Lock-Up of Emergency Diesel Generator and Load Sequencer Control Circuits Preventing Restart of Tripped Emergency Diesel Generator

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

                              January 31, 1991


Information Notice No. 91-06:  LOCK-UP OF EMERGENCY DIESEL GENERATOR 
                                   AND LOAD SEQUENCER CONTROL CIRCUITS
                                   PREVENTING RESTART OF TRIPPED EMERGENCY 
                                   DIESEL GENERATOR


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

Purpose: 
        
The information in this notice is intended to alert addressees of potential 
problems involving the restart of a tripped emergency diesel generator (EDG) 
because of "lock-up" of EDG or load sequencer control circuits.  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 do not 
constitute NRC requirements; therefore, no specific action or written 
response is required.

Description of Circumstances:

On March 20, 1990, operations personnel at the Alvin W. Vogtle Plant, Unit 
1, experienced difficulties when attempting to restart the in-service EDG.  
In response to a valid bus undervoltage condition, the EDG had automatically 
started and energized its associated 4.16 kV ac safety bus per design.  The 
EDG tripped after approximately one minute of operation.  A spurious jacket 
water high temperature signal is believed to have caused the trip.  The load 
sequencer had completed its cycle before the trip.  The EDG could not be 
restarted following the trip even though the condition that had caused the 
trip had cleared and a valid undervoltage condition and automatic start 
signal existed.  The EDG control circuits had functioned as designed but had 
locked-up such that subsequent restart attempts were prevented.  Although 
the EDG could have been started in the "emergency" mode by using the 
emergency start switch at the local panel, the local and remote control 
switches used for "normal" EDG start were ineffective.  No equipment 
failures or malfunctions were involved.  

The design of the load sequencer circuits that provide an automatic start 
signal upon sensing bus undervoltage and the interface between these 
circuits and the interlocks in the EDG air start system are such that the 
circuits will lock-up whenever a trip results in an undervoltage condition 
on the associated 



9101280052 
.

                                                            IN 91-06 
                                                            January 31, 1991 
                                                            Page 2 of 3 


safety bus.  During the March 1990 event at the Vogtle Plant, the initial 
attempts to reset the lock-up were unsuccessful, and the EDG was not 
restarted for approximately 18 minutes.  Had proper action been taken to 
reset the locked-up circuits, the EDG could have been restarted promptly. 
However, the operations personnel involved were not trained concerning the 
lock-up condition, and procedures providing guidance concerning actions 
required for restart did not exist. A detailed explanation of the design of 
the Vogtle Unit 1 load sequencer and EDG air start system is presented in 
NUREG 1410, "Loss of Vital AC Power and the Residual Heat Removal System 
During Mid-Loop Operations at Vogtle Unit 1 on March 20, 1990."

On November 14, 1990, the 1A EDG at the Kewaunee Nuclear Power Plant was 
being tested following replacement of the governor.  Because of an improper 
governor setting, the EDG tripped on overspeed before the load sequencer 
completed its cycle.  This trip caused the sequencer control circuits to 
lock-up in a condition that would have prevented loading of the EDG 
following a valid start.  Maintenance technicians present during the testing 
had to lift leads in the sequencer cabinet to clear the lock-up condition.  
The first attempt to clear the lock-up condition was unsuccessful because 
the leads were only removed momentarily, which resulted in remote 
indications that led the operators to believe that the sequencer had been 
reset.  However, the leads were relanded before time delay relays could time 
out and reset (approximately 90 seconds is required).  Approximately two 
hours were required for the plant personnel to troubleshoot the load 
sequencer and complete successful testing of the EDG.  Apparently, the plant
staff were not adequately trained to recognize and reset the lock-up 
condition, and plant procedures were not appropriately detailed to allow 
operators to cope with a situation where the EDG trips before the load 
sequencer completes its cycle.  The licensee is considering a modification 
to the sequencer to install a conveniently located reset switch to allow 
reset to be accomplished without having to lift leads.

Discussion: 

These incidents have raised concerns regarding the understanding of EDG and 
load sequencer control circuits and their interfaces, and the adequacy of 
procedures for restarting EDGs following unexpected trips.  EDG and load 
sequencer control systems are often complex.  As the result of either a 
valid or spurious trip signal, an EDG can shut down at any time before, 
during, or after cycling of the load sequencer.  Successful mitigation of 
the effects of accidents or transients and maintenance of the overall 
reliability of EDGs depends upon operations personnel having (1) sufficient 
knowledge of the associated instrumentation and controls; and (2) the 
ability to recognize and reset a lock-up condition.  
.

                                                            IN 91-06 
                                                            January 31, 1991
                                                            Page 3 of 3 


This information notice requires no specific action or written response. If 
you have any questions about the information in this notice, please contact 
the technical contact listed below or the appropriate NRR project manager.




                              Charles E. Rossi, Director
                              Division of Operational Events Assessment
                              Office of Nuclear Reactor Regulation


Technical Contact:  E. Nick Fields 
                    (301) 492-1173


Attachment:  List of Recently Issued NRC Information Notices 
.