Information Notice No. 91-13: Inadequate Testing Of Emergency Diesel Generators (EDGs)

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

                                March 4, 1991


Information Notice No. 91-13:  INADEQUATE TESTING OF EMERGENCY 
                                   DIESEL GENERATORS (EDGs)


Addressees:

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

Purpose:

This information notice is intended to alert addressees to inadequacies in 
the testing of emergency diesel generators at nuclear power plants.  
Specifically, some EDG testing has not adequately verified the capability of 
the EDG to carry its maximum expected loads and other tests have failed to 
properly verify the operation of the load shedding logic for the EDG.  These 
testing deficiencies indicate that other licensees may have similar 
deficiencies that have not yet been detected.  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:

Yankee Atomic Electric Company, the licensee for the Yankee Nuclear Power 
Station (Yankee Rowe), recently identified that the EDGs at the Yankee Rowe 
plant were undersized for their maximum accident electrical load demand.  
The EDGs failed to reach their technical specification (TS) required loading 
of 400 kilowatts (kW) while connected in parallel to the electrical grid.  
The licensee then retested the EDGs using a resistor bank in place of the 
grid.  Using this test configuration, Yankee Atomic determined that the 
frequency of the generator dropped to 58 hertz with a maximum electrical 
output of 385 kW.  The licensee attributed the reduction in power output, in 
part, to high ambient air temperatures that reduced engine power and 
cooling.  The licensee had not considered the effect of ambient air 
temperature during previous testing.  The existing air temperature was 
within design basis limits.  Although high ambient air temperature was the 
immediate cause of the failure to attain 400 kW, the root cause was the fact 
that the engines were undersized.  The manufacturer's rating of 400 kW also 
failed to consider the effects of ambient air temperature.  The licensee 
calculated that the dependable capacity of these EDGs was between 359 kW and 
399kW.  Yankee Atomic eventually replaced all 3 EDGs with models capable of 
producing 600 kW (450 kW continuous) because the 

9102260165 
.

                                                            IN 91-13
                                                            March 4, 1991
                                                            Page 2 of 3


estimated electrical load on the EDGs is very close to 400 kW and 
modifications to the original EDGs failed to raise their output 
capabilities.  The Omaha Public Power District, the licensee for Fort 
Calhoun Station, Unit 1, reported a similar problem with their EDGs.  At 
Fort Calhoun, the EDGs could temporarily achieve their design rating at high 
ambient temperatures, but they could not maintain this power level for an 
extended period of time.

At the Vermont Yankee Generating Station, during a safety system functional 
inspection, the U.S. Nuclear Regulatory Commission (NRC) staff determined 
that the surveillance testing of the station EDGs was inadequate.  The staff 
found that there was insufficient electrical loading during testing as well 
as insufficient compensation for the expected electrical power factor (pf).  
Specifically, the plant surveillance test for the EDGs required that the 
EDGs maintain a load of 2500 to 2750 kW, which is equivalent to 2500 to 2750 
kilovolt-amperes (kVA) at pf=1.0.  The licensee considered this limit suffi-
cient to meet the maximum load of 2471 kW specified in the final safety 
analysis report (FSAR).  However, during subsequent analysis, the licensee 
determined that the worst-case EDG loading could actually reach 2751 kW.  In 
addition, the EDG would realistically experience an electrical power factor 
of 0.85 lagging, causing the actual generator output current to be 
significantly higher (i.e., assuming generator output voltage is maintained 
roughly constant, generator output current and kVA at pf=0.85 are 
approximately 1.18 times the generator output current and kVA at pf= 1.0.)  
Therefore, the existing surveillance did not adequately demonstrate the 
capability of the EDG to carry its accident kW and current loading.  As a 
result, the licensee is modifying its EDG testing requirements to ensure 
that both the engine and the generator will be appropriately tested to 
demonstrate their capability to produce the accident electrical load.

At Millstone Nuclear Power Station, Unit 3, the Northeast Nuclear Energy 
Company (the licensee) determined that its testing of the EDG load shedding 
circuitry was inadequate.  In particular, the surveillance tests did not 
verify the load shedding of five nonsafety-related loads and two 
safety-related loads.  This condition was a violation of TS requirements and 
had existed from initial plant operation in April 1986 until January 1990.  
Subsequent testing verified that the load shedding circuits operated 
properly.

Discussion:

The events at Yankee Rowe, Fort Calhoun, and Vermont Yankee illustrate the 
failure of EDG testing to demonstrate properly the capability of the EDGs to 
assume the design-basis electrical loading during an accident.  The intent 
of the required surveillance testing is to assure that the EDG can 
dependably carry its accident loads.  It is important that the worst-case 
conditions (frequency, voltage, electrical power factor, and environment) be 
considered when the EDG is tested.  In particular, the effects of the 
electrical power factor and the environment may have been overlooked.  It is 
important that procedures give personnel performing the test sufficient 
information to allow them to load the EDG to simulate worst-case conditions.  
Since the simulation of environmental conditions is generally impractical, 
it may be necessary to use analysis to correct for environmental conditions 
or to show that the effect 
.

                                                            IN 91-13
                                                            March 4, 1991
                                                            Page 3 of 3


of such conditions is negligible for the EDG.  Information from the vendor 
may help in determining what changes, if any, may be necessary.  It is also 
important for procedures to give adequate guidance to ensure that the EDG is 
not inadvertently overloaded during the test.  

The last event illustrates the failure of EDG testing to verify adequately 
the proper operation of all of the logic necessary to support the EDG in an 
accident.  Millstone Unit 3 was given as an example, but other similar cases 
have been reported.  

Generic Letter 88-15, "Electric Power Systems - Inadequate Control Over 
Design Processes," discussed problems similar to those in this information 
notice. 

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 NRR project 
manager.




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


Technical Contacts:  C. H. Woodard, Region I
                     (215) 337-5261

                     A. J. Kugler, NRR
                     (301) 492-0834


Attachment:  List of Recently Issued NRC Information Notices
.