Information Notice No. 92-74: Power Oscillations at Washington Nuclear Power Unit 2

                                 UNITED STATES
                         NUCLEAR REGULATORY COMMISSION
                     OFFICE OF NUCLEAR REACTOR REGULATION
                            WASHINGTON, D.C.  20555

                               November 10, 1992


NRC INFORMATION NOTICE 92-74:  POWER OSCILLATIONS AT WASHINGTON NUCLEAR POWER
                               UNIT 2 


Addressees

All holders of operating licenses or construction permits for boiling-water
reactors (BWRs).

Purpose  

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees to a recent event involving power oscillations in
an operating region where instability had not been specifically predicted.  It
is expected that recipients will review the information for applicability to
their facilities and consider actions, as appropriate, to avoid similar
problems.  However, suggestions contained in this information notice are not
NRC requirements; therefore, no specific action or written response is
required.

Background

On March 9, 1988, a thermal hydraulic instability event occurred at LaSalle
Unit 2.  The NRC discussed this event in Information Notice 88-39, "LaSalle
Unit 2 Loss of Recirculation Pumps with Power Oscillation Event," and
Bulletins 88-07 and 88-07, Supplement 1, "Power Oscillations in Boiling Water
Reactors."  In the first bulletin, the NRC requested licensees to establish
procedures and give training to reactor operators to enable them to recognize
oscillations and to take appropriate actions.  In the supplement, the NRC
requested licensees to implement the General Electric (GE) Interim 
Recommendations for Stability Actions, designated the Interim Corrective
Actions (ICA).  GE defined the exclusion regions on the power/flow map in
which, with varying probability, instability might be expected.  Three regions
were defined in which operation was to be avoided (immediate exit if entered)
or limited (e.g., when required during startup).  These regions were based on
operating or test experience for reactors with GE fuel.  The exclusion regions
for new fuel designs were to be reevaluated and justified based on any
applicable operating experience, calculated changes in core decay ratio using
NRC-approved methodology, and/or core decay ratio measurements.   Since the
LaSalle event in 1988, the NRC and the BWR Owners' Group (BWROG) have
conducted extensive analyses and reviews of various aspects of stability while
developing long-term solutions to augment or replace the ICA.  On 
March 18, 1992, the BWROG sent a letter (BWROG-92030) to BWROG members



9211040206.

                                                            IN 92-74 
                                                            November 10, 1992
                                                            Page 2 of 5

 
transmitting "Implementation Guidance for Stability Interim Corrective
Actions."  In this letter, the BWROG emphasized the need for caution when
operating near the exclusion regions.  The BWROG also recommended reexamining
procedures and training to reflect uncertainties in the definition of
exclusion region boundaries.

Description of Circumstances

On August 15, 1992, Washington Nuclear Power Unit 2 (WNP-2) experienced power
oscillations during startup.  The event occurred early in cycle 8 operation. 
During cycle 8, the licensee had two previous startups without incident.  The
reactor core consisted primarily of Siemens fuel, with about 74 percent of
this fuel in 8x8 fuel assemblies and about 25 percent in 9x9 fuel assemblies,
and with the remainder of the core consisting of various lead test assemblies. 
The 9x9 fuel assembly used in WNP-2, designated 9x9-9x, has a higher flow
resistance than the 8x8 fuel assembly with a difference of about 10 percent in
pressure drop.  These 9x9 fuel assemblies were loaded during cycles 7 and 8.

About 33 hours before the event, the licensee commenced a controlled power
reduction from full power to 5-percent power to repair a valve packing leak in
the drywell.  After completing the repairs, the licensee began a return to
full power.  The licensee increased reactor power to about 34 percent and then
held it at that level for 3 hours to perform turbine bypass valve tests and
control rod drive system timing tests.  The recirculation system was operated
with flow control valves (FCVs) full open and pumps at slow speed.  

After completing the tests, the operators continued the restart up the
(approximately) 30-percent flow line to about 36-percent power (Figure 1). 
This is at a power above the recirculation pump cavitation region.  The
operators then began closing one of the two FCVs in preparation for shifting
the associated recirculation pump to fast speed.  During this change, in which
power and flow decreased along the 76-percent rod line to a power/flow of
about 34/27 percent, the operators observed power oscillations first on the
average power range monitors (APRMs) and then by local power range monitors
(LPRMs) downscale indications.  Upon recognizing the power oscillations, the
plant operators manually initiated a reactor scram.  Post-event review
indicated that the 2-second-period oscillations were in-phase (core-wide) and
had grown to a peak-to-peak amplitude of about 25 percent of rated power. 
Most of the oscillation amplitude increase occurred in an interval of about 
1 minute with the oscillations continuing at the limiting (maximum) amplitude
for an additional minute before scram.  The oscillations occurred while the
reactor was operating at a power about 4 percent of rated power below the
lower exclusion region boundary line (the nominal 80-percent rod line). 
During later review, the licensee found no indication that fuel had failed
because of the event. 
.

                                                            IN 92-74
                                                            November 10, 1992
                                                            Page 3 of 5


The NRC sent an Augmented Inspection Team (AIT) to the site to determine the
possible causes and relevant facts of this event.  The AIT concluded that the
primary cause of the oscillations was very skewed radial and bottom peaked
axial power distributions in the reactor (1.92 radial peaking factor and 
1.62 core average axial peaking factor).  These power distributions resulted
from (1) the control rod pattern that the shift technical advisor selected for
increasing the power and shifting the recirculation pump speed, and (2) the
relationship of this control rod pattern to the specific WNP-2 cycle 8 core
fuel loading configuration.  These rod patterns were primarily directed
towards achieving the target full power configuration and did not consider
stability concerns.  

The AIT also found, by analyses using the LAPUR code, that a contributor to
the oscillations was the core loading, consisting of a mixed core with
unbalanced flow characteristics between the new 9x9-9x fuel and the old 
8x8 fuel.  The analyses indicated that a full core of the 9x9-9x fuel would be
significantly less stable than the old 8x8 fuel, and that the mixed core was
less stable than a fully loaded core of either fuel type.  The analyses also
indicated that while the oscillations would be in-phase (core-wide), as
observed in the event, the out-of-phase (regional) instability boundary would
be very close to the in-phase boundary (Figure 1).  The AIT found that small
changes in operating conditions could have resulted in out-of-phase
oscillations, which would have been more difficult for the APRM system to
detect.

WNP-2 has a Siemens Advanced Neutron Noise Analysis (ANNA) monitor, a
stability monitor required by technical specifications only if the licensee
intends to enter the lower exclusion region.  Since the licensee did not
intend to enter the exclusion region during this startup, the ANNA monitor was
not put into the observation mode, although it was gathering data which was
used later to confirm stability calculations performed after the event.

The licensee successfully restarted the unit after implementing the following
restrictions for maintaining the limits on rod withdrawal patterns and power
distribution in the low flow regions of concern. 

�    The licensee analyzed the control rod patterns for stability before
     startup, and the operator could not change these patterns without
     analysis and review.

�    The calculated maximum total peaking factor was less than 3.4.
.

                                                            IN 92-74
                                                            November 10, 1992
                                                            Page 4 of 5


�    The calculated core average axial peaking factor was less than 1.45.

�    The Minimum Critical Power Ratio was greater than 2.2.

�    The licensee analyzed the conditions at FCV closure and found a decay
     ratio of less than 0.5.  The recirculation pump was shifted to fast 
     speed with the reactor power less than 33 percent and the feedwater
     temperature greater than 146.1 �C (295 �F).

�    The licensee continuously used the ANNA monitor when the reactor was
     operating above 25 percent power and below 50 percent flow.

Further detailed description of the event can be found in the AIT Inspection
Report No. 50-397/92-30.

Discussion

The WNP-2 power oscillation event indicates that the boundaries of the ICA
regions, or modifications approved for various reactor technical
specifications, do not necessarily encompass all stability limits. 
Instability may occur beyond these boundaries if the reactor is operated with
configurations outside those used to define the  boundaries.  This event
presented direct evidence that the following factors can be significant
contributors to the possibility of unstable operation.

�    Power distributions involving extremely skewed radial and axial peaking
     factors can induce unstable operation even in regions or with operating
     conditions not otherwise considered susceptible to oscillations. 

�    Core loading patterns involving a mixture of fuel types with differing
     flow resistances can contribute to instability.  

�    Reactors with two-speed recirculation pumps and FCVs can hinder stability
     because of the narrow range of operation between pump cavitation regions
     and possible instability regions. 

The event also indicates the value of operating a stability monitor.  The ANNA
monitor could have given the operators information that instability was
imminent, prompting them to alter operations to avoid the oscillations.

.

                                                            IN 92-74
                                                            November 10, 1992
                                                            Page 5 of 5


This information notice requires no specific action or written response.  If
you have any questions about the information in this notice, please contact 
one of the technical contacts listed below or the appropriate Office of
Nuclear Reactor Regulation (NRR) project manager.


                                        ORIGINAL SIGNED BY


                                   Brian K. Grimes, Director
                                   Division of Operating Reactors Support
                                   Office of Nuclear Reactor Regulation

Technical contacts:  Howard Richings, NRR
                     (301) 504-2888

                     Peter C. Wen, NRR
                     (301) 504-2832
                   
Attachments:
1.  Figure 1.  Best-Estimate Lines of Constant 
      Decay Ratio=1.0 for Actual Conditions of 
      WNP-2 8/15/1992 Event, Assuming Constant 
      Power Distribution
2.  List of Recently Issued NRC Information Notices  
.