ML18094A553
| ML18094A553 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 06/16/1989 |
| From: | Eapen P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18094A551 | List: |
| References | |
| 50-272-89-17, GL-89-04, GL-89-4, NUDOCS 8907110242 | |
| Download: ML18094A553 (18) | |
See also: IR 05000272/1989017
Text
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No.
50-272/89-17
Docket No.
50-272
License No.
License~:
Public Service Electric and Gas Company
P.O. Box 236
Hancocks Bridge, New Jersey 08038
Facility Name:
Salem Unit No. 1
Inspection At:
Hancocks Bridge, NJ
Inspection Conducted:
May 22-26, 1989
Inspectors: .P. H. Bissett, Senior Operations Engineer
W. C. Lyon, Senior Reactor Engineer, NRR
S. M. Pindale, Resident Inspector
Approved by:
~date -
Inspection Summary:
Special announced inspection on May 22-26, 1989
(Inspection Report No. 50-272/89-17).
Areas Inspected:
Loss of RHR pump event on May 20, 1989 due to inadvertent
nitrogen injection from an accumulator and the licensee action associated with
it.
- Results:
The inspectors concluded that the licensee responded reasonably to
this event with a few exceptions.
These exceptions included operator error in
selecting wrong switch from the rear panel to unblock the accumulator
isolation valve; operator action to drain the RCS initially to bring the
pressurizer level on scale; inadequate test procedures for conducting
accumulator check valve full flow tests; inadequate abnormal operating .
procedures that did not assist the operators in realizing that the loss of RHR
event is reportable under 10 CFR 50.72 and the event classification guide that
also did not assist the operators in classifying this event adequately .
8907110242 890703
ADOCK 05000272
Q
PNU
I !**
2
A complete engineering analysis of the event based on system response was not
initially developed by the licensee.
Recent -operator training to mitigate the
consequence of a loss of decay heat removal during shut _down conditions permitted
the operators to vent the RHR system using gravity fill from the RWST.
Two violations (inadequate review of the accumulator check valve test
procedure and an inadequate event classification procedure that improperly
classified this event) were identified .
DETAILS
1.0 Persons Contacted
1.1
Public Service Electric and Gas Co~pany
-
Moses L. Burnstein, Principal Safety Review Engineer
Vi Jay Chandra, Technical Consultant, Engineering Sciences
Brian Connor, Technical Staff Engineer
Mahesh R. Danak, Salem Mech. Group
Allen Ho, Technical Consultant, Sciences
John Hudson, Offsite Safety Review Engineer
Stan LaBruna, Vice President Nuclear Operations
Craig Lambert, Nuclear Engineering Sciences - Manager
C. P. Lashkari, System Engineer
E. A. Liden, Manager-Offsite Safety Review
Bill McTique, Engineer
L. K. Miller, General Manager - Salem Operations
Steve Miltenberger, Vice President & Chief Nuclear Officer
John Musumeci, Operations Engineer
Pete Ott, NSSS Group Head - Technical Department
D. A. Perkins, Manager Station QA
Bruce Preston, Manager - Licensing & Regulation
Glenn A.~Roggio; Station Licensing.Engieer
John Ronafalny, Manager-Nuclear Engineering Services
Pell White, Technical Manager
1:2 United States Nuclear Regulatory Commission
Kathy Ha1vey Gibson, Senior Resident Inspector
I i *
I :
4
2.0 Salem Unit 1 Loss of Residual Heat Removal (RHR) System Pumps on
May 20, 1989
On May 20, 1989 Salem Unit 1 lost both of its RHR pumps for about fifty
minutes due to an inadvertent injection of nitrogen from accumulator
No. 13 to the RHR pump suction piping.
This inadvertent injection took
place while the operators were conducting full flow testing of accumulator
check valves after recent maintenance.
Each accumulator test consisted of
unblocking the appropriate accumulator isolation valve (SJ54) from the
back of panel 1RP4, stroking the valve from fully closed to fully open and
returning the valve to the fully closed position after the test.
Accumu-
lators 11 and 12 had been satisfactorily tested, and the pressurizer water
level indication had increased from 10% to 34% due to the injected water .
. The Unit was recently refueled and the reactor vessel (RV) head was rein-
stalled.
The reactor coolant system (RCS) had been refilled to an indicated
cold calibrated pressurizer level of 10%.
Reactor coolant pumps (RCPs)
had not been started and consequently the steam generator (SG) tubes and
RV head ccntained air.
Two pressurizer power operated.relief valves (PORVs)
were open, the pressurizer relief tank (PRT) was essentially e~pty, and
the -PRT rupture di ap1ragm was removed.
A 11 accumulators were fi 11 ed to
normal operating level and pressurized to approximately 600 psig.
The seq~ence of events for this occurrence is listed below:
2.1
Sequence of Events
Initial Conditions
~ode 5 (Cold Shutdown)
Time
9:25 a.m.
Event
RCS average temperature = 92 F
RCS pressure
= 14 psi
RCS filled to greater than center line
No. 12 RHR pump in service at 3000 gpm
Pressurizer level
= 34%
RCS/Pressurizer vented to containment atmosphere
No. 13 accumulator pressure = 630 psig
Volume of nitrogen
in No. 13 accumulator = 491 cft at 630 psig
RWST level
= 38.3 ft.
Start discharge test of Ko. 13 accumulator.
wrong isolation valve unblocked.
discharge commenced, but operators could not reclose
isolation valve.
proper isolation valve unblocked.
9:26 a.m.
9:34 a.m.
9:35 a.m.
9:43 a.m.
9:45 a.m.
5
No. 13 accumulator isolation valve closed
(total stroke time= 70 seconds).
RCS pressure increased from 14 to 51 psig.
Pressurizer level increased to greater than 100%.
No. 13 accumulator pressure dropped from 630 to 62
psig.
Control room supervisor ordered RH21 (RHR return to
RWST) valve opened to drain pressurizer level on scale.
Pressurizer level back on scale.
RHR flow to RCS at 1000 gpm due to RH21 open
(approximately 2000 gpm diverted to RWST).
No. 12 RHR pump motor amps steady at 44.
RCS pressure at 35 psi.
RHR flow to RCS rapidly decreased to zero (RH21 still open).
No. 12 RHR pump motor amps at 21-24 (operator reports
pump sounds abnormal, however unlike cavitation).
RH21 ordered ~losed.
RHR flow paths, except recircula-
tion flow path, isolated.
(Control room supervisor
believed ~o. 12 RHR pump may have failed mechanically).
Pressurizer level continued to decrease (85%)
Operators placed No. 11 RHR pump in service.
No. 11 pump motor amps identical to No. 12 (21-24 amps).
No indicated RHR flow.
Core exit thermocouples slowly increasing (about 1
degree/hour).
AOP-RHl and AOP-RH2 initiated.
Both RHR pumps removed from service.
Operators realized that nitrogen had discharged from.
No. 13 accumulator and RHR system was air bound.
Control room. supervisor ordered the RHR system be
vented .
9:45 -
9: 58 a. m.
9:59 a.m.
10:05 a.m.
10:18 a.m.
10:23 a.m.
10:28 a. m.
10:37 a.m.
10:56 a.m.
2:50 p.m.
3:15 p.m.
6
Venting both RHR trains.
RH6 (pump casing vent) ~nd RH9 (discharge p1p1ng drain)
valves open on each train.
Could not verify system
venting since those valves were hard piped into the
auxiliary building sump (could not see into sump).
Operators could not immediately identify the location
of RH13 valves (discharge piping sample line) to enhance
venting.
RHR system vents ordered closed.
Started No. 12 RHR pump.
No flow, low amps, pump
removed from service.
RH13 valves located and opened, venting locally verified,
however it was very slow (low pressure).
Pressurizer level reached 23%.
Reactor head vents open, however differential pressure was
not sufficient to provide vent flow:
Core exit thermocouples at 122° F.
Control room supervisor ordered SJ69 (RWST to RHR pump
suction) valve open to flush air out of RHR system
since venting was slow.
Pressurizer level increased from 23 to 56%.
Air, then water, issued from RHR vents and drains.
SJ69 closed, pressurizer level stable at 56%.
No. 11 RHR pump placed in service.
No. 11 RHR pump flow restored to 3000 gpm.
Pressurizer level steady at 56%.
Core exit thermocouples at 93° F.
No. 12 RHR pump placed in service.
No. 11 pump removed
from service (RHR flow at 3000 gpm).
Reactor head vents closed, AOPs exited.
Surveillance test satisfactorily performed on No. 12 pump.
Surveillance test satisfactorily performed on No. 11 pump.
Final Conditions:
7
RCS average temperature
RCS pressure
Pressurizer level
RWST level
2.2 Evaluation of RCS/RHR System Responses
-= =
=
93° F
31 psi
57%
39.8 ft.
Preliminary analyses by the licensee indicated that about 1300 cft of
nitrogen was injected into the reactor coolant system while the isolation
valve for 13 accumulator remained open for about seventy seconds.
(Note:
491 cft of Nitrogen at 630 psig expands into 2690 cft at 62 psig.
The
accumulator, 1350 cft, and the piping system up to the reactor coolant
system, 35 cft, has~ combined volume of 1385 cft.) As nitr6gen entered
the RCS it initially expanded to about 2307 cft at the maximum observed
reactor pressure of 51 psig.
At atmospheric pressure this injected
nitrogen would have a volume of about 10,200 cft).
Figures 1,2 and 3
depict the estimated RCS fluid conditions at critical stages of this
event.
Figures 4 and 5 present the behavior of key reactor parameters
during this transient.
From these observations, the licensee postulated
the following behavior for the reactor coolant system during nitrogen
injection:
At 0925 hours0.0107 days <br />0.257 hours <br />0.00153 weeks <br />3.519625e-4 months <br /> the initial rapid water injection into the cold leg compressed
air trapped in the steam generators (SGs) and in the upper reactor vessel
(RV) head.
Water was forced into the pressurizer.
The high pressure nitrogen essentially emptied the cold leg at the injection
point, followed by depression of the RV downcomer level and draining of
the other cold legs into the depressed downcomer.
Nitrogen passed through
leak passages between the top of the RV downcomer and entered the RV head
region.
Displaced water continued to fill the pressurizer.
Water in the
cross-over pipes and in the cold leg side of the SG tubes probably prevented
large quantities of nitrogen from escaping directly into the hot legs by
way of the SG tubes.
Some nitrogen may have vented into the RV head via
the RV lower plenum. and the core.
At 0926, the open Power Operated Relief Valves (PORVs) allowed the pressurizer
.to continue to fill.
As water continued to flow into the pressurizer, the
Reactor Coolant system (RCS) depressurized and nitrogen expanded into the
SG tubes, the cold and hot legs, and the RV head, forcing more water out
the surge line into the pressurizer.
As nitrogen expanded into the upper
hot leg region, nitrogen separated the water in the pressurizer from the
RCS.
Pressurizer level indication reached 100% and went off-scale high.
Some water probably was carried over into the PRT.
RCS draining operations were initiated at 0929 and contributed to hot leg
inventory reduction as two thirds of the water entering the RHR suction
pipe was pumped into the RWST rather than being returned to the RCS cold
legs.
At 0935, continued reduction of water inventory in the hot legs led
to the loss of RHR flow due to gas binding of the 12 RHR pump.
However,
the operators did not realize the RHR pumps were gas bound.
At this time,
pressurizer level indication was 85% and indicated RCS pressure was about
33 psig.
8
Between 0942 and 0945, RHR pump noise led the operators to-incorrectly
believe the operating 12RHR pump was damaged, and they _started llRHR.
Since the RHR suction ljne was voided, llRHR immediately became gas bound.
The operators then realized the pumps were voided.
They tripped both RHR
pumps, terminated RCS draining, and initiated pump vehting operations.
About 25,000 gallons o.f water had been pumped from the RCS into the RWST.
Indicated pressurizer level was still above 60% at 09:45.
The operators followed RV temperature closely after loss of RHR.
They
correctly concluded the.slow temperature increase rate would allow adequate
time for RHR restoration efforts. (Water injection via charging or safety
injection pumps were available options, but the operators were con~erned -
about the voided suction line and the possibility of ingesting nitrogen
into these pumps as well.)
-
The early attempt~ to vent the RHR system were unsuccessful partially due
to under sized vent lines and the piping system that trapped water to
prevent venting of certain portions of the system.
Between 0945 and 0956,
pressurizer lev~l continued to decrease and the RCS pressure remained
constant at about 33 psig.
Between 0956 and 0959, an attempt to sta~t the 12RHR pump was unsuccessful.
However, pressurizer level dropped from about 38% to 23% and pressure
dropped from 33 psig to 18 psig during the start attempt.
Level and pressure
remained constant until RCS refill was initiated.
Between 1018 and 1023, operator initiation of gravity 'fill of the RHR system
from the RWST increased suction line pressure and RCS inventory.
Venting
was accomplished successfully and at 1023 RHR was restored.
Level and -
pressure were stable and constant at 56% and 18 psig, respectively.
2.3 Cause of Event
lhe root cause of the event was operator error due to inattention to
detail in that he unblocked the isolation valve for accumulator 14 instead
-of the isolation.valve for accumulator 13 from the rear panel 1PR4. *
Several additional factors were identified which contributed to this
operator error.
The surveillance procedure used (SP(0)4.0.5-V-SJ-6) did not require
specific sign-offs to verify that the correct accumulator isolation valve _
was unblocked from the rear panel.
IMterviews with several plant operators
indicated that the procedure steps requiring sign-offs received more
attention than those that did not.
The adequacy of this procedure is
discussed in more detail in Section 2.5
Another important contributor to this event was the human engineering
deficiency associated with the control room lay out. Specifically, the
unblocking switches were arranged in the control room panel in 11, 12, 14,
13 order.
When the third accumulator (13) _was being set up for discharge,
9
the operator proceeded to the switch in the third location (14) and actuated
i~. The licensee had already recognized the* layout of those switches to
be a human engineering deficiency (HED), and as such, plans to correct
- that HED during a subsequent phase of the control room modifications.
A third contributing factor was the statio~ policy whfch permits only
reactor operators (RO) to manipulate the controls.
In this case, however,
a senior reactor operator (SRO) performed that evolution and actuated the
wrong switch.
Although the SROs are licensed to perform this function,
their primary responsibilities are to direct the ac_tivities of the ROs,
who are more familiar with the details of the control boards.
Prior to this test, the licensee was performing full flow testing of the
high head safety injection system.
This safety injection system flow
testing was delayed for several hours.
Since plant conditions satisfied
the test requirements for SP(0)4.0.5-V-SJ-6, and the test was scheduled to
be performed prior to entering Mode 4, control room supervision elected to
perform the accum~lator discharge test during this delay.
Although ample
time appeared to be available to complete the test, interviews with licensee
personnel indicated that the test may hav~ been rushed.
(There appeared
to be no undu~ pressure placed on the operations staff to complete the
test quickly.) This may also have ~ont~ibuted to the operator error~
In summary, although the cause of the event was operator error, several
additional factors contributed to the error.
2.4 Reportability of Concerns
The loss of the RHR system occurred at approximately 9:45 a.m. on May 20.
10 CFR 50.72, Section (6)(2)(iii)(B), requires the licensee to notify the
NRC within 1our hours of any event or condition that alone could. have
prevented the fulfillment of the safety function of structures or systems
that are needed to remove residu~l .heat.
This event was not reported as
required by that section until about 11: 00 a. m. on May 22.
The inspector reviewed the details associated with the licensee's implementa-
tion of reportability requirements.
The inspector determined that following *
the event, operations personnel referenced the Event- Classification Guide *
(ECG) to determine event reportability.
Tab 17 (Safeguards) was referenced
first, which specifically stated that if the RHR system fails to attain/
maintain RCS at 200° F, then declare* an Alert.
Since 200° F was not reached
(122 F maximum), that declaration was not applicable.
The other section
that could have potentially applied was Tab 17 (Technical Specifications).
Item J, "Event or condition that alone could have prevented the function
of safety structures of systems,
11 was reviewed; however, it was not as
specific as 10 CFR 50.72.
Although item J contained implementing examples,
none appeared to specifically fit the event.
Therefore, the licensee did
not report the_ total loss of RHR event.
I
I
I i l
I
10
Followup investigation identified that although the event was not officially
reported, the licensee recognized the importance of the event and did notify
several organizations, including NRC Region I, later on May 20.
It was
not until Monday, May 22, that the licensee determined that the event was
a 50.72 reportable event.
The licensee stated that the ECG was designed
to contain all of the necessary information to determine reportability and
the operators are not expected to review 10 CFR 50.72 foi reportability of
an event.
The inspector noted that the licensee was in the process of implementing a
major change to the ECG procedure.
The revision which was reviewed for
approval by the Station Operations Review Committee on May 3, 1989, included
specific guidance in several appropriate locations that clearly identified
the proper reporting requirement for the May 20 event.
However, the revision
was not effective prior to the event.
The new ECG procedDre was subsequently
issued and became effective on May 26, 1989.
The ECG procedure was inadequate in that it did not provide the necessary
guidance to properly classify the event and report it to the NRC.
This is
an. apparent violation of NRC requirements (50-272/89-17-01).
2.5 Test Procedure Adequacy
The insp~ctor reviewed the procedure and testing requirements associated
with this event.
Surveillance procedure SP(0)4.0.5-V-SJ-6,
11 Inservice
Testing - Valves - Safety Injection
11 , prescribes the testing process for
the accumulator discharge check valves (two series check valves for each
of the four accumulators).
11 Rules for Inservice Inspection of Nuclear Power Plant
Components", Subsection IWV-3520 (Check Valve Tests) and the licensee's
approved Inservice Testing (IST) Program require that the above check
valves be full-stroke tested during each cold shutdown.
Section XI
specifies that the test can be performed with or without flow through the
valve.
Additionally, IWV-3200 (Valve Replacement, Repair and Maintenance)
requires that valves be tested following specified activities to ensure
that performance parameters are within acceptable limits.
The above referenced SP(O) was performed on March 31, 1989, to satisfy the
licensee's !ST Program requirements.
During the outage, all eight check
valves were modified, and as such were required to be tested following the
modification.
The licensee performed the SP(O) on May 20 to satisfy the
post-modification/ maintenance requirement.
The inspector reviewed the surveillance procedure and found that the licensee
performs that test at normal system operating parameters for the accumulators
(595.5 to 647 psig pressure and ~l to 65% level).
Psrforming the discharge
test at full accumulator pressure and a significantly low RCS pressure
resulted in the nitrogen injection and the associated problems of the May 20
event .
11
The inspector also identified that there were no sign offs for the action
steps prescribed by the procedure, which could potentially lead to the
omission of steps or performing evolutions out of sequence.
The inspector
interviewed several operators and found that the operators felt that steps
which require their sign off received more attention than those that did
not.
This may have contributed to the operator error in performing the
test.
The licensee stated that they recognized that certain procedures
. did not contain the necessary sign off steps and these procedures were
slowly upgraded as a part of an integrated procedure upgrade program.
The
inspector found, however, that the two-year review was performed for the
above referenced SP(O) on December 2, 1988, and no changes were made.
Technical Specification (TS) 6.8.1 requires that written surveillance and
test activity procedures shall be established, implemented and maintained.
TS 6.8.2 requires that those procedures shall be independently reviewed by
a station qualified reviewer (SQR) and approved by the appropriate station
department manager prior to implementation.
Administrative Procedure No.
32,
11 Implementing Procedures Program,
11 specifies that the SQR will prepare
a safety evaluation for those procedures which involve a significant safety
issue (SS!) and. the implementing department manager will document the
determination as to whether the procedure contains an SS!.
If a procedure
contains an SS!, then a 10 CFR 50.59 safety evaluation must be completed
and Station Operations Review Committee (SORC) review must be performed
per TS 6.5.1.6 requirements,
Inspector review of SP(0)4.0.5-V-SJ-6 identified that no SORC review had
been performed.
Additionally, the yes/no classification for the SS!
determination was not checked.
The licensee* stated that the SS! classifi- *
cation was no, and therefore no SORC review was required.
The licensee
failed to properly evaluate the safety implications of performing this
procedure under specific plant conditions.
The failure to perform the
appropriate safety review for the surveillance test procedure is a violation
of Technical Specification requirements (50-272/89-17-02).
2.6 Abnormal Operating Procedure Adequacy
Following identification of RHR pump problems and the subsequent stoppage
of both pumps, the licensee initiated Abnormal Operating Procedure AOP-RHR-1,
11 Loss of RHR Cooling.
11
Entry conditions that precipitated usage of this
procedure included an abnormal change in RHR pumps motor current and the
subsequent loss of both RHR pumps.
AOP-RHR-1 usage for the given plant conditions would eventually lead to
.the establishment of RCS feed and bleed operations.
Since incore thermo-
couple temperatures were rising at a slow rate (- 1/2° F per minute) other
actions were being taken simultaneously with setting.up for feed and bleed
operations.
This included continuing efforts to vent the RHR pumps in an
attempt to reestablish RHR flow.
' *-
1
12
In conjunction with the performance of AOP-RHR-1, personnel were also
reviewing AOP-RHR-2, Loss of RHR Cooling - RCS Level Below Pressurizer -
EL104.
11
AOP-RHR-2 applies to the loss of RHR cooling during mid-loop
operations.
Since the RCS level was above the 104 feet level (greater
than mid-loop)~ entry into this procedure was inappropriate.
However, as
a result of continuing training received for mid-loop operation, shift
personnel knew that level in the RCS could be maintained by gravity drain
from the refueling water storage tank (RWST) and therefore could be used
as a means for possibly filling and venting the RHR system.
Operation of
valve SJ-69 for gravity draining of the RWST is designated in step 3.8.6
of AOP-RHR-2.
In completing AOP-RHR-1, step 3.13.5, which essentially completed the
actions necessary for feed and bleed, with the exception of starting a
charging pump, the shift supervisor opened SJ-69.
Pump venting efforts
were successful at this time and a restart 6f the RHR pumps subsequently
took place, whereupon RCS conditions were returned to normal.
In summary, it was apparent that the established procedures were not adequate
for the plant conditions during this event.
The licensee identified these
deficiencies, and the applicable procedures were being revised.
However,
operator actions taken during the recovery of this event were considered
appropriate for the conditions that existed.
2.7 Adequacy of Training
A review was performed to determine the extent of training received by
licensed operators both prior to and after the event.
The extent of the
training department's involvement during and after this event was also
assessed.
As a result of this review, it was determined that previous RCS mid-loop
operation training, conducted during the week of Aprtl 3, 1989, helped the
operators during the attempted recovery of the RHR pumps.
It eventually
provided a course of action for them to take instead of feed and bleed of
RCS as designated in AOR-RHR-1.
Following the event, the operation's department provided informal training
to all licensed operators not involved in the event prior to their assuming
any watchstanding positions.
These briefings included a review of the
fol lowing:
Sequence of events
Inappropriate operator actions
Inadequacies of AOR-RHR-1 for given plant conditions
Injection of Nitrogen into RCS
It was not apparent that the training department provided any oversight or
assistance to the operations department during the conduct of these briefings.
It was determined that the training department did not become aware of the
event until three days later.
Even after becoming aware, training involve-
u
13
ment was minimal.
Not until the fifth day following the event, did the
training department meet with the operations department to review the
sequence of events and formulate desired courses of action in regards to
training.
These courses of action, both short and long term include the
following:
Develop or revise licensed and non licensed operator training materials
to support corrective actions identified by the ongoing investigation
of this event.
Incorporate training on the event in continuing ltcensed and non
licensed training programs.
Review the event and follow-up investigations with entire shift
compliment during segment 1 of the 1989-90 continuing training program.
Develop and conduct training of various pump failures/cavitation.
Upgrade the capability of the Salem simulator to provide
11 hands-on
training of abnormal RHR events.
Review the design basis and specifications for the, ECCS accumulators
and their applications.
In summary, it was evident that there was a definite lack of initiative to
get the training department involved in the training related to this event.
Even though the operations department representatives stated that they
felt no need for immediate assistance from the training department, they
did not inform the training department of the events shortly after it had
occurred.
The training department appeared not to be kept current of all
operational events regardless of the outcome or severity.
The training department also was not aggressive in offering their assistance
and becoming involved once they were informed of this op~rational event.
As such, training personnel were not involved in the investigation that _
followed the event, and the training department was not kept abreast of
the conclusions of the investigation to develop meaningful short term and
long term training programs to preclude the reoccurrence of this event.
Licensee management involvement appeared to be lacking in their use of
resources available in the training department to effectively investigate
plant occurrences and to develop meaningful short and long term training
programs.
2.8 Licensee Corrective Actions
The
license~ initiated the following short term corrective actions:
1.
Training of oncoming licensed shift personnel was conducted as detailed
in section 2.6 of this report.
14
2.
Use of th~ accumulator discharge test procedure, SP(0)4.0.5.V-SJ-6
was discontinued pending reevaluation of testing requirements.
3.
Abnormal Operating Procedure AOP-RHR-1, Loss of RHR cooling was revised
to incorporate the conditions present in* the RCS at the on set of the
loss of RHR event, with reactor vessel head installed, RCS vented,
and level in pressurizer.
4.
AOP-RHR-1 was revised to incorporate use of additional vent paths,
including reactor head vents as recommended by Systems Engineering.
5.
The Training Center- was developing training plans to incorporate the
lessons learned from this event.
6.
The Unit 1 Emergency Core Cooling System Surveillance test procedure
(SP (0) 4.5.2h) was revised to incorporate the enhanced guidance for
removing excess RCS inventory.
Additionally, Unit 1 and 2 Cold Shutdown
to Hot Standby Integrated Operating Procedures (IOP-2), Draining Reactor
Refueling Cavity were identified as requiring the same procedural
enhancement.
Revision to these procedures will be completed prior to
June 15, 1989.
7.
Flow testing of the rema1n1ng accumulator check valv~s was ordered
stopped by the General Manager - Salem Operations, pending initial
review of the event, SORC approval and GM - Salem Operations approval
to resume flow testing.
8.
Evaluation of the event from a systems engineering/analysis perspective,
was conducted.
9.
Notified the industry through the Nuclear Network.
10.
Submitted Operating Experience Report to INPO for Industry Experience
Information.
11.
Event Classification Guide (ECG) was revised and issued on May 26, 1989.
Additional review of the new revision will be conducted by licensing
personnel.
The licensee's planned long term corrective actions are:
1.
Accumulator block out switch reconfiguration to incorporate previously
identified Control Room Human Factors deficiencies will be completed
in accqrdanc~ with the Control Room redesign schedule (i.e., next
refueling outage).
2.
Develop specific training for this event to:
a.
Discuss Sequence of .Events
- -
15
b.
Discuss root cause of identified inappropriate actions taken
during event.
c.
Review the design basis and specifications for the ECCS accumulators
and their application.
d.
Review revised procedures:
1.
2.
3.
IOP-2
4.
SP(0)4.0.5.V-SJ-6
5.
Other applicable procedure changes
This will be completed by September 1989 and be incorporated into the
continuing training process.
3.
Review the procedure weaknesses of SP(0)4.0.5.V-SJ-6 and AOP-RHR-1
with Station Procedure Writers by July 1, 1989.
4.
Conduct an ind2pendent investigation of the event and submit the results
to the Vice President and Chief Nuclear Officer by June 16, 1989.
5.
Develop or revise the licensed and non licensed operator training
materials to support the corrective a~tion~ identified by the ongoing
and independent investigation of the event.
6.
Develop and conduct improved training with regard to various pump
failure/cavitation conditions.
7.
Review the need to upgrade the capability of the Salem simulator tti
provide
11 hands-on training
11 of abnormal RHR events.
8.
Evaluate the need for improved venting capabilities on the RHR piping.
9.
Evaluate the need to provide physical in plant training and/or detailed
description in procedures of selected valves and other components
important for maintaining core inventory.
10.
Evaluate IST testing to determine the best test method for the accumu-
lator check valves.
The guidance of Generic Letter 89-04 will be
used in this effort.
11.
Evaluate the use of SPDS for shutdown operations and identify potential
areas of enhanc-ement.
16
3.0 Conclusion
The NRC inspectors concluded that the licensee's engineering evaluation
adequately addressed the system responses d~ring this event.
However, the
licensee did not assess the event from a system response perspective until
NRC personnel got involved in the event follow-up.
Except for the initial
operator error and a questionable decision to drain RCS upon high pressurizer
level, the operator actions demonstrated excellent skills and knowledge of
the system.
Specifically, the operators used the knowledge gained from
mid-loop operation training while venting the RHS system during this event.
Procedure deficiencies played a significant role during this event.
The
surveillance test procedure for testing of the check valve did not appear
to alert the operators about nitrogen intrusion.
The abnormal operating
procedures did not guide the operators to declare an alert when both RHR
pumps were lost.
The event classification guidelines were not adequate to
inform the operator that a 10 CFR 50.72 reportable condition existed when
the facility lost both *RHR pumps.
The above procedural deficiencies constitute a violation of Technical
Specification 6.8.1.
The inspectors found the licensee:s short and long
term corrective actions effective in addressing the circumstances that led
to this event.
The use of resources available in the licensee's training
department was not apparent during licensee's investigation of this event.
With the exception of the findings discussed in this report, the inspectors
found the licensee's response to this event, engineering analysis and short
and long term correctiv~ actions to be effective in mitigating the
consequences of this event and precluding this event from occurring in
the future.
4.0 -Exit Meeting
The findings of this inspection were discussed with the licensee represen-
tatives (denoted in Paragraph 1) at the exit interview on May 26, 1989.
No written material was provided to the licensee by the inspectors.
The
licen*see did not indicate that proprietary information was involved within
the scope of this inspection.
' ; '
r
'
.
Sl::.AM
00.JERATOR
/
/
/
I
LOOP 11
FIGURE 1
PRESSURIZER
VESSEL ,
YI-%
TIME
9:25 am
R't\\ST
FIGURE 2.
TIME
9:27 am
sl::'.:.u
C:::!-.f.Foi:;..iOR
RV/ST
ACOJUULA TOR
P~RIZER
ST'"JJI.
CBI O:.A i\\Y.
LOOP 13
S'TEAM
GENERATOR
LOOP 11
LOOP 13
FIGURE 3
TIME
9:35 am
RCS VOLUME:
11,905 CFT
AIR IN RCS:
4,086 CFT
NITROGEN IN RCS:
0 CFT
NITROGEN IN RCS:
2,307 CFT
RHR PUMP IS GAS BOUND
,.....
IL.
Cl
Cl
Cl ...
Cl ,,
'-'
w
0:::
- J
~
w
a..
- i: w
I-
,..... ., .
.....
'-'
Iii
3::
- 0:::
,...,
"
._,
0:::
N
a..
,...,
ii
Q.
'-'
ln
(.)
0:::
122
120
118 -
116
114
112
110
108
106
104
102
100
98
96
94
92
90
I f\\.
v
88
09:25
(
I
,...v
/\\ I
09:40
INCORE TEMP T21
(
r'
r
/
.
CORE LOC E-11
I
n.r-1
.v
,... r
-
. .
09:55
TIME
FIGURE 4
..
~
r
-
10:10
/'1
\\ v \\
\\
\\
'--\\
10:25
RWST .& PZR LEVEL
and
RCS PRESSURE
110
-
.
100
90
80
. 70
60
50
40
30
20
'\\.
-
I
~ \\
\\.
\\
- )
" ~
""'
n r
'"""
""'
I
I
\\__
'I
I
{>-<
~-
1
J
)
\\
-
L
?
10
0
..,
J
09:13
09:28
09:43
09:58
10:13
10:28
TIME
RWST LVL
~PRESS
o
PZR LVL
FIGURE 5
r
-
. .