow rates through the CCHX and RHRHX are based on one CC pump
operating near runout.
In response to these issues, the licensee contacted* the pump manufacturer to
obtain further information related to CC pump performance under runout conditions.
In a letter to PSE&G, dated December 11, 1996, the manufacturer stated that:
The minimum flow for this pump is 600 gpm for a maximum of 60 minutes.
Maximum continuous flow is 5600 gpm. Based on testing done on same
size pumps, a maximum flow rate of 6370 gpm may be tolerated for up to
10 minutes. However, NPSH required is 26 pounds per square inch absolute
(psia) and brake horsepower is 362 hp.
At the time of the inspection, PSE&G did not have a formally documented hydraulic
analysis or a field benchmarked and issued flow model that establishes the
maximum possible CC pump flow that could be achieved for all worst case system
alignments. Except for the estimates and extrapolations reflected in this calculation
and the independent analysis, there was no documented hydraulic analysis to
establish the maximum possible CC pump flow that could be achieved, or to
confirm that the manufacturer's NPSH requirements would be satisfied.
25
CC Flow Diversion
The team also noted that when the outlet valves open on both RHRHXs (on low
RWST level) and the CC pump is at runout, CC flow may be diverted from safety-
related components (e.g., SI, charging, and RHR pumps) to the RHRHXs. There
was no documented analysis or testing to assure that adequate CC flow would be
supplied to the safeguards pumps during these conditions. In addition, with two
RHRHXs operating, more heat may be rejected to the CC system. There was no
documented analysis to establish what flow would be supplied to the RHRHXs, and
that the CCHXs could maintain CC supply temperatures below the design limit
(126°F).
EOG Loading
The pump manufacturer determined that the CC pump motor would be required to
produce* approximately 362 brake horsepower at pump runout conditions. This is
an increase of approximately 62 brake horsepower above that assumed in the EOG
loading calculation for a CC pump motor. The team noted that the impact of the
increased horsepower *requirement on the EOG loading had not been evaluated.
Summary
In summary, the team found that there was no documented analysis to confirm
that:
Minimum CC pump flow will be greater than 600 gpm in all cases, e.g.,
during injection.
The maximum flow will not be exceeded in any mode of operation or system
alignment.
Sufficient NPSH is available to support operation at the maximum permissible
runout flow rate. It is not clear that operation at this flow rate, even for a
short duration, will not result in cavitation that could compromise the
capability of the CC pump to continue performing its safety function.
The emergency diesel generator has adequate capability to accommodate the
increased loading from a CC pump operating at the specified runout flow.
Adequate CC flow is supplied to safeguards pumps when the outlet valves
on both RHRHXs are open at low RWST level; or, that sufficient flow would
be supplied to both RHRHXs to maintain CC supply temperatures at or below
126°F.
The licensee has issued AR 961212085 to evaluate CC pump operation at the
maximum flow rate of 6370 gpm .
26
c.
Conclusions
The team identified a condition where the operation of the CC pumps appeared
inconsistent with documented design limits. The team concluded that the CC
-pumps would probably be at or near runout conditions when the RHRHX outlet
valves are automatically opened during a postulated LOCA. CC pump operation at
runout during these conditions had not been adequately analyzed by PSE&G.
Consequently, the CC pumps may be adversely affected if sufficient NPSH is not
available, and the pumps are subjected to the effects of cavitation. An unresolved
item for this issue is described in Section 03.1 of this inspection report.
E1 .2
CC Pump Room Ventilation
a.
Inspection Scope
- At the completion* of the inspection, PSE&G' had not issued a verified analyses of
CC room temperatures under design basis conditions with postulated single failures
- of the auxiliary building ventilation room coolers. The team was informed of
preliminary room .temperature results by PSE&G. The team reviewed an interim
design calculation completed to establish the maximum outside air temperature that
would permit simultaneous operation of the 22 and 23 CC pumps assuming failure
of ~he 22/23 pump room cooler during Mode 6 (refueling operations).
b.
Observations and Findings
There was no documented analysis to demonstrate that sufficient cooling would be
provided to the 22/23 CC pumps to permit satisfactory pump operation with the
22/23 CC pump room cooler (2VHE34) out-of-service. In response to the team's
concerns, PSE&G performed a preliminary analysis to determine CC room
temperatures under design basis accident conditions assuming the single failure of
these room coolers. The analysis -assumed the door to the pump room with the
failed room cooler is open. Preliminary results indicated temperatures as high as
138°F in the room with the failed room cooler, which was in excess of current
room temperature design limits. PSE&G is evaluating the permanent removal of the
22/23 CC pump room door, and is developing analyses of room temperatures for
these accident conditions.
The team also reviewed interim design calculation S-2-ABV-MDC-1666, Revision 0,
Maximum Outside Air Temperature for Mode 6 Entry. The calculation was
developed in response to the team's concern with the single failure of CC room
ventilation equipment. * This calculation was performed using a GOTHIC computer
model. Since the room coolers are thermostatically controlled to start at 100°F,
and calculated temperatures in the areas never reached 100°F, without modeling
the coolers, no room cooler operation was assumed in the analysis. The doors of
the 22/23 CC pump room were removed in the model. The calculation determined
that the maximum permissible outside air temperature for operation under these
conditions is 67°F. The interim analysis was performed to allow refueling activities
to proceed prior to the final resolution of this issue.
- .: .. ':..*.
27
c.
Conclusions
Since the analyses of cooling available to the 22/23 CC pump room for design basis
accident conditions had not been completed, the team was unable to assess the licensee's
evaluation. PSE&G is continuing efforts to complete and verify these calculations and to
resolve the issues related to the single failure of the room coolers and its impact on CC
system performance. The team found the interim design calculation on 22/23 CC pump
room temperatures was acceptable in that it adequately represented the scenario described
. for Mode 6 operation. An unresolved item for the ventilation system issues is described in *
Section 03.1 of this inspection report.
E1 .3
Pump Seal Water Cooling
a.
.,
. .. * .... * ... " .. *.
!* :.
- ...... ,,
.*. '*
- * . * * *
- The*team reviewed the technical basis for assuring that ade*quate CC water flow will be
provided . to emergency core cooling pump seal water coolers following the initiation of an
accident .. *
b.
Observations and Findings
A Westinghouse letter BURL-3824, dated May 14, 1980, ESF Pump Operation Without CC
indicates that the centrifugal* charging, safety injection, and RHR pumps can be operated
without CC being supplied to the seal water heat exchangers for 15-20 minutes following
an accident or blackout provided that lube oil cooling* (service water) is automatically
started within 50 seconds. A memorandum attached to the letter recommended that
procedures should be changed to start a CC pump in the event of a small break LOCA that
could result in extended RWST drawdown time beyond the 15-20 minute criterion. The
team questioned whether this issue had been addressed to assure that adequate cooling
water is supplied to the safety-related pump seal coolers*within the prescribed 20 minutes.
The licensee had issued PIR 950814345, dated February 1996, which raised similar
questions on this issue. However, resolution of this PIR was .not completed and the
licensee was not viewing closure of the PIR as a restart issue.
PSE&G engineers stated that, a CC pump will be started within 20 mi~utes after the
initiation of any accident event, including any size LOCA. PSE&G is further evaluating the
time required to start a CC pump using the current EOPs*to confirm that a CC pump can be
started within the required 20 minute time frame.
c.
Conclusions
The licensee's operations staff indicated that the EOPs would direct the operators to
manually start a CC pump in less than 20 minutes after the initiation of any accident event.
However, PSE&G was unable to provide documentation to support this assessment.
Consequently this item remains unresolved pending the completion of the PSE&G
evaluation and NRC review for enforcement action (URI 50-311/96-81-10).
28
E1 .4
Electrical Protective Devices
a.
Inspection Scope
The team reviewed the electrical protective devices selected for CC system pumps
and MOVs to ensure that the components were adequately protected and the
device setpoints were appropriate. The team also reviewed the calculations, single
line drawings and MCC pan descriptions and performed walkdowns of associated
MCC pans to verify the as-built CC system MOV power supplies and protective
devices.
b.
Observations and Findings
...... }; .. ,
. CC MOV !hermal Overloa~ ff<?U H:eaters ...
.. * ..
Heater Selection
The team reviewed calculation ES-18.006, Rev. 0, dated June 16, 1994,
Selection of TOL Heater Elements for Safety Related MOVs, to confirm the
design was in conformance with the licensee's commitn:ients to Regulatory
Guide 1 . 106. The calculation indicated a change from using option 1 a of the
Regulatory Guide, continuously bypassed except during testing, to option 2,
trip setpoints established with all uncertainties in favor of completing the
safety-related function.
The team identified that the calculation incorrectly used the manufacturer's
data for the trip characteristics for the TOL relays. The manufacturer's
information that was included as a reference was incomplete in that it did
not include the current range tables required for proper selection of the
heater elements. The calculation did correctly include information that
indicated the trip point was 125% of the heater element minimum current,
but never defined minimum current. The licensee incorrectly used digits
included as part of the heater model number as the trip point. The team
noted that the licensee's technical standard for TOL sizing was issued one
month after the calculation had been issued and correctly addressed the
manufacturer's heater selection criteria for trip point determination.
The team questioned the source of the MOV motor data because the
referenced motor curves listed in the calculation spreadsheet were not
included in the referenced motor data packages. PSE&G was able to find
motor data sheets for the CC system valves in question under another
vendor document not referenced in the calculation.
29
The team noted that the calculation methodology adjusted the thermal
withstand data for the MOV motors based on applied voltage. Motor thermal
withstand is not directly affected by applied voltage. In addition, the
calculation also adjusted the motor time current characteristic curves for
voltage. This unrealistic double compensation for voltage was a
conservative error. However, the team did find that the calculation failed to
address the ambient temperature of the MOVs which could affect their
thermal withstand capability. The motor data that the licensee found
indicated the motors were rated for 40 degrees centigrade (°C). The
calculation indicated some of the MOV motor ambient could be as high as
50°C.
The team observed that the time current characteristic curves for the heater
elements selected for 6 of the 14 CC valves would not provide locked rotor
protection. This failure to meet one of the goa_ls of the calculation was not
- * addressed. *
The team observed that the calculation did not address the position of the
TOL adjustment knob, located on the face of the TOL relay. This adjustment
could affect the trip point by + /-10 % . This was also was not addressed in
the assumptions. The team walked down several CC system motor control
. center (MCC) compartments and confirmed the adjustment knobs were set at
100%. Therefore, this did not affect the results of the calculation.
The calculation was based on un-compensated TOL relays and adjusted the
trip curves for a 50 °C ambient. The team's walkdown confirmed that there
was a mix of un-compensated and ambient compensated TOL relays.
The licensee documented the calculational weaknesses identified by the
team in AR 961212226. The AR was designated as requiring completion
prior to Mode 4, hot shutdown.
Heater Design Control
The team observed that design change DCP 2EC-3249, approved
July 9, 1994, was initiated to provide adequate protection for the power
conductors feeding the safety-related valves. This was to be accomplished
by resizing the thermal overload relay heaters, breaker sizes and removing
the jumpers around the TOL contacts. Calculation ES-18-006 was the base
design document for the TOL heater selection.
The team identified inconsistencies between the calculated TOL heater size,
the heater size listed on the 230 Volt MCC one line diagrams and the heater
size listed in the Maintenance Management Information System (MMIS) data
base. These inconsistencies affected 3 of the 14 CC system MOVs. In
addition, the team identified an additional MOV which did not have any TOL
data listed in .MMfS and two MOVs that had two different TOL heater sizes
listed for the~same valve.
30
The team identified the following examples where the TOL calculation and
the TOL heaters installed in the plant did not agree:
CC118
CC136
CC190
C2.68A
C2.60A
C5.92A
Installed
C3.01
C3.56A
C6.30A
In response to the difference between the calculation and the installed
heaters, the licensee* performed an extent of condition evaluation and found
28 other discrepancies in safety-related systems. The licensee identified that
a change document (CD) No. 509/0 had been written against calculation
ES-18.006, as a result of design change 2EC-3249. The CD was issued to
revise the calculation for 30 heaters, including_ 2 of the 3 CC system MOVs
- identified above: The CD did not address one of the* cc system MOVs
,identified .as having a discrepancy. In addition, the team noted that for one
. CC system MOV, discrepancies in heater size existed between the installed
- heater, calculation, and the CD.
The CD failed to provide a technical justification for the change in TOL heater *
size. The team found that 10 of the 30 TOL heater changes involved
reducing the size of the heaters without any design calculation. Smaller size
heaters could result in premature tripping of the safety-related MOVs. This
was in direct conflict with the one of the stated goals of modification
_ 2EC-3249 to remain in conformance with Regulatory Guide 1 . 106 by
selecting heaters based on the conservative methodology.
2.
CC MOV Molded Case Circuit Breaker (MCCB) Magnetic Setting
. The team noted that the time current curves developed as part of calculation
ES-18.006 to demonstrate adequate protection for the MOV and the
connected power cables did not include the MCCB that forms part of the
combination motor starter with the TOL and the contactor. The TOL
manufacturer's information included as a reference to the calculation stated
that short circuit protection must be provided for the TOL and its associated
controller [contactor]. Selected MCCB curves were included in Calculation
ES-13.006(0), Rev. 2, dated October 18, 1995, Breaker and Relay
Coordination. The team noted that the MCC one line diagrams indicated a
variety of MCCBs had been used in the MCC compartments supplying the
CC system valves. These included thermal-magnetic (T /M) breakers that
respond to both overload and fault conditions and adjustable magnetic-only
(MAG) MCCBs that respond only to fault (instantaneous) conditions. The
team found examples where four CC valves with the same size MOV motors
and TOL heaters (C5.92A) were protected by 15 Amp T/M MCCB
(2CCV117, 2CCV131 ), a MAG set at 42 Amps (2CCV17) and a MAG set at
128 Amps (2CCV18). The team identified the 128 Amp setting was
excessive because the ratio of protection of the MCCB to the TOL was
,
.. *
31
greater than 20 and should normally be in the range of 7 to 10. This high
. setting provided insufficient protection for the combination starter. The team
also found that Calculation ES-13.006, Attachment E2, page 109, plotted
this breaker at 42 Amps, not 128 Amps.
In addition, the team found two pairs of valves (21 CC3, 22CC3 and 2CC30,
2CC31) with TOL size C10.48 heaters and 52 Amp RMS inrush currents that
had instantaneous MAG settings of 68 and 75 Amps, respectively. Standard
industry practice (e.g., ANSI C37 .96-1976, AC Motor Protection) would
maintain a ratio between inrush and MAG breaker setpoint of 175 -200%.
These low settings below 150% could result in premature tripping of the
MCCB resulting in failure of the valve to perform its intended safety function.
The licensee responded that they had never experienced a trip of the MCCB
during valve testing. Nevertheless, the licensee agreed that setpoint to
locked rotor ratios of less than 150% was not. prudent and initiated a change
to these four MCCB setpoints to reduce the* risk of premature failure*:
3.
. CC Pump Overcurrent Relay
The team reviewed the CC pump overcurrent relay setpoints as documented
in the maintenance department's relay test orders and compared those
values with the relay settings depicted in calculation ES-13.006(0), Breaker
and Relay Coordination, Rev. 2, dated October 18, 1995, and Drawing
203117, Rev. 24, dated March 11, i 996. The team observed that the relay
settings were consistent between the two documents. However, the team
noted the CC pump motor data (full load and locked rotor amps) contained
on the drawing differed slightly from the motor outline drawing 209C219,
Rev. 6, and the motor nameplate data. The team noted that the coordination
curves used generic motor data for motor acceleration, and the time current
plots failed to contain any motor thermal capability information. The team
reviewed the assumed motor acceleration data to verify the use of generic
curves was acceptable. The calculation assumed all pumps accelerated in
one second, but the overcurrent relays for all the safety-related 4160 Volt
motors (except 21 AFW) were set to allow locked rotor current for
approximately 15 seconds. The team reviewed traces of the
January 23, 1996, loss of offsite power loading test of EOG 2A and
confirmed the CC pump accelerated to full speed within one second with an
average voltage of 4200 Volts applied. The team also estimated the
acceleration of the CC pump with a minimum specified 70% voltage would
be less than four seconds. Therefore the use of generic motor accelerating
curves was acceptable for the CC pump motor.
The calculation did not include motor thermal damage capability curves.
Therefore, it was not possible to determine if the relays provided sufficient
running or locked rotor protection. The team reviewed the motor
specification 78-1303, dated July 26, 1978, and specification 85001, dated
December 13, 1985, and confirmed that motor thermal capability information
should have been available. The licensee was not able to find motor thermal
32
damage criteria in its document files. The motor manufacture was also
unable to find Salem specific CC pump motor information. However, the
motor manufacturer was able to find motor data for a motor similar to the CC
pump motor supplied to another nuclear facility. The team found the thermal
damage curve for the similar motor was not fully protected in the locked
rotor condition by the settings used for the Salem CC pump motor
overcurrent relays. However, as noted above, the relay setting would ,allow
greater than 15 seconds locked rotor condition prior to tripping. This
discrepancy would only affect the CC pump motor if it had failed to start and
was not considered an operability concern by the team because in order for
the relay to malfunction the CC pump would have already failed to start for
another unrelated reason. The licensee considered the existing relay setting
provided adequate motor protection during normal operation. The team
noted that the long time overcurrent setting would permit a 200% overload
and would not inhibit operation of the CC pumps. The licensee considered
some potential' rnotor *damage *under *a* 1ocked rotor condition to *be
acceptable.
c.
Conclusions
The team concluded that there. were significant weaknesses in the calculation for
the selection of the thermal overload relays for the CC system MOVs. The selection
of the new TOL heaters involved a lack of design and calculation control. The lack
of quality in the calculation indicated that the calculation preparer and the reviewer
failed to fully understand the operation of the equipment or the referenced
manufacturer's information.
The design change to place the TOL heaters in service resulted in the installation of
TOL heaters without a documented design basis. The team concluded that the
licensee had not maintained document control of the TOL relay heaters associated
with the CC system and other safety-related systems because heater sizes existed
in MOV circuits that were not based on the existing calculated basis. The team also
concluded the change document to the design calculation did not provide any
documented basis to accept the irn~talled TOL heaters for 30 safety-related MOVs.
The team found inappropriate design control in the selection of the molded case
circuit breakers associated with the CC system valves. The team identified four
magnetic trip MCCBs that had the potential for a premature trip (Valves 21 CC3,
22CC3, 2CC30 & 2CC31 ). The licensee completed corrective actions to reset
these breakers prior to the conclusion of this inspection with minor modifications to
S-96-025 and S-96-026.
The team also identified some minor weaknesses in the documentation and
assumptions made in the calculation for the selection of the CC pump motor
overcurrent relays .
E1 .5
a.
33
The team concluded that these issues remain unresolved pending the completion of
the licensee's corrective actions and the review of these issues by the NRC for
potential enforcement action (URI 50-311196-81-11 ).
Setpoint Control
Inspection Scope
The team reviewed selected instrument setpoint calculations to ensure that the
setpoints had a technically sound basis.
b.
Observations and Findings
1.
2RM17 - Surge Tank Vent Isolation Radiation Monitors
The UFSAR, Section 9.2.2.4:3; indicates that the CC surge tank is open to
.the atmosphere, but if high radiation is detected in the recirculation system,
the vent line is automatically closed.- The inspectors questioned the setpoint
basis for the radiation monitor alarm and safety function and were informed
that the setpoint was established to detect when the concentration of
. radioactive material in the component cooling system was approximately that
of primary coolant. The basis for the setpoint was to ensure that gaseous .
releases via the waste gas header were low.' In response to the team's
questions, the licensee's engineers stated that the radiation monitors
associated with the CC surge tank vent line appear to be set substantially
above the setpoints documented in either the Westinghouse Precautions
Limitation and Setpoint Document, Table 4.5, (Rev. 7, August 1979) (VTD-
304209) or the CC system configuration baseline document, DE-CB.CC-
0023(0), Rev. 1, Table T-1. Both of these documents indicated a setpoint
was 0.5 decade above minimum sensitivity. UFSAR Section 11.4 describes
these process radiation monitors, and UFSAR Table 11.4.2 indicated the
setpoint for 2-R17 A and 2-R178 was 1.0 E(-)7 micro curies per cubic
centimeter. Maintenance Procedure S2.IC-CC.RM-0027(0), Rev. 3,
December 11, 1996, which indicated a trip setpoint of 1.8E4 counts per
minute. PSE&G informed the team that an evaluation of the radiation
monitors setpoints and their bases would be conducted.
2.
2Ll-628A.C Cooling Component Cooling Surge Tank Level Alarms
The team reviewed Calculation SC-CC003-01, Rev. 1, Setpoint Relationships
- CC Surge Tarik Level, dated April 23, 1996, which established the
uncertainties associated with the high and low level alarms for the CC
system surge tank. The range for the alarms were established by the
Westinghouse Precautions Limitations and Setpoints document (Rev. 7,
34
August 1979) (VTD-304209) for the prime purpose of detecting leakage into
or out of the CC system. The team noted that the calculation concluded that
the process limits were not defined and required future configuration to
assure the existing setpoints were adequate for the alarm limits. The team
found that no action had been established to resolve this (or other) open
items from the setpoint calculations.
c.
Conclusions
The team concluded that the design basis documentation for the CC system
radiation monitors were inconsistent. The team also determined that the CC
radiation monitor setpoints may be inappropriately set to high. These radiation
monitors are not safety-related and are not used to calculate offsite radioactive
releases.
The'team* also identified *that' the design 'basis setpoint calculation for the *surge tank
level alarms .contained missing information that was identified in the body of the
calculation, but had not bee_n included in a system to track its resolution.
The calculation deficiencies are an unresolved item pending the completion of the
licensee's corrective action, including an evaluation to determine if the radiation
monitor setpoint was consistent with the UFSAR, and the review of these
deficiencies by the NRC for potential enforcement action (URI 50-311/96-81-12).
E1 .6
Equipment Power Supplies
a.
Inspection Scope
The team reviewed the power supplies for selected CC system and supporting
system components to confirm proper electrical train separation.
b.
Observations and Findings
1.
CC Pumps
The team reviewed the 4160 Volt and 230 Volt MCC one line diagrams. The
team confirmed the three CC pumps were powered from individual safety-
related 4160 Volt ac busses .. These busses are fed from the preferred offsite
power supply and can be fed from the onsite emergency diesel generators if
required. The pumps are automatically loaded on the EDGs following a
LOOP and are manually loaded on the EDGs following a LOCA and
appropriate load shedding. The team questioned the emergency operating
procedure (EOP) guidance for the restoration of the CC system. *The team
noted that procedure 2-EOP-APPX-1, Rev.20, dated October 14, 1996,
Step 5, required load shedding the 21 cooler and three fans (21 switchgear
supply fan, 21 auxiliary building exhaust fan and the 21 containment fan
, cooling unit) prior to starting the 21 CC pump. The team noted the
.. : corresponding section in the EOG loading calculation (ES-9.002) did not
c.
2.
3 .
35
account for these load changes at the 600-second mark, but instead load
shed the containment. spray pump (317 kW) at the time 21 CC pump was
- added at 59 minutes. The team therefore questioned the availability of the
onsite power supply for CC pump 21 at the assumed 10-minute switchover
point.
The licensee agreed that the EOG loading calculation failed to address this
.scenario and issued an AR (961223215) to resolve this issue. This will
-remain -an unresolved item pending the licensee's incorporation of this
scenario into the EOG loading analysis and the review of this issue by the
NRC for potential enforcement action (URI 50-311196-81-13).
CC Motor Operated Valves
The team confirmed the motor operated valve~ were powered from 230 Volt
ac vital motor control centers (MCCs)* powered from the* A and C safety * *
. divisions. The one exception found by the team was the flow control valve
- for the number 22 RHR heat exchanger (22CC16), which was powered from
the B safety division. This .arrangement of power supplies permits isolation
of the two CC flow headers and allows CC flow to both safety divisions of
the CC system on loss of any one power supply.
CC Instruments
The team reviewed selected CC system instruments powered from the 11 5
Volt instrument buses and confirmed that those instruments were powered
from redundant vital instrument buses.
4.
Supporting Systems Power Supplies
The team reviewed the power supply for the fan units cooling the CC pump
rooms 21 and 22 and confirmed they were powered from safety related 230
Volt ac MCCs. The team noted that the number 22 pump (Train 8) room
also contains the number 23 CC pump (Train C) and was cooled by fan unit
2VHE34 powered from Train 8. However, the team found that the number
21 CC pump (Train A) was cooled by fan unit 2VHE33 powered from Train
C. Therefore loss of Train C could result in loss of both the number 21 and
23 CC pumps. The effect of single failures on multiple CC pumps are also
discussed in section 03.1, E1 .1, E1 .2 of this inspection report.
Conclusions
The team concluded that proper train separation had been provided for the CC
system components. However, the team concluded that the emergency power
supply for the 21 CC pump to support EOP APPX-1, step 5, had not been included
in the EOG loading analysis. Also, the team concluded that the association of two
fan units for the three CC pumps provided the opportunity for an additional single
failure not previously analyzed by the licensee.
-
--~ ------------
36
E3
Engineering Procedures and Documentation
E3.1
Electrical Calculations
a.
Inspection Scope
The team reviewed selected design basis calculations and analyses for the ac and
de electrical systems that support CC system operation. The de system is required
for pump control, pilot solenoid valve power and control, control center low voltage
control, and EOG control and loading. The ac system is required for CC pump
power and MOV power and control.
b.
Observations and Findings
1.
CC Control Circuit Voltage Drop
The team observed that calculation ES-15.006(0), Rev. 2, dated
January 23, 1995, 230 Volt Vital AC Bus Control Power Circuit Voltage
Drop Study, did not address the MOV control circuit voltage drop for all
circuits including 7 of the 14 CC system MOV circuits. The licensee
responded that the calculation was based on the "worst case" circuits; The
team noted that the calculation failed to document and verify this
assumption. In response to the SSFI team finding, the licensee reviewed the
circuit lengths of the missing circuits and confirmed that these circuits were
enveloped by other analyzed circuits with similar size one motor starters .
. Therefore the voltage drop assumptions for the CC system MOVs were
acceptable.
2.
Load Flow CC Pump Starting Capability
The team observed that calculation ES-15.004(0), Rev. 2, dated
October 7, 1996, Load Flow and Motor Starting, was based on a degraded
voltage condition of 0.932 Per Unit (PU). This value was less than the
present degraded voltage setpoint and adds conservatism to the analysis.
However, the team noted that the CC pump motor was not included in the
analysis which included other safety-related 4160 Volt motors. The licensee
responded that the 300 horsepower (hp) CC pump motor would be
enveloped by the larger motors (400 and 600 hp) on the 4160 Volt buses.*
Therefore the calculation was also acceptable for the CC pump motor. The
calculation had not documented this engineering judgement .
37
3.
Battery Charger Recharge Load on the EOG
The CC pumps are automatically loaded onto the EOG following a loss of
offsite power (LOOP) and manually loaded onto the EDGs following a safety
injection signal coincident with a LOOP. The team observed that calculation
ES 9.002(0), Rev. 2, dated October 14, 1994, Emergency Diesel Generator
Loading, indicated very little margin between the required load and the 2
hour EOG rating.
The team observed that the calculation included detailed loading analysis for
all three EDGs for several different cases including LOOP, LOOP/LOCA with
all EDGs available, and LOOP/LOCA with one EOG not available. The loading
analysis for the 28 EOG was the most limiting because the calculated margin
below the 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> EOG rating was 68 kilowatts (kW) at a nominal frequency
of -60 hertz (Hz).
The team reviewed the calculation of the battery charger contribution to the
EOG loading. This calculation was based on measured battery charger
output during normal plant operation. The team observed the calculation
incorrectly adjusted the measured output of the battery charger resulting in
an incorrect conclusion that the input power was less than the output power.
The licensee had previously acknowledged this error and had appropriately
issued a change against the calculation prior to this inspection.
In addition, the team noted that the calculation was based on a 250 Amp
charger and the chargers had been replaced with 300 Amp chargers under
design change (DCP) 2EC-03332, approved April 1, 1996. The EOG loading
calculation had been identified as a document that required change as part of
that modification ancl included a change document (CD E511 /0} against the
calculation identifying the change details.
Following a loss of offsite power and prior to the diesel generators supplying
power to the battery chargers, the batteries begin to discharge to supply the
de loads. Calculation ES-4.006(0), Rev. 0, 125 Volt de Component Study
and Voltage Drop, indicates the initial battery voltage could drop to 113.16
volts and the battery cable resistance is less than .0007 ohm. When volt(!ge
is returned to the battery charger, its output voltage will attempt to return to
132 Volts direct current (Vdc), resulting in the charger going into current
limit. The team found that the calculated battery charger load on the EDGs
neglected any battery recharging current, following a partial battery
discharge, that could result in the battery chargers going into a current limit
mode. This additional load was not included in either the original calculation
or in the two changes that were outstanding against the calculation .
c.
"~38
The team considered the lack of adequate justification for not including the
battery recharging current in the battery charger load in the diesel generator
load calculation to be an unresolved item pending the licensee's evaluation of
this issue and the NRC review for potential enforcement action. (URI 50-
311196-81-14).
4.
DC System Time Constant
The team observed that calculation ES-4.003(0), Rev. 1, dated
January 18, 1996, 125 Volt DC Short Circuit and System Voltage Drop,
included as .an attachment a letter from the manufacturer of the battery main
fuses. This letter provided confirmation for the de interrupting rating for the
fuses. The team noted the letter included a statement that the rating was
only applicable if the de system time constant was within the requirements
of the Underwriters Laboratory (UL) standard {UL-198L, DC Fuses for
- lndustrial""Use) used to test* the fuses. The* licensee* could *not produce any
evidence that the circuit time constant was ever reviewed to address the
- fuse manufacturer's caution. The fuses in question were the main fuses
from the battery. They would* only be called on to operate for a major fault
on the de bus. While the team does not consider this to be an operability
concern, the team did considered this to be an undocumented engineering
judgement. The licensee initiated an action request (AR 961130114) against
the calculation to document the time constant of the de system .
Conclusions
The team concluded that the licensee failed to document a number of engineering
judgements and assumptions. The calculation review and approval process also
failed to identify and correct the lack of rigor in documenting assumptions and
engineering judgements. While the missing engineering judgements and
unsubstantiated assumptions did not invalidate the results of these calculations,
other unsubstantiated assumptions, used in the CC NPSH and TOL heater
calculations, did invalidate the calculation results.
E3.2
Technical Standards
a.
Inspection Scope
As part of the team's review of issues identified during this inspection, the team
reviewed the guidance available to the engineering staff in the form of technical
standards.
b.
39
Observations and Findings
1.
Low Voltage Circuit Breakers and Combination Starters
The team reviewed technical standard ND.DE-TS.ZZ-2012(0), Rev. 0, dated
July 7, 1994, Low Voltage Circuit Breakers and Combination Starters during
the review of the selection* of thermal overload relays and molded case
circuit breakers for the CC system motor operated valves. The team noted
the technical standard was generally consistent with present engineering
practice endorsed by recent IEEE standards (IEEE-741-1990). However, the
section on MCCB selection recommended an instantaneous setting range of
185% to 235% of the motor's locked rotor current. While this range of
settings would have avoided the risk for premature tripping of CC system
MOVs, it did not address the TOL manufacturer's requirement for short
circuit protection for the combination starter components. This requirement,
contained* in the licensee's vendor technical *data (VTD) number 317-235-01'
(GE.instruction GEH-5091 ), and included as a reference to Calculation ES-
18.006, Rev. 0, June 6, 1994, indicated the importance of providing short
.Circuit protection for the thermal .overload relay.
The team also noted that Attachment 4 to the standard listed a number of
TOL heaters as non-safety related. Two of those heaters were used in the
CC system and included in the ES-18.006 calculation. The licensee could
not identify the bases for the non-safety related label, but did confirm that
the TOL heaters of concern in the CC system were listed in the Salem Unit 2
. Bill of Material as safety-related.
2.
Medium Voltage Motor Protection
The team reviewed technical standard ND.DE-TS.ZZ-2014(0) , Protective
Relaying for 4.16 kilo Volts (kV) and 7 .2 kV Susses, Rev. 1, dated
December 13, 1995, as part of the review of the protection* for the CC pump
motors. The team noted that the standard's recommendations for the long
term and instantaneous relay settings agreed with industry recommendations
(ANSI C37 .96) and the actual settings on the CC pump motor overcurrent
relays generally agreed with the standard. The team noted that the
standard's recommendation for the time dial selection correctly stated that it
should be picked to lie between the motor acceleration curve and the motor
thermal damage curve. This information was missing from the Salem
calculation ES-13.006, Rev. 2, dated October 18, 1996, and contributed to
the discrepancy noted in Section E1 .4 of this report. The standard did not
address selection guidelines for using generic motor data. The Salem
calculation assumed the same one second motor acceleration curve for all
4. 16 kV motors without any basis. The team confirmed this was a good
value for a 100% voltage start of the CC pump required by the Salem motor
specifications number 78-1303, Rev. 0, dated July 26, 1978, and number
I
L
40
85001, Rev. 0, dated December 13, 1985, from a review of the EOG loading
. test data but also noted that the service water pump required two seconds
to come up to speed. This did not address minimum voltage starting at 70%
voltage with the potential for a longer acceleration time.
3.
Electrical Installation
The team reviewed technical standard SC.DE-TS.ZZ-2034(0), Rev. 3, dated
July 30, 1996, following a walkdown of the electrical distribution equipment
associated with the CC system. The team observed numerous examples of
power cables, in the 84 foot elevation de equipment area, that were
unsupported for distances of six feet or more, including the ac power feeds
to the new 300-amp battery chargers recently installed under DCP-2EC-
3332, Rev. 0, dated April 1, 1996. The team noted that the standard,
paragraph 5.2.27, specified that the maximum. free air length [of cable]
- should be nominally 3'-0"*or twice the*min'imum bend .. radius,'whichever is
greater and that this paragraph had not been changed in the latest revision.
The licensee responded to these discrepancies between the recently issued
technical standards and the existing plant conditions by stating that the
intent of the standards was for new work. Although the new battery
charger did not change the ac power cables, the team felt the cables should
have been supported between the raceway and the new chargers to the
recent guideline. The licensee prepared an Action Request (961212228) to
address the generic concern of these discrepancies.
c.
Conclusions
The team concluded that the development of the technical standards program was a
positive initiative by the licensee. However, the team noted that the standards did
. not include a technical justification for the acceptability of existing conditions in the
plant. The team considered this to be a program weakness. .In addition, the team
identified one case where work in progress, during the development of the technical
standards, was not coordinated with the technical standard the licensee was
developing at the same time. The licensee issued an Action Request to address the
practice of not evaluating existing conditions.
E3.3
Drawing Control
a.
Inspection Scope
The team conducted walkdowns of the CC system to verify that plant drawings
were consistent with the installed plant equipment.
41
b.
Observations and Findings
The team noted several minor discrepancies with component description on the
labels attached to CC components. For example, the team noted that the label on
- the steam generator blowdown cooler inlet valve was incorrectly labeled as the
outlet valve. The licensee appropriately issued ARs to track the resolution of these
deficiencies.
The piping and instrument diagrams (P&ID) for CC were generally accurate. The
team noted that the labels on the steam generator blowdown cooler heat
exchangers were not consistent with the P&ID. The licensee revised the P&ID
(Sheet 2, Rev. 33) to properly document the component identification numbers.
The team also noted that the P&IDs did not indicate identification numbers for
instrumentation located on the post accident sample .coolers. The licensee issued
an AR to track and resolve this discrepancy.
The team identified that DCP 2SC-2154 did not update the P&ID to remove FM
601 A&B and DCP 2EE-0248 did not update the instrument loop diagram for FIC
642A&B. In response~ the licensee issued ARs 961130118 and 961206159 to
address these and related drawing errors.
Section. E1 .4 of this report also addresses discrepancies between the 230 Volt vital
MCC bus one line diagram and the installed TOL heater and related design
documents.
c.
Conclusions
The team concluded tha*t the CC system drawings were generally accurate. The
licensee initiated actions to correct the minor discrepancies identified by the team
for both the. drawings and plant equipment labels. *
E3.4
Configuration Baseline Document
a.
Inspection Scope
The team reviewed selected sections of the Configuration Baseline Document (CBD)
for the component cooling system. The team also sampled several design
calculations, engineering evaluations, and other design documents to assess the
accuracy of the CBD and its supporting design inputs.
b.
Observations and Findings
The CBDs were developed and issued final for use during the 1988 to 1992 time
frame. However, in some cases CBDs were found to be impacted by processes
outside of the design change process (e.g., revisions to calculations and engineering
evaluations). These types of changes to the CBD were incorporated without design
verification. In a memorandum, dated July 1, 1996, all licensing and engineering
personnel were directed to use qualified source documents (calculations,
42
engineering evaluations, etc.) in CBDs rather than the CBDs themselves to make
engineering decisions. The CBDs were to be used as a "Road Map" to identify
these source documents. The CBD validation effort is in process to fully validate
the design basis information contained in those documents.
The CBD provided a comprehensive index of CC design basis calculations. In
general the team found that calculations were readily available and, in the case of
those maintained on the Document Management System (OMS), easily retrieved.
c_.
Conclusions
The team concluded that the CBD was a good source of design information and was
properly controlled by the licensee. The team found that CC design calculations
were referenced in the CBD. Calculations on the OMS were readily available.
Miscellaneous* Engineering Issues
E8.1
Post Accident Sampling System Heat Exchangers
a.
Inspection Scope
The team reviewed the interface between CC and the PASS to verify that operating
practices were consistent with the licensing basis.
b.
Observations and Findings
. _ The UFSAR, Sec~ion 9.3.6.1, states that, "The PASS provides the capability to
obtain, under accident conditions, a containment air grab sample, liquid and stripped
gas reactor coolant grab samples ... " Secti_on 9.3.6.2 states that, "Ten gallons per
minute of component cooling water is supplied to the sample cooler rack to cool
reactor*coolant samples. n
Emergency Operating Procedure, 2-EOP-TRIP-1, Reactor Trip or Safety Injection,
Step 17, provides instructions requiring the closure of the boric acid evaporator CC
outlet valve (2CC48). Isolating the boric acid evaporator also isolates CC water
from the PASS heat exchangers. The boric acid evaporator CC outlet valve remains
closed throughout the duration of an accident. Therefore, in accordance with _the
current EOPs, CC is not available to provide PASS heat exchanger cooling flow
during an accident.
In response to this concern, the licensee provided procedure SC.CH-AB.CC1155(Q),
Revision 0, Temporary Cooling of PASS Cooler Rack 811, which is used to restore
cooling to the PASS cooling rack (Panel 811 ). The team reviewed this procedure
and found that it provides for installation of temporary hoses to supply cooling
water (demineralized water) to the PASS when component cooling water is not
available. However, this temporary installation is not consistent with the
description of the cooling water supply provided in the UFSAR or with NUREG 0737
evaluations for the PASS. The licensee issued AR00961212177 to track the
resolution of this discrepancy.
43
c.
Conclusions
The team concluded that providing PASS heat exchanger cooling water from the
demineralized water system is inconsistent with the UFSAR. This issue remains
unresolved pending: (1) the completion of the licensee's evaluation, including an
assessment to determine if the operating procedure change for using deminerlized
water instead of CC was conducted in a manner consistent with 1 OCFR 50.59 and
50. 71 (e); and, (2) review by the NRC for potential enforcement action
(URI 50-311196-81-15).
E8.2
Licensing Basis Verification *
a.
Inspection Scope
The team compared the UFSAR description of the CC (Section 9.2.2, Component
Coating System) and suppc>rt systems with the design 'basis information to verify
that the UFSAR descriptions were accurate. The team also reviewed the licensee's
adequately reviewed the UFSAR.
b.
Observations and Findings
The team identified examples where the description material in the UFSAR did not
Clearly reflect the CC system design. For example:
Section 9.2.2.8.1,. states in part that "The reactor coolant pump bearing
temperature alarm is set at 175 degrees Fahrenheit." The next paragraph
which also discusses the reactor coolant pump bearing temperature alarm
states that "The maximum test temperature of 185 °F is also the suggested
alarm setpoint ... ". The team found that these two statements were not
consistent.
Section 9.2.2.3, states in part that "The operation of the system is
monitored with the following instrumentation: 3. A temperature indicator in
the outlet line from each heat exchanger". The team noted that the steam
generator blowdown sample heat exchangers did not have a temperature
indicator in the outlet line.
Section 11.4.2.2, states in part that "These channels (CC radiation monitors
2-R17 A,8) continuously monitor the component cooling water for radiation."
The team noted that plant procedures do not require one or both CC
radiation monitors to be continuously inservice when the CC system is in
operation. The team also noted that both CC radiation monitors* had been
out-of-service for an extended duration at the start of this inspection (See
Section M2.1 ).
The team reviewed the UFSAR Macro-Review for the CC system. The Macro-
Review was one part of the licensee initiative to validate the information provided in
44
the UFSAR. The CC Macro-Review was conducted by one engineer during a one
week duration. The CC Macro-Review verified 57 UFSAR attributes and identified 6
discrepancies. The identified discrepancies were generally descriptive errors that
did not adversely affect the CC system function. For example, the Macro-Review
identified that the UFSAR incorrectly stated that CC provides makeup water for the
waste gas compressor seals. The Macro-Review correctly identified that the CC
system is not capable of performing this function. The identified discrepancies for
- . the CC system were all appropriately placed in the corrective action program.
The attributes were verified by identifying a document which substantiated the
statement in the UFSAR. It was not the intent of the Macro-Review to conduct an
in-depth validation of the supporting documentation. For example, if the attribute
was each CC heat exchanger is designed to remove 1 /2 the decay heat 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />
after plant shutdown, then the calculation validating this information would be
referenced in the Macro-Review. The engineer conducting the Macro-Review would
- *not *necessarily revrew the talcl..llation to verify that the* assumptions were valid and.
the calculation was based on sound engineering principles.
The team compared the administrative guidance provided in procedure S2.SE-
DD.ZZ-0008(Z), System Engineering Final System Readiness Review UFSAR Macro-
Review Desk Guide, with the CC Macro-Review. The team noted a few examples
where the attributes recommended for selection by the desk guide, such as
setpoints, were not selected for verification during the CC Macro-Review. For
example, the RCP bearing alarm setpoint was not selected. The team also noted
one example, regarding the absence of the outlet temperature indicator on the
steam generator. blowdown sample heat exchanger, where the attribute was
selected; however, the Macro-Review did not identify the discrepancy between the
pl~nt and UFSAR description. *
The team reviewed the auxiliary building UFSAR project Vertical Slice to determine
if the single failure of the CC room ventilation issue identified during this inspection
was identified during the Vertical Slice. The team concluded .that this issue was not
explicitly identified during the Vertical Slice review. However, the licensee had
initiated a single failure evaluation to review ventilation systems for single failures.
The licensee's ventilation engineers stated that the preliminary evaluation had
identified that a single ventilation failure would affect multiple CC pumps.
c.
Conclusions
The team concluded that the CC licensing basis descriptions (UFSAR) were, with a
few minor exceptions,* consistent with the actual plant design. The team concluded
that the CC UFSAR Macro-Review was a good initiative and identified and corrected
several UFSAR discrepancies. However, the team noted that there are significant
scope and methodology differences between the Vertical Slice/Macro-Review and
an SSFI. Therefore, it was not the primary purpose of the Vertical Slice/Macro-
Reviews, to identify design issues, such as, the ventilation and pump runout issues
that were identified during* this inspection.
-.
45
E8.3
Licensing Basis Updates
a.
Inspection Scope
The team reviewed the timeliness for two licensing basis changes.
b.
Observations and Findings
During the team's review of the EDG loading Calculation ES-9.002, the team noted
the analysis concluded that the load margin must be maintained by restricting the
frequency of the EDG to no greater than 60.5 Hz. A maximum frequency would
limit the maximum pump and fan speeds and limit their driven load. This then
would limit the increase in load to below the EOG' s 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> limit. The calculation
referenced an incident report (94-301, dated October 13, 1994) and a licensing
change request (LCR) 94-40 which addressed the change in the EOG technical
speeification. Technical Specification 4:8:1. l prese-ntly liniits the acceptable
frequency range to 58.8 - 61.2 and the proposed change would set the allowable
range to 58.8 - 60.5 Hz. The LCR 94-301 had not been issued but had been
replaced by LCR S95-36. LCR S95-36 had been sent. to the NRC for approval on
September 26, 1996. The team was concerned with the delay in issuing the LCR
which supports the EDG operability. In response to the team's concern, the
licensee indicated the delay was due to combining the original LCR 94-40 with
another LCR into the new LCR S95-36. There was no explanation why this new
LCR was not released until September 1996.
The team reviewed the licensee's EDG governor test and setup procedure SC.MD-
CM-OG-0006(0), Rev. 8, dated June 12, 1996~ and noted that the allowable
frequency range of the latest governor setup procedure would have supported the
proposed technical specification. (Refer to Section M3.2 of this report for a
discussion on the as-found acceptance criteria.)
The team also noted that a proposed 1995 LCR S95-31, Component Cooling - Add
Third Pump, has not been submitted to the NRC. In addition, actions had not been
initiated to reflect this administrative control requirements in the UFSAR or technical
specification bases. The licensee's system readiness review indicated that the
resolution of this item was scheduled for prior to plant restart.
c.
Conclusions
The team concluded that PSE&G was untimely in submitting their licensing change
request for the EDG frequency limit requirement. In addition, incorporation of the 3
operable CC pumps administrative requirement into the licensing basis was also
untimely.
46
E8.4
Probabilistic Safety Assessment
a.
Inspection Scope
The team reviewed the Salem Probabilistic Safety Assessment (PSA) fault trees to
verify that the CC system design was properly modeled in the PSA fault trees. The
team also verified that dependencies between CC and other safety-related systems
were *consistent with the design information.
b.
Observations and Findings
The team noted that the PSA description of the CC system operation and fault tree
was not consistent was the system design basis. The licensee's PSA engineers
made the following changes to address the team's findings:
.*
Separate. CC fault tre*es* were developed for GC normal operation and CC .
operation when entering cold leg recirculation following a loss of coolant
.*accident. The cold leg recirculation fault trees included separating the two *
- CC headers as required by the emergency operating procedures.
The Human Error Probability for the operator action to transfer to cold leg
recirculation was recalculated to reflect the train separation.
A new operator action was created to repre~ent restarting a CC pump and
un-isolating the CC heat exchanger service water after a loss of offsite
power and a safety injection signal both occur.
The team also noted that the PSA model incorrectly assumed that the CC to the
charging pump seals was not required for pump operation. In response to this
finding, the licensee's PSA group created a new event tree to model the response to
a loss of all CC as an initiating event. This had previously been excluded as an
initiating event because of the incorrect assumption that the charging pumps were
not dependant on CC. There is a potential that the loss of CC will result in a loss of
the charging pumps and reactor coolant pump thermal barrier cooling. Under certain
conditions, this could result in a RCP seal LOCA.
The preliminary CC model update resulted in a core damage frequency (CDF)
change from approximately 4.4E-5/year to 8.13E-5/year. The new model assumed
that a loss of the 22/23 CC pump room cooler would result in both CC pumps *
failing to function. The licensee's engineering staff was conducting an analysis to
determine what action*s could be implemented to break the dependency between the
CC pump room coolers and CC pumps. A decrease in CDF will be achieved if
corrective actions eliminate this dependency.
The PSA group with the assistance of operations staff reviewed the PSA
descriptions for several safety-related systems to determine if the modeling errors
identified were prevalent throughout the PSA. The operations staff review did not
identify any other significant discrepancies.
47
c.
Conclusion
The team concluded that there were CC modeling errors that adversely affected the
calculated plant CDF. The licensee corrected the identified errors and performed a
review to identify similar errors in the individual plant evaluation (IPE). The team
determined that the licensee's actions to resolve this issue were appropriate.
V. Management Meetings
X1
Exit Meeting Summary
The team discussed the team findings with the licensee staff and management
before leaving the site on December 13, 1996. The team presented the inspection
results to members of licensee management at the conclusion of the inspection on
January 8, 1997. The exit meeting was open for public observation. The slides
used* at the. exit meeting* are provided* as *Enclosure 1 *to this report. *The licensee
acknowledged the findings presented.
No proprietary material was knowingly retained by the team* or disclosed in this
- inspection report. The SSFI team question data base that was developed by the
licensee will be maintained onsite as a quality controlled record .
I L
l
48
PARTIAL LIST OF PERSONS CONT ACTED
V.J. Chandra, Engineering
T. DelGaizo, Contractor
D. Dodson, Licensing
J. Dunn, Westinghouse
L. Ford, System Manager
M. Hoskins, Engineering
K. King, Engineering
D. Lounsbury, Operations
W. Maher, Engineering
D. McHugh, Senior Engineer
G. Overbeck, Director System Engineering
D. Powell, Licensing
J. Raymond, Westinghouse
Drawings
49
DOCUMENTS REVIEWED
205331 A8763-46, Rev. 46, dated 7/3/96, No. 2 Unit Component Cooling, Sheet 1 of 3
205331 A8763-32, Rev. 32, dated 6/21/96, No. 2 Unit Component Cooling, Sheet 2 of 3
205331 A8763-35, Rev. 35, dated 6/27/96, No. 2 Unit Component Cooling, Sheet 3 of 3
205337 A8763-19, Rev. 35, dated 12/5/96, No. 2 Unit Auxiliary Building Ventilation,
Sheet 2
205337 A8763-2.1, Rev. 21, dated 12/29/96, No. 2 Unit Auxiliary Building Ventilation,
Sheet 3
207491 A8803-24, Rev. 24, No. 1 Unit-Auxiliary Building Component Cooling Piping,
Plans Elev. 45', 55', 64', 100', & 122'
205328-SH1, Rev. 46, Chemical & Volume Control System P&ID
205328-SH2, Rev. 53, Chemical & Volume Control System _P&ID
205328-SH3, Rev. 39, Chemical* & Volume Control System P&ID
.205342-SH3, Rev. 62, Service Water Nuclear Area P&ID
205342-SH4, Rev. 51, Service Water Nuclear Area P&ID
209C219, Rev. 6, CC Pump Motor Outline Drawing
203828, Rev. 20, SWP 21 125V DC Schematic
203834, Rev. 17, SWP 22 125V DC Schematic
220942, Rev. 16, SW Inlet Control Valves to CC HT EX 21 & 22
322509, Rev. 21, 2B Ventilation 230 V Vital CC One Line*
322510, Rev. 20,-2c Ventilation 230 V Vital CC One Line
211517, Rev. 9, .21 CC Pump 125V Schematic
211518, Rev. 4, 22 CC Pump 125V Schematic
-
-
211520, Rev. 4, 23 CC Pump .125V Schematic
211516, Rev. 4, 21 CC Pump 28V Schematic
211519, Rev. 9, 22 CC Pump 28V Schematic
211521, Rev. 7, 23 CC Pump 28V Schematic
601685, Rev. 4, 2CC17, 21CC3, 2CC30 MOV 230/115V Schematic
601683, Rev. 5, 22CC3, 2CC18, 2CC31 MOV Schematic
211527, Rev. 16, 2CC117, 2CC136 MOV Schematic
211528, Rev. 24, 2CC118, 2CC113, MOV & SOV Schematic
216911, Rev. 14, 2CC131 MOV Schematic
218846, Rev. 15, 2CC187 MOV Schematic
21884 7, Rev. 16, 2CC190 MOV Schematic
224384, Rev. 9, 2CC215 SOV Schematic
211526, Rev. 4, 2CC117, 2CC136 28V Control
211 530, Rev. 14, 22CC 16 MOV Schematic
211529, Rev. 15, 21CC16 MOV Schematic
601686, Rev. 1, 22CC3, 2CC18, 2CC31 28V Control
601684, Rev. 0, 2CC17, 21 CC3, 2CC30 28V Control
211524, Rev. 10, 2CC149 28V Control
242625, Rev. 4, RM System Alarms - Sheet 1
242626, Rev. 3, RM _System Alarms - Sheet 2
211522, Rev. 11, 21 Surge Tank and Header Pressure Alarms
211523, Rev. 10, 22 Surge Tank and Header Pressure Alarms
50
220178, Rev. 18, Interface Racks 41 & 126
236256, Rev. 7, Safeguard Equipment Control System
236259, Rev. 8, Safeguard Equipment Control System
236262, Rev. 8, Safeguard Equipment Control System
211357, Rev. 9, 28V DC Oneline
220804, Rev. 8, 2ADE 28V Distribution Cabinet Oneline
220805, Rev. 9, 2BDE 28V Distribution Cabinet Oneline
220806, Rev. 7, 2CDE 28V Distribution Cabinet Oneline
223720, Rev. 22, 125V DC Oneline
220812, Rev. 21, 2A 115V AC Vital Instrument Bus Oneline
220813, Rev. 20, 2B 115V AC Vital Instrument Bus Oneline
220814, Rev. 18, 2C 115V AC Vital Instrument Bus Oneline
222483, Rev. 29, 2A West Valve 230V Control Center Oneline
222484, Rev. 30, 28 West Valve 230V Control Center Oneline
222485, Rev. 37, 2C West Valve 230V Control Center Oneline
- 222505*, Rev. 23,* 2A East Valve 230V Control *center Orieline
222507, Rev. 24, .2.C East Valve 230V Control Center Oneline
203063, Rev. 31, 460V & 230V Vital and Non-Vital Oneline
601400, Rev. 13, 2A-230V AC Vital Bus Oneline
601401, Rev. 13, 2B-230V AC Vital Bus Oneline
601402, Rev. 11, 2C-230V AC Vital Bus Oneline
203061, Rev. 30, 4160V Vital Buses Oneline
601701, Rev. 13, Salem-Hope Creek 500kV, 138kV, 4.16kV One Line
203117, Rev. 24, 4160V Vital Buses Relay Settings
EOG 2A, 11 /23/96, Loading Test Visicorder Plot
203666, Rev. 9, Safeguards Emergency Loading Sequence SH1
203667, Rev. 7, Safeguards Emergency Loading Sequence SH2
203668, Rev. 6, Safeguards Emergency Loading Sequence SH3
203669, Rev. 7, Safeguards Emergency Loading Sequence SH4
203670, Rev. 11, Safeguards Emergency Loading Sequence SH5
203673, Rev. 6, Safeguards Emergency Loading Sequence SH6
228477, Rev. 14, Control Console Component Coding Water
622031 D, Rev. 0, Loop Diagram Vent Valve 2CC149
6220290, Rev. 1, Loop Diagram Excess Letdown Ht Ex Outlet 2CC113 (3 sheets)
6220300, Rev. 1, Loop Diagram Excess Letdown Hx Ex Inlet 2CC215 (3 sheets)
622013, Rev. 2, Loop Diagram, 2FT601 A (3 sheets)
622014, Rev. 2, Loop Diagram, 2FT601 B (3 sheets)
622017, Rev. 1, Loop Diagram, 2F1 C613
622020, Rev. 1, Loop Diagram, 2F1 C622
622019, Rev. 1, Loop Diagram, 2FIC619
622018, Rev. 1, Loop Diagram, 2FIC616
622021, Rev. 0, Loop Diagram, 2FIC645
622022, Rev. 0, Loop Diagram, 2FIC646
622027, Rev. 0, Loop Diagram, 2FIC643A
622028, Rev. 0, Loop Diagram, 2FIC643B
622032, Rev. 0, Loop Diagram, 2FIC642A
622033, Rev. 0, Loop Diagram, 2FIC642B
622034, Rev. 0, Loop Diagram, 2FIC625
..
51
622015, Rev. 1, Loop Diagram, 2LT628A (2 sheets)
622016, Rev. 0, Loop Diagram, 2LT628B (2 sheets)
622023, Rev. 1, Loop Diagram, 2PC600A
622024, Rev. 1, Loop Diagram, 2PC600B
622000, Rev. 0, Loop Diagram, 2TE672P
622001, Rev. 0, Loop Diagram, 2TE672Q
622006, Rev. 0, Loop Diagram, 2TA8463
622002, Rev. 0, Loop Diagram, 2TE672U
622003, Rev. 0, Loop Diagram, 2TE672V
622007, Rev. 0, Loop Diagram, 2TA8464
622004, Rev. 0, Loop Diagram, 2TE672L
622005, Rev. 0, Loop Diagram, 2TE672M
622008, Rev. 0, Loop Diagram, 2TA8465
622035, Rev. 1, Loop Diagram, 2TIC627 A
622010, Rev. 0, Loop Diagram, 2TE602C
- * 622011, Rev: 3, Loop Diagram, 2TE602A (2 sheets)
622026, Rev. 1., .Loop Diagram, 2TA9286Z
622036, Rev. 1, Loop Diagram, 2TIC627B
622009, Rev. 0, Loop Diagram, 2TE602D
622012, Rev. 2, Loop Diagram, 2TE602B (2 sheets)
622025, Rev. 2, Loop Diagram, 2TA9264Z
622037,. Rev. 1, Loop Diagram, 2TIC623
622038, Rev. 1, Loop Diagram, 2TIC624
21839-A-8902, Rev. 19, Salem Unit 2 Auxiliary Building Component Cooling Piping,
Mechanical Arrangement, Plans and Elevations
21839-S-8902, Rev. 20, Salem Unit 2 Auxiliary Building Component Cooling Piping,
Mechanical Arrangement, Plans and Elevations
Calculations and Engineering Evaluations
S-C-CC-MDC-0879, Revision 1, 6/23/92, Maximum and Minimum CC Pump Flow
Requirements and NPSH
S-C-CC-MDC-0860, Revision 0, 3/5/92, CC System Design Temperatures
S-C-CC-MDC-0575, Revision 0, 7/16/90, 11 and 21 CCW Heat Exchanger Data Sheet
Revision for Titanium Tubes
S-2-CC-MDC-0559, Revision 0, 6/28/90, 22 CCW Heat Exchanger Performance Evaluation
S-2-ABV-MDC-1622, Revision 0 IR1, 10/24/96, Auxiliary Building Pump Room
Temperatures
S-2-ABV-MDC-1666, Revision 0, 12/9/96, Interim Design Calculation-Maximum Outside
Air Temperature for Mode 6 Entry
S-C-VAR-MEE-1146, Revision 0, 11/1/96, Review Component Cooling & Service Water
System Piping Classifications - Salem
S-C-CC-MEE-0606-0, 7 /29/91, Service Water Pipe Cracks in Component Cooling (CC) Heat
Exchanger 12/22 Cubicle
S-C-CC-MEE-0596-0, 7/15/91, Containment Isolation Valves for Component Cooling
System
S-C-CC-MEE-0880-0, 2/25/94, Evaluation of Component Cooling System Operability with
Valves CC125 & CC146 Open
52
S-C-CC-ME-0605, 7/29/91, Component Cooling (CC) System Surge Tank Relief Valve
CC147 Set Pressure
S-C-CC-ME-0602-0, 7 /29/91, Component Cooling System* Isolation Following Thermal
Barrier Rupture
.
S-C-N21 O-MSE-269, 7 /31 /84, Potential Overpressurization of the Component Cooling
Water System
Westinghouse Calculation 3/10/67, Relief Valves for IPP#2, ACS (applicable to PSE&G)
S-C-4kV-JDC-959, Rev. 4, 6/18/93, Degraded Vital Bus UV Setpoint
ES-4.003(0), Rev. 1, 1 /18/96, 125 Vdc Circuit and System Voltage Drop
ES-4.004(0), Rev. 3, 5/29/96, 125 Vdc Battery and Battery Charger Sizing
ES-4.006(0), Rev. 0, 1 /18/96, 125 Vdc Component Study and Voltage Drop
ES-9.002(0), Rev. 2, 10/14/94, Emergency Diesel Generator Loading
ES-13.005(0), Rev. 5, 3/27/96, Penetration Overcurrent Protection
ES-13.006(0), Rev. 2, 10/18/95, Breaker & Relay Coordination Study
ES-15.004(0), Rev. 1, 10/7/96, Load Flow & Motor Starting.
ES-15J)06(0); Rev. 2, 1 /23195, *23ov Vitat* MCC Power Circuit Voltage Drop
ES-15.008(0), Rev. 2, 12/22/95, Degraded Grid Study
ES-18.006(0), Rev. 0, 6/6/94, Selection of TOL Heater Elements
SC-CC001, Rev. 1, 6/14/94, CC Ht Ex Outlet Temp Setpoint
SC-CC002, Rev. 1, 4/27/96, CC RHR Outlet Flow Setpoint
SC-CC003, Rev. 1, 4/23/96, CC Surge Tank Level Setpoint
S-2-CC-MDC-0898(003) and (004), MOV Capability Assessments for 21 and 22 CC16
Correspondence and Other Documents
DE-CB.CC-0023(0), Revision .2, 1 /5/94, Configuration Baseline Documentation for
Component Cooling Water
NLR-190194, 5/17 /90, B. A. Preston to V. Polizzi, Licensing Position on Passive Failures
Associated with the Salem Service Water System
NLR-194224, 6/14/94, D. A. Smith to H.G. Berrick, Service Water and Component Cooling
Water Design Bases Accident Conditions
LR-196102, 7/25/96, Updated Licensing Position on Passive Failures.Associated with the
Salem Generating Station Service Water System
Westinghouse PSE-89-744, 11/8/89, Salem CCW Calculation Summaries
TS2.SE-SU.CC-0001 (0), Revision 0, 1 OCFR50.59 Applicability Review, CC System Flow
Balance
Westinghouse PSE-84-802 (Re-ISSUED), 7 /26/84, Component Cooling Water System
Potential Overpressurization Notification
Westinghouse BURL-4031, 10/21 /81, Reactor Coolant Pump CC & Seal Injection Water
Loss
Westinghouse PSEB0-96-040, 9/3/96, Single Train Cooldown Analysis Report, Revision 1
Westinghouse PSE-94-605, 5/25/94, Revision 1 of Salem CCWHX SWS Flow Margin
Report (Post-LOCA Mode)
Westinghouse 20223, 5/26/89, Type 93A Reactor Coolant Pump for Surry Units 1 and 2
Estimated Flow Through Thermal Barrier Cooling Water Outlet Caused by Rupture of One
Heat Exchanger Tube
VTD No. 322553, dated 11/21/96, Evaluation of Component Cooling Pump at Runout
Condition, Revision 1, November 7, 1994 (MPR Associates, Inc.)
53
Bechtel Letter GSA-3679, 2/14/91, SWS Heat Exchanger Performance Evaluation -
Computer Runs with Dummy Values
DES-93-0300, 12/30/93
PIR 00950814345 CR
PIR 00960216322
PIR CR 00960925135
Letter LR-N96228, 9/25/96, PSEG to NRC LCRS95-36
~etter LR-N95042, 4/4/95, PSEG to NRC LCR 93-27
Letter NLR-N94169, 9/13/94, PSEG to NRC LCR 93-27
Letter NLR-N94108, 6/28/94, PSEG to NRC LCR 93-27
Letter NLR-N93196, 1/21/94, PSEG to NRC LCR 93-27
Letter 95-1158, 9/19/95, NRC to PSEG 125V DC TS 3.8.2.3
Letter, 12/6/96, C&D to PSEG 1993 Performance Test Evaluation
.Performance Improvement Request 960709221, Concern with Setting CC Flow through
RHRHX, issued July 18, 1996
PSE&G *Audit No. *95..:0125 of tn*service Testing Program*
Nuclear Training Center Lesson Plan 0299-095.04H-PANFLO-OO, Panametrics Ultrasonic
Portable Flowmter
.
..
.
.
.
Panametrics* Letter, dated November 22, 1996, forwarding .Certificate of Calibration for PT
868 Instruments - Serial Numbers 702, 157 and 158
.
Performance Improvement Request 961026073, Discrepancy with RHR HX CC Flow
Measurement
PSE&G Letters NLR-N90021, dated January 26, 1990, and NLR-N90165, dated
August 31, 1990, regarding Commitments to Generic Letter 89-13
Action Request 96031914 7 Regarding Repeat' Failures of Containment Isolation Valves
Action Request 961121204 Regarding Corrective Actions to Control Configuration of
Ventilation System Equipment
Action Request 961202179 Regarding Corrective Actions to Address the Manual Valves
Inconsistency between EOPs and the IST Program
Configuration Baseline Documentation
DE-CB-CC-0023(0), Rev. 0, 1 /15/94, CC System Configuration Baseline Document
Vendor Technical Document
304209, Rev. 7, August 1979, Westinghouse Precautions Limitations & Setpoints
317235, Rev. 1, 4/4/95, General Electric Thermal Overload Relay Data
317227, Rev. 1, 4/4/95, Reliance Motor Data 10 HP
317229, Rev. 1, 4/4/95, Electric Apparatus Motor Data 10 HP
317233, Rev. 1, 4/4/95, Limitorque Motor Data 0.2 HP
175421, Rev. 1, 9/12/94, Westinghouse CC Pump and Motor Data
910-142A-2, Two Channel Transport Model 2 PT868 Portable Flowmeter
54
Technical Standards
ND.DE-TS-ZZ-2012(0), 7/13/94, Low Voltage Circuit Breakers & Combination Starters
.ND.DE-TS.ZZ-2014(0), 12/13/95, Protective Relaying for 4.16kV & 7.2kV Buses
SC.DE-TS.ZZ-2034(0), 7/30/96, Construction of Electrical Installation
OPS Procedure
Emergency Operating Procedures
CC Abnormal and Normal Operation Procedures
2-EOP-APPX-1, Rev. 20, Component Cooling Water Restoration
Miscellaneous
SWEC TOL Walkdown, 11 /95 - 3/96, Fuse & breaker Verification Data Sheets
MMIS Component Data; CC system 230V *Mcc Pan Data Records
Specification
S-C-1978-DSP-1303, Rev. *o, 7126178, Spare Motors (78-1303)
S-C-1970-EGS-0048, Rev. 0, 6/25/70, Alternating Current Motors (70001-A)
S-C-EOOO-:-EGS-0115, Rev. 0, 12/13/85, Alternating Current Motors (85001)
Work Order 960214251, 6/25/96, 2C EOG MOP Replacement
960805170, 8/8/96, 2A EOG Governor Corrective Maintenance
960214210, 7/9/96, 2B EOG MOP Replacement
931124002, 5/11 /93, 2A 125V Station Battery Performance Test
950521003, 11 /14/94, 2C 125V Station Battery Performance Test
931124004,.5/12/93, 2B 125V Station Battery Performance Test
880412133, 10/18/88, .2A 125V Station Battery Replacement
880412134, 9/17/88, 2B 125V Station Battery Replacement
950725034, Rebaseline 21 CC Pump per Procedure S2;0P-ST.CC-0001 (0)
960202028, Rebaseline 22 CC Pump per Procedure S2.0P-ST.CC-0002(0)
950622037, Rebaseline 23 CC Pump per.Procedure S2.0P-ST.CC-0003(0)
960928055, Authorizing Troubleshooting Procedure to Collect Data for Evaluating
Electrical Vibration Levels for 21 , 22, -23 CC Pump
961112091, Authorizing Troubleshooting Procedure to Collect Data for Calibrating RHR HX
CC Flow Elbow Meters
960608023, Regarding Inspection and Cleaning of 21 CC Heat Exchanger
960528046, Regarding Inspection and Cleaning of 22 CC Pump Room Cooler
950924133, Correct Bent Spring Rod Pipe Support 2P-CCH""332
960919250, Replace CC Surge Tank Vacuum Breaker Valve 2CC148
960422158, Perform Set Pressure Test on Relief Valve 2CC112
- .*"
- ,
55
Modifications
2EC-3585/Pkg 1, 1 OCFR50.59 Safety Evaluation for CCW Letdown Temperature Control
Valve (CC71)
2EC-3249, 7/8/94, Cable Protection (TOL Replacements)
2EC-3332, 4/1 /96, 125V Battery Charger Replacement
2EZ-1478, 7/13/93, Valve 2CC71 Temperature Interlock
2EC-1015, 10/20/82, Addition of motor operators to CC Valves
2EC-0348, 8/3/83, Addition of motor operators to CC Valves
2EC-1014, 10/12/82, Addition of motor operators to CC Valves
Design Change 2E0-2340, Regarding Replacement Wedte Shoes for MOV 2CC117
Design Change 2E0-2425, Regarding Replacement of 1-inch Globe Valves 2CC281 and
282.
Evaluations .
s..:C-230-EEE-0753, 5/18/93, 230V Motor Operation During Degraded Grid
S-C..:230-EEE-0790-2, 7 /20/93, Motor Starting & Running During LOCA Block Start
8.3, Opsit~ Power System
9.2.2, Component Cooling System ..
11 .4~2.2, Process Radiation Monitor
Chapter 1 5, Accident Analysis
Technical Specification :
2/4.8. 1, AC Sources
3/4.8.2, Onsite Power Distribution
Procedures
S2.IC-CC.RM-0027(0), Rev. 3, 12/11 /96, 2R178 CCW Process Radiation Monitor
SC.MD-ST.ZZ-0005(0), Rev. 2; 5/2/96, MCCB Maintenance
SC.MD-PT.230-0001 (0), Rev. 1, 5/22/96, TOL Overcurrent Trip Testing
SC.MD-CM-DG-0006(0), Rev. 8, 6/12/96, DG Speed/Load Control System Alignment
SC.MD-ST.125-0004(0), Rev. 8, 5/16/96, 125V Station Batteries 18-Month Service Test
SC.MD-FT.125-0002(0), Rev. 4, 11 /15/95, 125V Station Batteries Performance Discharge
Test
82.MD-ST.125-0001 (0), 7/27/96, 125V Battery Charger Maintenance
TS2.SE-SU.CC-0001 (0), CC System Flow Balance
S2.0P-ST.CC-0001 (0), -0002(0), -0003(0), lnservice Testing - 21, 22, 23 Component
Cooling Pumps
S2.0P-ST.SW.0014(0), lnservice Testing, Room Cooler Valves
S2.0P-PT.SW-0026(Q) and -0027(0), 21 and 22 CC Heat Exchanger Heat Transfer
Performance Data Collection
Opened
50-,311/96-81-01
50-311 /96-81-02
£?0-311 /9 6-81-03
50-31 1,196-81-04
50-.311196-81-05
50-311196-81-06
50-311196-81-07
50-311196-81-08
50-311196-81-09 .
50-311/96-81-10
50-311/96-81-11
50-311196-81-12
50-311196-81-13
50-311/96-81-14
50-311196-81-15
56
ITEMS OPENED, CLOSED, AND DISCUSSED
CC pump room ventilation deficiency prior to 1995
.
Current EOPs are inconsistent with single CC pump room ventilation
failure
Current EOPs allow CC pump to runout which is not supported by
pump design documentation
No documented basis for CC flow balance acceptance criteria
No documented basis for CC heat exchanger performance test
assumptions and analysis
Lack of acceptance criteria for CC room ventilation coolers
CC radiation monitors not restored in a timely manner
CC pump room ventilation damper position is not controlled
Battery surveillance -test inadequacies
CC supply to pump seal water cooling heat exchangers
Inadequacy in TOL heater calculation and control
Inadequacy in setpoint calculations .for radiation monitors and surge
tank level alarm
EOG loading study discrepancy* with loading CC pump
EOG loading study discrepancy with battery charger
PASS operation inconsistent with UFSAR
BEP
CBD
cc
CCHX
ESQ
ft
gpm
hp
NRC
PSE&G
- psia
_
RHRHX
SGS
OF
oc
MAG
CV
A
v
kV
kW
w
C&D
CBD
ac
de
0/L
EOG
VTD
MOP
PU
57
LIST OF ACRONYMS USED
best efficiency point
Configuration Baseline Documentation
Component Cooling Water System
Component Cooling Heat Exchanger
Emergency Operating Procedures
Emergency Safeguards
feet
gallons per minute
horsepower
Loss of Coolant Accident
Net Positive Suction Head
United States Nuclear Regulatory Commission
Post Accident Sampling System
Pubtic Service* Bectrlc & Gas * *
pounds per square inch absolute
Residual Heat Removal .System
. Residual Heat Removal Heat Exchanger
Refueling Water Storage Tank
- Spent Fuel Pool Heat Exchanger
Salem Generating Station
Safety Injection
Updated *Final Safety Analysis Report
degrees Fahrenheit
thermal overload relay
degrees Centigrade
magnetic only (instantaneous) MCCB
chemical and volume control system
service water system
ampere
volt
kilo Volt
kilo Watt
C&D Charter Power Systems
radiation monitoring system
Configuration Baseline Document
alternating current
direct current
one line
motor control center
Licensing Change Request
Vendor Technical Document
motor operated potentiometer
per unit
.
"
ENCLOSURE 1
.
.
...... *
EXIT MEETING *S*LIDES.- * * * * * * *
- * * * *
SALEM UNIT 2
SAFETY SYSTEM
FUNCTIONAL INSPECTION
. .
.
.
. .. *.
. .
. .
.
COMPONENT COOLING
WATER.
NRC INSPECTION
50-311 /96-81
DECEMBER 2-13, 1996
OBJECTIVE
Determine if the system will perform its
intended safety function
SCOPE
Verify the system has a technically sound
d*esi"gri and* l"icensing basis
- * * *
Verify that system components are tested
to demonstrate design requirements
Verify that system operating practices are
consistent with the design
Review licensee's efforts to validate
licensing basis
1
OVERALL CONCLUSIONS
Significant CC system improvements
were made during the current outage
The licensing basis descriptions {FSAR)
were, with some exceptions, consistent
with actual plant conditions
- * * * ** ** Design issues were identi.fied by* the* team
that raise questions regarding CC system
capabilities
Contingent upon the satisfactory
resolution of the SSFI findings,
The SSFI team has concluded that the
Unit 2 CC system can perform its
intended safety function
2
- - - - - - - -
..
CC PUMP ROOM VENTILATION
Operating Practice Are Inconsistent with
Design
Historically, the single failure of CC room
ventilation was not properly addressed
Current EOPs do not support CC system
- opetation with a single *venti'lation *.
component failure
N.o CC ventilation design analysis to
support current EOPs
.
This issue should have been identified in
1995 when administrative requirements
were changed
Based on the SSFI finding, an AR was
issued to track the resolution of this issue
3
- ,
Ventilation Configuration Control is
Inadequate
No ventilation damper administrative
controls
Dampers found mis-positioned
Ventilation Testing... * * * * * **
The CC pump room ventilation testing
does not measure design parameters
necessary to demonstrate function
4
...........
CC PUMP OPERATION
Operating Practices Inconsistent With Design
EOPs allow a CC pump to operate beyond
its measured pump curve
The calculation for CC pump runout/NPSH
in:chJded some unsu*bsta**ntiated
=**
..... '*
- assumptions * * *
An adequate analysis was not completed
to support the runout of the CC pump
Based on the SSFI finding, an AR was
issued to track the resolution of this issue
5
.:.- .
.,
ENGINEERING DESIGN EVALUATIONS
Calculations
Calculations were generally easily
retrievable and available
.... . =.: ***=*--:-*e*: . .---*T*he,*brea:ker ._overcutretit r.-el-a*y,setting:for****** * * * ._ .. ,,~
4 *MoVs Were not *conservative* and*
required resetting
Multiple inconsistencies were noted in the
MOV thermal overload calculation
The setpoint calculation and actual CC
radiation monitor setting were not
appropriate
The setpoint calculation for the surge
tank level alarm was not complete
6
.,
ENGINEERING QUALITY VERIFICATION
Licensing Basis Validation
The PASS system is not operated in
accordance with the FSAR description
Several other minor descriptive FSAR
- * * * ** * * ***** *. * .. *d*iscrepanc1e*s.' *were***i*denti*f-ied** .* * * **
= =*~ *.**
- ... - * **
.. *. * * * ... ** .. : *
Minor discrepanci~s were noted between
the implementation of the FSAR project
CC Macro Review and the administrative
guidelines
The CC Macro review fulfilled its intended
function
Two changes to the TS were not
submitted in a timely manner
7
.,
Probabilistic Safety Assessment
Dependencies between the CC
pumps/charging pumps and ventilation
were not modelled correctly
.*'*- .... *.* ..... ** ... Th:~se .. errors. irDP.~.G~e..d .. th~ G~!~~_l9_ted _99re .. . ...
.
damage frequency. . * * .** *.
. *
- * *.* *
- * * ***
- * =
8
.,
.,
SURVEILLANCE AND TESTING
Test Program Scope
In general, CC system components were
included in the test program
One exception was the Spent Fuel Pool
- **. :, -. :and* Bori*c. ***Acid .-*Evaporato*r .manual .. ,.:., .... >
., .*:: * '; ..... :-*
isolation valves
Procedures
In general, the surveillance test
procedures reviewed were appropriate
One exception was the battery
performance test procedure acceptance
criteria were not consistent with the TS
9
- ""
.......
, .
Acceptance Criteria
The basis for the CC flow balance test
acceptance criteria was not documented
The basis for the CC heat exchanger
performance test acceptance criteria was
- * ** ., __ : ....... * ...... ** ... *not.-:documented:-. * :*_ **- ,. *-. * ..... , *.* *: .*- *- .* *.: :: -
._-*, .. ,: *. * *** * * * -* * *.: **; *.* *
Test Program Implementation
In. general, the test program
implementation was good
-
The 1993 battery surveillance test data
were not properly evaluated .
10
OPERATIONS
Procedures
Operating procedures were generally of
high quality and consistent with the CC
system design
~_ .. ,:, -* ..... = .*. ***:=.*Exception *n*oted we*-re*-the * EO*Psrr.elated to--**.:.**.-* .. -**
. ventilation failures or cc pu'tnp runoLit ..
. .
and the RCP trip criteria was inconsistent
between Abnormal Procedures
Training
The CC training material was of high
quality
The simulator properly reflected the CC
design with the exception of the CC
radiation monitor setpoints
11
"-.
l
'
....
Equipment Configuration Control
The CC radiation monitors have been out-
of-service for nearly 1 year
CC system drawings are generally
accurate
- ..:*:-:*.*:***:.'\\ ......... ,. .. : ....... .:.:=** ..
-
- - ........ ,,.*_ .. * ......
- _. ................. - .. **: .:* .. :*, :* .;
- .*
- . *.~*** ... * .... *****
'* ** .. ::.
Corrective actions *reviewed* for cc**
system component failures were.
appropriate
12