05000271/LER-2001-004
| Vermont Yankee Nuclear Power Station (Vy) | |
| Event date: | 08-21-2001 |
|---|---|
| Report date: | 10-19-2001 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications |
| 2712001004R00 - NRC Website | |
DESCRIPTION:
On 08/21/01, with the reactor at 100% power, it was discovered by System Engineering that during the period of 08/17-08/21/01, Core Thermal Power (CTP) as calculated by the process computer, was non-conservatively low by 3.5 Megawatts Thermal (MWth). This calculation (used to determine reactor power), allowed Vermont Yankee to unknowingly exceed its licensed limit of 1593 MWth by a maximum of 3.5 MWth. This event is reportable under 10CFR50.73(a)(2)(i)(B) as a condition prohibited by Vermont Yankee's Technical Specifications. Although the maximum power level is stated in the Operating License, this has typically been interpreted to include both the License and the Technical Specifications by the industry.
A calibration of the feedwater nozzles was performed on 08/21/01 using the Crossflow Ultrasonic Flow Meter, after the control rod pattern exchange down power of 08/16-08/17/01 to 65%. The feedwater nozzles, FE-6- 11A&B were found out of calibration, indicating a 0.22% lower than actual flow. This signal is used as an input to the Process Plant Computer (PPC) CTP calculation, which caused the calorimetric calculation to indicate 0.22% lower than actual. Therefore, reactor power as adjusted by plant operators, was higher than the licensed power level of 1593 MWth and calculated power of 1592.56 MWth by 3.5 MWth.
Following the discovery of the error in the CTP calculation on 08/21/01 at 11:30 AM, reactor power was decreased by 4 MWth at 11:50 AM. New calibration numbers for the nozzles were calculated by the guidance provided within OP2001 by System Engineering and placed in the computer by Reactor Engineering allowing the return to 100% CTP by 12:33 PM.
BACKGROUND:
Corrosion product fouling of the feedwater flow-measuring element has been recognized as an industry problem since the 1970's. The fouling preferentially occurs on the necked (narrowed) down part of the flow nozzle. The additional constriction has the effect of indicating higher than actual feedwater flow and hence a calculated CTP that is lower than indicated.
Vermont Yankee installed a Crossflow Ultrasonic Flow Meter in May 1999 to manage the fouling effects by allowing online calibration of the installed flow meter. Evaluations performed during Crossflow testing and Cycle 22 operation, indicated that the correction is valid for all routine power maneuvers, such as control rod pattern exchanges. Past practice has been to perform the calibration monthly and supply Reactor Engineering the results to input into the reactor heat balance.
Prior to Crossflow, the cycle based transmitter calibration was based on flow data obtained from nozzle testing at Alden labs in 1970.
CAUSE OF THE EVENT:
The cause of this event is mainly attributed to fouling on the feedwater nozzles which changed during a plant down power to 65% CTP. Contributing factors include the first application of noble metal chemistry during RF022 (May 2001) which could have altered the oxide layer inside the feedwater nozzles. In addition, previous operational history indicates that nozzle fouling was stable for any percentage down power that did not take the unit off line.
ANALYSIS OF THE EVENT:
Prior to the control rod pattern exchange down power, the feedwater nozzles were reading within calibration, based upon a review of the data performed by System Engineering. Power ascension from the 08/16-08/17/01 down power and previous down power maneuvers relied on the calculated CTP to be accurate. Alternate power indications are not as precise an indicator of CTP. Additionally, interviews with plant and peer plant personnel indicate changes in chemistry, noble metals, zinc, oxygen and hydrogen can alter feedwater-fouling characteristics, thereby affecting flow nozzle measurements.
A review of the Monthly Core Performance Log indicates the highest thermal limits achieved during this event were 0.883 for MFLCPR, 0.960 for MFLPD and 0.890 for MAPRAT. For small power changes, a change in power will cause a similar percentage change in thermal limits. A 0.22% rated error in the calculation of core thermal power could not have caused thermal limits to exceed 1.000. The 100.22% power level achieved during this event was fully bounded by plant analyses (Fuel Reload and Transient Analyses). The Fuel Reload Analysis assumes a CTP level in excess of 102% for all FSAR accident analysis and feedwater flow uncertainty of 1.8% for Transient Analysis. The plant's safety significant protection systems were unaffected and the limiting condition achieved during this event was fully bounded by plant analyses and accident analyses.
CORRECTIVE ACTIONS:
1. The operating crew reduced reactor power by 4 MWth and the nozzle calibration was implemented (08/21/01).
2. Daily calibration checks of feedwater nozzles with Crossflow were implemented for two weeks following the event to ensure that no other issues or equipment were contributing factors to this event.
3. The Crossflow Calibration check frequency was increased to weekly.
4. Interim Guidance for power ascension was provided to operators.
5. Procedures and interim guidance will be revised to address improved flow determination capability resulting from Crossflow Calibration and the impact of chemical addition upon plant systems.
ADDITIONAL INFORMATION:
In the past ten years, similar events have been reported as follows:
LER No. 1 Date 1 Title 97-12 1 10/02/97 1 Excess Core Thermal Power 94-12 1 10/24/94 Operated Above License Limit for Core Thermal Power