Information Notice No. 94-83: Reactor Trip Followed by Unexpected Events

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

                               December 6, 1994


NRC INFORMATION NOTICE 94-83:  REACTOR TRIP FOLLOWED BY UNEXPECTED EVENTS


Addressees

All holders of operating licenses or construction permits for nuclear power
reactors.

Purpose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees to a recent reactor trip followed by a series of
unexpected events and equipment failures.  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.

Description of Circumstances

On September 8, 1994, the River Bend Station experienced an automatic reactor
trip on a reactor water level high (Level 8) trip signal.  A Level 8 trip
occurs at 51 inches, but the operators verified that the level was and had
been stable at 36 inches.  After implementing the appropriate scram recovery
procedures, the operators observed that the expected turbine trip had not
occurred; therefore, they tripped the turbine manually about 10 minutes into
the event.  The operators also opened the generator output breakers which had
failed to open.  After the output breakers were opened, electric power was
unexpectedly lost to the following systems and components:  safety parameter
display system, emergency response information system, reactor protection
system (RPS), feedwater pumps, condensate pumps, reactor recirculation pumps,
a normal service water pump, and two circulating water pumps.  In addition,
due to the loss of the RPS, a containment isolation occurred which isolated
both main steam and feedwater systems.  In spite of these problems, the
operators determined that no emergency level had been reached.  

During the recovery period, the operators manually opened one safety/relief
valve for pressure control and initiated operation of the reactor core
isolation cooling (RCIC) system; however, the RCIC pump turbine tripped on
overspeed.  The operators then initiated the high pressure core spray (HPCS)
system to provide condensate makeup to the reactor vessel.  Because the
reactor vessel level was maintained above the emergency core cooling system
(ECCS) actuation setpoints, no automatic initiation of the ECCS occurred.  In
addition, because no power was lost to the safety-related busses, no emergency
diesel generators started.

9411300353.                                                            IN 94-83
                                                            December 6, 1994
                                                            Page 2 of 3


Discussion

The anomalous occurrences began with a high water level signal, which was
subsequently determined to be a spurious signal received by RPS channels C 
and D.  The reactor scram initiated on a 1-out-of-2-taken-twice logic.  The
licensee root cause determination concluded that Rosemount Model 1153 level
transmitters with no or minimal damping are susceptible to providing spurious
trips as a result of signal noise in process variables.  The licensee replaced
these model transmitters due to incompatibilities between damping and
instrument response time.  

General Electric Nuclear Energy (GENE) addressed the concern about signal
noise in process variables in SIL No 463, R1 dated July 9, 1991 and presented
recommended actions.  Since the transmitters are not the source of the noise,
GENE recommends that a root cause evaluation be performed to determine the
source of the noise and that, if there are differences between the required
instrument response time and the measured response time, an evaluation be
performed and appropriate adjustments made.

The anticipated turbine trip, which normally follows a valid high water level
scram, did not occur because this actuation is based on a 2-out-of-3 logic and
only channel C of channels A, B, and C had received a trip signal.

Usually, an automatic reactor trip will lead to motoring of the main generator
and an automatic trip of the generator on reverse power when the reverse power
exceeds the set point of approximately 3 MW (million Watts) on either one of
two reverse power relays.  However, because of an abnormally high inductive
load (about 200 MVAR [million volt amperes reactive]) on the generator prior
to the reactor scram, the reverse power trip set point of both relays had been
effectively shifted to about 10 MW on one relay and 20 MW on the other relay. 
Because the highest reverse power attained was about 10 MW, the automatic
turbine and main generator trip did not occur.  

The anticipated automatic trip of the turbine and main generator would have
led to a fast transfer (within 6 to 10 cycles) of power from the normal
station service (main generator) to the preferred station service (grid).  The
manual trip, however, resulted in the plant load being carried by the main
generator while it was losing voltage and frequency which, in turn, led to a
slow transfer on undervoltage about 1.3 minutes later.  This series of events
led to a loss of power to certain non-safety equipment and the RPS motor
generator sets.  The loss of power to the RPS led to containment isolation. 
Plant response during a slow transfer had not been addressed in operator
training nor was it modeled on the control room simulator.

The RCIC turbine tripped on overspeed because the governor valve had failed in
the open position.  Attempts to manually stroke the valve were unsuccessful. 
After the reactor was shut down, the valve bonnet was disassembled and the
valve stem was found to be stuck.  This event and others are described in
Information Notice 94-66, "Overspeed of Turbine-Driven Pumps Caused by
Governor Valve Stem Binding," issued on September 19, 1994.  One potential
cause for the stuck valve stem is galvanic corrosion of the valve stem
material while the system is in its standby condition..                                                            IN 94-83
                                                            December 6, 1994
                                                            Page 3 of 3


Other failures that occurred after the reactor scram involved operational
failures of one safety-related motor operated valve (MOV) and four non-safety
related MOVs.  These failures, in this sequence, did not cause other event
consequences.  The failure of the safety related valve was determined to have
been caused by an electrical lead within Limitorque actuator, Model SMB-OO,
which had been sharply bent with the bend adjacent to the cover.  Vibration
associated with valve operation caused a chafing action which over time
permitted arcing and a fuse to blow.  Contact positions LS-1 and LS-9 were
determined to be the only positions with insufficient clearance and thus
susceptible to this type of failure.  

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.

                                    /s/'d by BKGrimes


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

Technical contacts:  T. Stetka, RIV 
                     (817) 860-8247

                     J. Carter, NRR
                     (301) 504-1153

Attachment:
List of Recently Issued NRC Information Notices