05000440/LER-2001-004
| Perry Nuclear Power Plant | |
| Event date: | 10-01-2001 |
|---|---|
| Report date: | 10-30-2001 |
| 4402001004R00 - NRC Website | |
I. Introduction On October 1, 2001, it was determined that a non-conservative assumption for the steam carryover fraction of main steam has been applied in the General Electric methodology for calculating reactor core thermal power at the Perry Nuclear Power Plant (PNPP). This was documented in a General Electric (GE) Report entitled, "Impact of Steam Carryover Fraction on Process Computer Heat Balance Calculations, September 2001." This report states that the assumed carryover fraction in later model Boiling Water Reactors (BWRs) is non-conservative with respect to heat balance calculations. The process computer determines reactor thermal power using a heat balance calculation. This calculation is based on a summation of all heat sources raising the inlet feedwater, and other cold water sources, to steam exiting the reactor pressure vessel. The effect of this non-conservative carryover fraction assumption is that the calculated core thermal power could be as much as approximately 0.082% lower than the actual core thermal power.
Consequently, the plant has potentially operated at reactor core power levels in excess of the licensed power level by approximately 0.082%, or approximately 3 megawatts thermal (MWth).
Condition 2.F of the PNPP Operating License requires that violations of the operating license be reported to the NRC with a 24-hour Emergency Notification System call and a written follow-up report within 30 days. This issue was reported to the NRC Operations Center via the Emergency Notification System at 1914 on October 1, 2001 (Notification number 38337).
At the time of the event, the plant was operating in Mode 1 at 100% power, at normal operating temperature and pressure.
II. Event Description
Through an industry questionnaire, a question was raised about how plants determine main steam [SB] moisture content. During the evaluation of the question by PNPP staff, it was noted that the moisture content, or steam carryover fraction, for PNPP had been measured during the startup test program in 1987. The primary purpose of this startup test was to demonstrate compliance with warranty requirements, and was not a verification of heat balance calculation input parameters. The measured value was determined to satisfy the acceptance criteria at that time. However, the carryover fraction used in the core thermal power calculation was higher than this measured value, which results in an under-calculation of core thermal power. As the investigation progressed, an administrative 8-hour average power limit of 3754 MWth (a reduction of 4 MWth from the licensed maximum power level) was implemented as a conservative measure. On October 1, it was determined that the potential to have exceeded the licensed maximum power level existed.
III. Cause of Event
During the GE development of the core thermal power calculations (heat balance), a value for the steam carryover fraction was established based on steam dryer specifications. This value corresponded to the performance requirements specified for most BWR steam dryers. Recent moisture carryover measurements in BWRs have generated questions about steam dryer performance and the resulting impact on process computer core thermal power calculations. Data also indicates differences in steam dryer/steam separator performance between the various BWR models, with newer models reporting smaller carryover fractions. GE has indicated that the difference in dryer performance between older and newer model plants is likely due to evolutionary design improvements made to the steam dryers installed in the later designs. This has resulted in a carryover fraction of essentially zero for the newer design plants, which means the calculated core thermal power is approximately 0.082% lower than actual core thermal power. Although the steam carryover fraction had been measured during the startup test program in 1987, the primary purpose of this startup test was to demonstrate compliance with warranty requirements, and was not a verification of heat balance calculation input parameters. Actual plant testing data was not incorporated into the core thermal power calculation.
IV. Safety Analysis
The PNPP licensed power level of 3758 MWth is the analysis basis for the Cycle 9 core (the current operating cycle).
This power level was used for the initial conditions for the design basis accident and transient analyses performed.
Most design basis accident and transient analyses are performed at 102% of rated thermal power. As discussed in the GE report, "Impact of Steam Carryover Fraction on Process Computer Heat Balance Calculations, September 2001", most reactor thermal heat balances have an uncertainty value of 1.8%. The uncertainty value of 1.8% coupled with the moisture fraction carryover bias of 0.082% would still be bounded by the 102% rated thermal power assumed in various design basis accidents and transients analyses. Use of the original carryover fraction, while non-conservative, does not represent a safety issue, however. The change in core thermal power is an order of magnitude less than the precision of the Maximum Critical Power Ratio (MCPR) safety limit evaluation process. Further, the change in core thermal power is a factor of 18 less than the precision of the process computer core thermal power estimate. Applying the change as a carryover fraction bias of 0.1% in the core thermal power evaluation represents a small increase (less than 1%) in the probability that the core thermal power will exceed the nominal rated power by more than 2%. If reactor power would be in excess of 102% at the start of a transient or accident, then the consequences of the accident may exceed the results estimated by the analysis. Although the thermal power potentially exceeded the maximum power level specified in the operating license, it is not likely that thermal power ever exceeded analyzed limits (102%). The Updated Safety Analysis Report (USAR) Chapter 15 contains bounding analyses for power levels of 102%.
The recent PNPP power uprate project (License Amendment 112, issued May 3, 2000) resulted in a power uprate to 105% of original rated thermal power. The carryover fraction issue also existed prior to the power uprate. It is not expected that power uprate improved the moisture carryover fraction, but instead degraded it towards the assumed value of 0.1%. If the power uprate improved the moisture fraction carryover, however, this improvement would be bounded by assuming no carryover. With no assumed carryover, calculated thermal power would be 0.082% less than actual reactor thermal power. The power uprate analysis basis maintained the 102% analysis basis discussed above (i.e., 102% of 105% of original rated thermal power). This power uprate analysis demonstrated all relevant acceptance criteria were satisfied. It should be noted the 102% analysis basis was also applied to the pre-power uprate analysis.
The Anticipated Transient Without Scram (ATWS) is a non-design basis accident and as such its analysis is performed at rated conditions. The recent 105% power uprate ATWS analysis was reviewed. Based on the change resulting in uprating the reactor to 105%, the increase in reactor power of 0.082% would not cause any acceptance criteria to be exceeded.
Thermal power distribution limits are established within PNPP Technical Specifications and are identified within the Core Operating Limits Report. These ensure that the reactor fuel is operated within design limits for both normal power operations as well as transient situations. Review of reactor parameters during Cycle 9 indicated that there was still margin to actual thermal power distribution limits. By maintaining margin to real thermal power distribution limits despite the reactor thermal power calculation error, operation of the fuel stayed within fuel design limits, and did not exceed Technical Specification limits.
Reactor power is also monitored using Average Power Range Monitors (APRMs) [IG]. The surveillance requirements for these nuclear instruments use calculated thermal power. The bias introduced by the non-conservative carryover fraction should not have resulted in any APRM inoperability. Surveillance procedures which verify APRM operability ensure the APRM readings are within 2% of the indicated percent core thermal power, or adjustment of the APRM is required to restore operability. For APRM readings that differ from the indicated power level by 1% to 1.9%, the APRMs are operable but the surveillance instruction states that the APRMs should be adjusted, and that is the standard operating practice. Therefore, there are no periods of time where the carryover fraction bias should have resulted in APRM inoperability.
Based on the above, this event had no safety significance.
V. Similar Events A search of licensee event reports from PNPP over the past five years found one event where exceeding operating license thermal power limits was reported (LER 1999-007, and LER 1999-007-01). In that event, a modification to a database within a software code providing input to thermal power calculations set specific constants to zero, which impacted the calculation of core thermal power when feedwater temperature was less than 420 degrees Fahrenheit. As a result, a historical review of operating data identified that the plant had reached a maximum power level that exceeded licensed reactor thermal power limits. That event had no safety significance because the thermal power levels did not exceed limits in the accident analyses. The cause of that event was a weakness in the administrative controls in place for the review of the software revision. Although similar consequences occurred, the current event was not the result of a change to the software.
As discussed previously, the non-conservative carryover fraction affects reactor power calculations at other plants, and several plants have reported potential operating license violations to the NRC.
VI. Corrective Actions
Core thermal power was administratively restricted to 3754 MWth (a reduction of 4 MWth) until the value of the moisture fraction in the process computer was changed based on GE input and industry performance through a software change in accordance with plant procedures. Corresponding changes will also be made to the program used for manual heat balance calculations.
A design basis document will be generated for the reactor heat balance to identify source documents for each term in the heat balance equation.
This event has been documented in the PNPP corrective action program. The corrective actions will be tracked and implemented in accordance with processes and requirements of the corrective action program.
NOTE: Energy Industry Identification System Codes are identified in the text by square brackets (e.g., [XX]).