Information Notice No. 96-41: Effects of a Decrease in Feedwater Temperature on Nuclear Instrumentation

                                 UNITED STATES
                         NUCLEAR REGULATORY COMMISSION
                         WASHINGTON, D.C.  20555-0001

                                 July 26, 1996

                               ON NUCLEAR INSTRUMENTATION


All holders of operating licenses or construction permits for pressurized
water reactors (PWRs).


The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees to the potential for operation above licensed power
as a result of a decrease in feedwater temperature event affecting nuclear
instrumentation.  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 February 14, 1996, the licensee for the Comanche Peak Steam Electric
Station was operating Unit 2 at 95 percent rated thermal power near end-of-
core life when a significant reduction in feedwater temperature occurred
because of the loss of feedwater heaters.  This reduction, in turn, caused a
reduction in the reactor coolant system cold-leg temperatures.  The colder
reactor coolant temperature, with a large negative moderator temperature
coefficient, caused reactor power to increase to approximately 102 percent
according to ex-core nuclear instrumentation.  The nitrogen-16 (N-16)
detection system reached the overpower turbine runback setpoint (109 percent)
and initiated a turbine runback.  The N-16 detection system measures N-16
activity in the primary coolant as a measure of the total power generation. 
This system is a substitute for the resistance temperature detector over-
temperature and over-power reactor trip functions used at other Westinghouse
PWRs.  The plant stabilized at an indicated power of approximately 97 percent
according to the ex-core nuclear instrumentation.

After approximately 90 minutes, a second similar turbine runback occurred
while restoring balance-of-plant equipment.  Following this runback, reactor
power was stabilized at approximately 100 percent according to nuclear
instrumentation.  During the next 30 minutes, the reactor was operated at
approximately 100 percent power as indicated by nuclear instrumentation, with
reactor coolant temperatures below normal.  The licensee noted that the N-16

9607220160.                                                            IN 96-41
                                                            July 26, 1996
                                                            Page 2 of 3

detection system indicated approximately 106 percent power and the computer-
based plant calorimetric system indicated approximately 102 percent power. 
Subsequently, the reactor power was reduced to less than 100 percent by all


There are three aspects of this event which have generic implications.  First,
with a loss of secondary plant efficiency, programmed TAve can no longer
reliably represent core thermal power.  Second, the venturi-based input into
the computer-based calorimetric system may not be accurate with cold
feedwater.  And third, the final safety analysis report had not analyzed this
transient accurately.

Following the second runback, operators noted that reactor power indicated
�100 percent according to nuclear instrumentation.  Although the operators
knew that cold feedwater could cause an increase in the amount of neutron
attenuation, they believed that the nuclear instrumentation indicated
conservatively (i.e., higher than actual) because they were maintaining TAve
approximately 1.7 �C [3 �F] above TRef.  The licensee could not use the
computer-based calorimetric until some time after the second turbine runback
due to maintenance activities.  TRef, based on the main turbine impulse
pressure, is programmed as a function of turbine load and, for normal
efficiency, is a good representation of thermal power.  When the unit lost the
feedwater heaters, the plant efficiency decreased.  Because the main turbine
electro-hydraulic control system maintained generator output, core thermal
power increased to account for the loss of efficiency, and thus, TRef no
longer accurately represented the core thermal power.

The cold-leg temperature is a more appropriate indicator of the accuracy of
the nuclear instrumentation than programmed TAve.  As the cold-leg temperature
decreased, the amount of neutron attenuation in the downcomer area surrounding
the core increased and hence affected the amount of neutrons reaching the
detectors.  The licensee analysis showed that for every 0.6 �C [1 �F] of cold-
leg temperature change, the nuclear instrumentation was affected by 0.6 to 0.8
percent power.  A review of the second transient showed that the cold-leg
temperature was approximately 2.5 �C [4.5 �F] lower than when the detectors
were last calibrated.  This corresponded to a 3 to 4 percent error, which
corresponded to the difference in the actual versus the indicated power (104
percent actual versus 100 percent indicated).

During the review, the licensee noted that the computer-based calorimetric was
4 percent lower than the actual thermal power (N-16 power monitor).  The
calorimetric was based on feedwater flow measured by venturis.  Although the
calorimetric calculation used feedwater temperature as an input, temperatures
significantly different than the normal 227 �C [440 �F] introduced errors into
the calculation.

Finally, the actual events involved temperature and power levels that exceeded
those in the analysis of the "Decrease in Feedwater Temperature" event
presented in Chapter 15 of the licensee final safety analysis report.  In that
.                                                            IN 96-41
                                                            July 26, 1996
                                                            Page 3 of 3

analysis, the inadvertent opening of the low-pressure heater bypass valve,
coupled with the trip of the heater drain pumps, resulted in a feedwater
temperature drop of less than 19 �C [35 �F], and a corresponding power
increase of less than 10 percent.  In the actual event, the feedwater
temperature dropped by approximately 111 �C [200 �F], and the licensee
calculated that reactor power would have increased by approximately 35 percent
without operator or protective actions.  The licensee determined that although
the initiating events were the same, the Chapter 15 analysis did not account  
for the loss of extraction steam to the high-pressure heaters, which was the
cause of the temperature difference.  During the event, a level imbalance
occurred between the two heater drain tanks, which resulted in the isolation
of extraction steam.

The NRC staff review of analyses of feedwater temperature events at similar
facilities revealed that most of these analyses assumed similar initiating
events as the Comanche Peak analysis and had similar conclusions concerning
the amount of feedwater temperature drop.  The licensee has reanalyzed the
event to include a 119 �C [246 �F] feedwater temperature drop and concluded
that all accident analysis parameters remained within requirements.

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 project manager.

                                       signed by

                                    Brian K. Grimes, Acting Director
                                    Division of Reactor Program Management
                                    Office of Nuclear Reactor Regulation

Technical contacts:  Harry A. Freeman, RIV
                     (817) 897-1500

                     Chu-Yu Liang, NRR
                     (301) 415-2878

Attachment:  List of Recently Issued NRC Information Notices


Page Last Reviewed/Updated Friday, May 22, 2015