ML20149E337
| ML20149E337 | |
| Person / Time | |
|---|---|
| Site: | Cooper  |
| Issue date: | 05/16/1994 |
| From: | Collins E, Gagliardo J, Harrell P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20149E325 | List: |
| References | |
| 50-298-94-10, NUDOCS 9405310023 | |
| Download: ML20149E337 (10) | |
See also: IR 05000298/1994010
Text
.
.
-
.
- , . .- - . .
.
~ .
. - ~ . ~
..
-.
.
.
.
-
.
, . -
APPENDIX
U.S. NUCLEAR REGULATORY COMMISSION
3
REGION IV
-Inspection Report:
50-298/94-10
License:..DPR-46
Licensee: Nebraska Public Power District
P.O. Box 499
y
Columbus, Nebraska
,
Facility Name:
Cooper Nuclear Station
e
. Inspection At:
Brownville, Nebraska
i
Inspection Conducted:
March 3-7, 15, and 29, 1994
'
Inspectors:
J. E. Gagliardo, Former Chief, Project Branch C
'
E. E. Collins, Project. Engineer
J. L. Pellet, Chief, Operations Branch
,
/s
'
Approved:
k \\${h
blh44
1
P. H. Harre
- hyroject. Branch C
.Date
Insnection Summary
,
'
- Areas Inspected:
Special, unarmounced team inspection of the licensee's
response to the events resulting in a. reactor scram'on the evening of March.2,
1994, and results of the licensee's investigation efforts 'and resulting
i
corrective actions taken by the licensee to resolve the identified problems-to-
j
prevent. recurrence.
Results:
l
The' event investigation process was initially-informal and unstructured.
)
.
1
The licensee's' efforts lacked the rigor and systematic approach that
would ensure success in identifying the'cause(s)'of'reacter trips and'
,
prevent their recurrence (Section 3.2. 1).
The posttrip review. completed by the' licensee was not effective in that-
+
it did not. address plant. response, equipment' performance, and the.cause
,
of.the trip. This was caused,.in part, because the objectives for the
!
.
review were poorly defined and procedural- requirements were- not well-
understood (Section 3.2.3).
,
s
2'-
1
.
l
J
'I
~
9405310023 940525
PDR- ADOCK 05000298i
..G
ppg
'
,
.yw,c
n-
.,
,.r~.
m
,
.
-<
.E
F!
~
.
>-
-2-
p
.
The information initially provided to the NRC was misleading and
contributed to the NRC's concern, which resulted in a special inspection
team being sent to the site.
Periodic updates to the.NRC during the
unusual event were not provided (Section 3.2.5).
Management oversight and self-assessment efforts related to this event
were poor (Section 3.2.6).
The inspection team noted that it appeared that the licensee was focused
on early startup of the plant rather than on thoroughly or
systematically evaluating the transient and determining the root cause
of the scram (Section 3.2.6).
Summary of Inspection Findings:
None
Attachment:
Persons Contacted and Exit Meeting
l
1
i
i
.
!
j
.j
g;_
~.
-
e
'
.
-3-
r
DETAILS
-
1 PURPOSE OF SPECIAL INSPECTION
The purposes of this inspection were to review and evaluate:
(1) the
licensee's response to the events associated with the reactor scram on
March 2, 1994, (2) the licensee's investigative efforts to determine the
causes of the events, and (3) the licensee's actions to correct- the identified
problems and prevent recurrence.
,
.
2 EVENT
On March 2, 1994, at 6:09 p.m., the licensee notified the NRC Headquarters
Operations Officer that the plant had experienced a reactor scram at-
5:47 p.m., with the plant at full power. The instrumentation in the control
room indicated that the scram was caused by high reactor flux.
The cause of
the indicated power increase was not known at the time initial notification
.
was made to the NRC. The initial notification stated that reactor vessel
water level dropped to -37 inches, which automatically actuated the high
!
pressure coolant injection (HPCI) and reactor core isolation cooling (RCIC)
!
systems. The initial notification also stated that these systems had injected
approximately 16,000 gallons of water into the reactor vessel and that the
posttrip shrink of water level was more than was expected for this type of-
The licensee indicated that a Notice of Unusual Event had been declared per
',
Emergency Action Level 3.1.2 because of the HPCI and RCIC system initiations.
The plant was stable in Mode 3 at 5:58 p.m., with decay heat being removed
through the condenser bypass valves.
3 EVENT REVIEW
3.1 Licensee Response
i
'
The inspection team interviewed the Plant Manager, Plant Engineering Manager,
and Problem Resolution Team (PRT) leader on March 3 and 4 to determine- the
,
'
licensee's initial response to the event. The Plant Manager was at the site
when the scram occurred and immediately responded to the control room to
>
determine what had happened and monitor the crew's actions. He established
,
communications with the Vice President-Nuclear at the corporate offices and
notified the Senior Resident inspector.
The Plant Engineering Manager was at the site when the reactor scrammed. He
called the system engineer, who became the leader of the licensee's event
review effort, which was later formally designated as a PRT. Together, these
individuals plotted a strategy for determining the cause of-the event.
They.
walked down the turbine and the digital electrohydralic control system (DEHCS)
-
system and found no anomalies that might have caused the scram.
Because the
,
r
p
.
.
-4-
operators were surprised by the magnitude of the water level swing during the
event, they also began looking at the feedwater system as a possible cause for
the event.
The Plant Engineering Manager and the system engineer identified potential
modes of DEHCS failure for evaluation. These included possibilities such.as:
(1) crud (foreign material) in the DEHCS control valves, (2) hydraulic leak in
the DEHCS, (3) electronic control failure (loss of control signal),
(4) throttle pressure transmitter failure, (5) servo valve failure or
malfunction, or (6) pressure controller malfunction. Their initial
evaluations eliminated all items, except the DEHCS.
The licensee stated that
i
an expert on the DEHCS would be brought to the site to assist them in
determining the problem.
The licensee performed a posttrip review, as required by Administrative
Procedure 2.0.6.
The posttrip review was completed on the afternoon of
March 3 and was reviewed and approved by the Station Operations Review
Committee (SORC) on March 3, prior to the inspection team's arrival on site.
The SORC had reviewed the computer print outs and data sheets for the scram
and had approved the review, with one open item. The open item was assigned
to the Plant Engineering Manager and required him to review the reactor water
level response to the transient to ensure that it was in conformance with the-
analyzed transients in the design basis of the plant.
3.2 NRC Review of the Event
3.2.1
PRT Response
The inspection team found that the licensee's initial review effort was not
formalized.
Later on March 3, after the inspection team's arrival on site,
the licensee's review efforts were formally designated as a PRT.
The inspection team asked the licensee if a sequence of events had been
developed for the event.
Licensee representatives stated that the computer
had generated a set of printouts of the event sequence and that they had
directed that a simple sequence of events be developed.
At the time of the
question, the licensee had not developed or planned to develop a' detailed
sequence of events.
The inspection team interviewed the members of the PRT responsible for the
DEHCS problem on the afternoon of March 4 and found that they did not have
detailed guidance on how they were to systematically approach the testing of
the DEHCS to ensure that the failure mechanism (s) were clearly understood and.
no other likely cause was overlooked. The system-engineer, tasked with
determining the problem with the.DEHCS, had been given very little specific
guidance and the engineers assigned to assist him were not' dedicated to this
effort. The inspection team discussed this concern with licensee management
and a comprehensive action plan was subsequently developed and implemented.
The action plan included a logic diagram for DEHCS testing and involved'a
systematic approach to the overall testing scheme.
--
=
.
.
-.
p
.
.
-5-
3.2.2 Control ~ Room Operator Response
On March 3 the inspection team interviewed the operations crew that was on
shift at the time of the event.
The crew had been preparing to do a test run
of a diesel generator when the event occurred. The first indications they had
of the event were the reactor scram annunciator actuations and the red
rod-inserted lights on the full core display. The operators noted that the
HPCI and RCIC systems actuated and originally presumed that there was a
problem with the feedwater pumps, because the HPCI and RCIC actuations, for
this type of transient, were not expected.
The operators verified tnat the
feedwater pumps were providing sufficient makeup water and immediately took
control of the HPCI and RCIC system pumps and secured them. The crew stated
that there were no previous indications of an impending problem. The crew did
express concern that there had been a red-arrow log entry regarding an
abnormally high difference on the output of the two pressure controllers in
the DEHCS. The crew stated that they had seen differences as.high as 50 psid,
when the actual difference should be only about~ 3 psid. The. inspection team
concluded that the operator response to the reactor scram was appropriate.
3.2.3
Posttrip Review
The ins- " ion team (stablished that Procedure 2.0.6 required that the
posttr4
tvh w include:
(1) a determination that the plant was in a safe
condith
., an evaluation of the cause for the trip, (3) a determination
that corrective actions were identified and appropriately implemented,' and
(4) a determination that proper operation of plant safety-related system:, had -
.
been observed.
The inspection team noted that the data sheets that were
reviewed by the SORC did not address these items.
The licensee indicated that these posttrip review items would be addressed
,
before the startup authorization procedure (Procedure 2.1.1.1) was
implemented, as part of the open item that was assigned by the S0RC. The
inspection team noted that Procedure 2.1.1.1 did not require the determination
of the cause of the trip or the other posttrip review items.
The inspection
team further noted that the open item wording appeared to be limited to
addressing feedwater system performance.
The inspection team concluded that the post-trip review that was completed and
approved by the 50RC on March 3, 1994, was not effective in that it did not
verify plant response, identify the cause of the scram, or verify
implementation of corrective actions.
Licensee representatives stated that
this issue would be resolved and more effectively communicated to those having
responsibilities in this area.
3.2.4
Plant Response
The inspection team generated a sequence of events from the computer printouts
provided by the licensee and found that the plant had responded as expected,
with the exception of the HPCI and RCIC system initiations. The inspection
team identified that the computer data showed that the HPCI and RCIC system
& w
g
-
.
.
-6-
.
initiations occurred at a reactor water level of -17 inches, while the
calibration data for the trip switches indicated that the switches were set to
actuate at a nominal level of -25 inches.
The licensee evaluated the actual initiation points of the HPCI and RCIC
systems after the issue was developed by the inspection team, and concluded
that the initiation points were correct.
The apparent start of the system
when reactor level indicated -17 inches was caused by the differences in the
mechanical response of the level indicating switches and the level
transmitters.
The inspection team noted that the curves, documented in the Updated Safety
~
Analysis Report for this type of transient, did not indicate a vessel level
transient as severe as the one experienced after the scram. The inspection
team interviewed the PRT members that were tasked with resolving the reactor
water level issues.
This inspection team included a General Electric (GE)
representative to assist the PRT in analyzing the data.
The GE representative
developed a set of curves from the computer data for the March 2 scram and
showed that th curves from this transient closely paralleled those of a
closure of one min steam isolation valve with the reactor at full power.
The inspection team observed that the only parameter that varied significantly
q
from those of the model (one main steam isolation valve closure) was that for
reactor water level. The magnitude of the reactor water level drop for the
March 2 scram was significantly more than that in the model. The GE
representative stated that he'did not understand why the scram caused such a
level transient and acknowledged that the design basis for the plant stated
that HPCI and RCIC actuations should not occur for a transient of the type
experienced on March 2.
Licensee representatives stated that they had seen
.
HPCI and RCIC actuations on trips such as this in the past and were. aware that
J
it was not desirable to inject cold water into the hot. vessel for this type of
i
transient. They noted that the level setpoint for HPCI and RCIC actuations
j
had been changed (made less negative) in the recent past because of
reliability concerns regarding the level instruments.
j
'
At the management meeting conducted on March 8, 1994, the licensee committed
to determine what corrective actions would be necessary to address the fact
i
that the HPCI system initiated and injected for this transient.when it was not
in the design basis for the system to initiate.
3.2.5' Licensee Event Notification 1
The inspection team found that the information provided by the licensee to the
I
NRC in their initial notification was misleading.
The licensee's initial
notification stated that reactor water level dropped to -37 inches and that-
16,000 gallons of water were injected into the reactor vessel by the HPCI and
RCIC systems.
In actuality, the level dropped to only -25 inches and the
amount of water injected into the vessel by the HPCI and RCIC-systems was only
j
about 1000 gallons. The licensee had not verified the accuracy of the
'
.
.
-
_
l
p
.-
1
-7-
I
information that had been provided and did not provide periodic updates, to
the NRC, of plant conditions while the plant was in an unusual event.
3.2.6 Assessment of Licensee Review Efforts
At the time the inspection team arrived on site (the afternoon of March 3),
.,
the licensee had issued an outage schedule that indicated a planned reactor
!
startup on March 5.
The scheduled date, the fact that the posttrip review had
.
- already been completed and approved by the SORC (with only on'e open item), and
the initial lack of a formal event review process gave the inspection team the
impression that the licensee was focusing on an early startup rather than on
thoroughly or systematically evaluating the transient and determining the root
cause for the scram.
The inspection team noted that there had been little or no oversight of the
ongoing activities in the early stages of the event followup. There was
little evidence of quality assurance oversight in the early stages of the
process; however, quality assurance involvement was more visible after the
t
inspection team arrived.
The inspection team asked if the offsite review committee (SRAB) had been
i
involved to any extent in evaluating the activities associated with the scram
followup. The SRAB chairman stated, on March 4, that they had not been
involved, but a SRAB meeting (by telephone conference call) was convened on
'
March 5.
The SRAB was briefed on the issues and the licensee's' approach to
resolution of the issues. The SRAB identified no concerns with the licensee's
approach to resolving the issues.
4 LICENSEE CONCLUSIONS
The licensee concluded that the most likely cause of. the transient was a
failure of a relay in the valve transfer control card of the DEHCS.
The
licensee was not able to duplicate the event, but concluded that it was the
only likely failure that could give the transient response observed.
On March 8, 1994, the Vice President-Nuclear and members of his staff met in
,
the Region IV office with Regional management to discuss the results of their
troubleshooting efforts. The licensee was advised at the meeting that the NRC
had no concerns that should prevent them from starting up the reactor when the
licensee was satisfied that all the problem and safety concerns have been
identified and corrected.
,
The suspect cards in the DEHCS were replaced and a reactor startup was
initiated on March 12. During plant start up, with reactor pressure being
controlled by the bypass valves, a pressure transient was experienced when all
three bypass valves went shut twice within 3 minutes from no immediately
identifiable cause.
The licensee investigated the cause of the bypass valve
failure and determined that the problem was with the +26/+24-Vdc power supply
to the DEHCS.
They found that the primary power supply had failed to +21
+-s
--
g-
_
-
,.
-8-
volts and the secondary (backup) power supply had a continuous negative
spiking condition in its output.
The inspection team inquired as to why the power supply problem was not'
. detected during the earlier testing. A licensee representative stated that
the power supplies had been tested and both were stable at that time.
The
simulator power supplies had then been used for the remainder of the testing.
The licensee replaced the power supplies and subsequent testing resulted in no
further problems being identified. A reactor startup was initiated on
March 23 and the DEHCS operated satisfactorily.
The inspection team expressed concern that the power supplies to the DEHCS
were not monitored or alarmed.
If either one of the power supplies were to
fail again, the licensee would have no indication that the' plant was
vulnerable to another transient if the one remaining power supply experienced
a problem.
Licensee representatives stated that they would evaluate the
concern, but made no specific commitment regarding the monitoring of the power
,
supplies.
i
1
i
i
4
-
-
- -
-
a
m
a
h
(
p
,-
."
ATTACHMENT
1 PERSONS CONTACTED
1.1 Licensee Personnel
R. L. Beilke, Acting Radiological Manager
L. E. Bray, Regulatory Compliance Specialist
S. L. Bray, Quality Assessment Supervisor
R. Brungardt, Operations Manager
J. W. Dutton, Nuclea'r Training Manager
C. M. Estes, Corrective Actions Program Overview Group (CAP 0G)
R. L. Gardner, Plant Manager
G. R. Horn, Vice President, Nuclear
J. E. Lynch, Plant Engineering Manager
E. M. Hace, Senior Manager Site Support
J. M. Heacham, Senior Nuclear Division Manager of Safety Assessment
M. E. Unruh, Maintenance Manager
D. A. Whitman, Division Manager of Nuclear Support
V. L. Wolstenholm, Division Manager of Quality Assurance
1.2 NRC Personnel
J. E. Gagliardo, Chief, Project Branch C
E. E. Collins,' Project Engineer
J. L. Pellett, Chief, Operations Branch
R. A. Kopriva, Senior Resident Inspector
The personnel listed above attended the exit meeting.
In addition to the
personnel listed above, the inspection team contacted other licensee personnel
during this inspection period.
2 EXIT MEETING
An exit meeting was conducted on March 7, 1994. During this meeting, the
inspection team reviewed the scope and findings of this report. During a
management meeting conducted on March 8, 1994, the licensee committed to
determine what corrective actions would be necessary to address the initiation
of the high pressure coolant injection system during the transient, when the
design basis for the system indicated that injection should not have occurred.
The licensee did not identify as proprietary any information provided to, or
reviewed by, the inspection team.