Information Notice No. 91-52: Nonconservative Errors in Overtemperature Delta-Temperature (OTdT) Setpoint caused by Improper Gain Settings

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

                              August 29, 1991 


Information Notice No. 91-52:  NONCONSERVATIVE ERRORS IN OVERTEMPERATURE
                                   DELTA-TEMPERATURE (OTdT) SETPOINT CAUSED 
                                   BY IMPROPER GAIN SETTINGS 


Addressees:

All holders of operating licenses or construction permits for Westinghouse 
(W)-designed nuclear power reactors.  

Purpose:

This information notice is intended to alert addressees to potential 
problems resulting from improper voltage gain settings applied to components 
in the overtemperature delta-temperature (OTdT) reactor protection system 
trip circuitry at Westinghouse (W)-designed nuclear power reactors.  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 July 17, 1991, Duke Power Company notified the NRC of a problem with the 
OTdT reactor trip circuit at McGuire and Catawba Nuclear Stations which may 
have prevented the system from operating correctly over its entire operating 
range for reactor coolant loop average temperature (T-avg).  Engineers dis-
covered the problem at McGuire while attempting to rescale their OTdT 
setpoints for the upcoming new fuel load.  The engineers found that the gain 
applied to the lead/lag amplifier for the T-avg portion of the OTdT circuit 
was set too high (Westinghouse 7300 series process equipment card TY412A), 
based on the current scaling of the hardware.  This caused the T-avg input 
to "saturate" at approximately 597�F for McGuire and 592�F for Catawba which 
prevented the OTdT circuit from reducing the OTdT setpoint as the T-avg 
increased above these temperatures.  The required input temperature range 
for T-avg is 530�F to 630�F.  Therefore, the OTdT setpoint calculation was 
in error above these "saturation" temperatures since the lead/lag amplifiers 
would not produce a proportional output.

Following discussions with Westinghouse Electric Corporation, the licensee 
corrected the problem at both McGuire and Catawba by redistributing the gain
associated with the T-avg term of the OTdT setpoint.  This was accomplished 
by replacing resistors in the lead/lag circuit to ensure that the gain was 
not so high that it could cause "saturation" of its output over the entire 
input 

9108230188 
.

                                                            IN 91-52
                                                            August 29, 1991
                                                            Page 2 of 3


range.  On July 19, 1991, the licensee modified the OTdT instrumentation at 
both facilities.

On July 31, 1991, the Commonwealth Edison Company informed the NRC that it 
had reviewed the gain settings for the OTdT circuits at its Braidwood and 
Byron Stations and had discovered conditions similar to those at McGuire and 
Catawba.

Discussion:

The OTdT reactor trip protects against departure from nucleate boiling 
(DNB).  The trip function operates by comparing indicated loop 
delta-temperature (dT) to the calculated setpoint, causing a reactor trip 
when two or more dT channels exceed their setpoint.  Analog circuitry for 
each reactor coolant loop continuously calculates the setpoint for this trip 
according to the temperature in the loop, the neutron flux distribution in 
the reactor, and reactor coolant system (pressurizer) pressure.  The 
principal concern for the function of the OTdT trip is for certain rod 
withdrawal events in which OTdT could be the primary safety-related trip 
mitigating the event.  It also functions as a safety-related backup for 
several other transients.

The problem involves the scaling used in the OTdT hardware.  If the hardware 
is scaled such that the output gain on the T-avg lead/lag amplifier is 
greater than unity (1.0), the amplifier will saturate at some T-avg and 
prevent further adjustment of the OTdT setpoint.  The equation for the OTdT 
setpoint at McGuire and Catawba includes as parameters lead-lag and lag 
constants, a bias (K1), and gains (K2 and K3).  Gains K2 and K3 are the 
adjustments for the effects of temperature and pressure, respectively, on 
DNB limits. 

The licensee indicated that the OTdT hardware had operated in the degraded 
condition since 1983 for McGuire and since startup for Catawba.  The 
original K2 value, provided by Westinghouse, was 0.0133/�F for McGuire.  The 
hardware was scaled by Westinghouse such that the lead/lag amplifier gain 
corresponding to this K2 was less than 1.0.  With a gain setting of less 
than 1.0, the amplifier does not overrange since the gain output voltage 
cannot exceed 10 volts.  The degraded condition was introduced at McGuire 
when the licensee rescaled the OTdT trip hardware as a function of 
performing a core reload analysis to allow for the use of Westinghouse 
"optimized" fuel.  As part of this change to the OTdT parameters, K2 was 
increased to its current value of 0.0222/�F.  Using the Westinghouse scaling 
methodology, however, the licensee scaled the OTdT hardware such that the 
output gain on the T-avg lead/lag amplifier was greater than 1.0, which 
introduced the problem. 

Preliminary information concerning Catawba indicated that the original K2 
value, supplied by Westinghouse, was also 0.0133/�F.  Sometime in 1983 or 
1984, following a McGuire transient which indicated the need to modify the 
OTdT and overpower delta-temperature (OPdT) setpoints, K2 was increased to 
its current value of 0.02401/�F.  It was then that the licensee incorrectly 
scaled the OTdT hardware such that the output gain on the T-avg lead/lag 
amplifier was greater than 1.0. 

At some plants licensees have avoided the problem by using different scaling 
methodologies or have identified the problem during the scaling process and 
.

                                                            IN 91-52
                                                            August 29, 1991
                                                            Page 3 of 3


modified the hardware (redistributing the gain) similar to the corrective 
measures performed by Duke Power Company.  

This information notice requires no specific action or written response.  If
you have any questions about the information in this notice, please call 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:  William Orders, RII
                     (803) 831-2963

                     Thomas Koshy, NRR 
                     (301) 492-1176 

Attachment:  List of Recently Issued NRC Information Notices