05000445/LER-2004-002
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications |
| 4452004002R00 - NRC Website | |
I. DESCRIPTION OF REPORTABLE EVENT
A. REPORTABLE EVENT CLASSIFICATION
Any operation or condition prohibited by the plant's Technical Specifications.
B. PLANT OPERATING CONDITIONS PRIOR TO THE EVENT
On April 2, 2004, Comanche Peak Steam Electric Station (CPSES) Unit 1 was in Mode 6 during the tenth refueling outage and Unit 2 was in Mode 1 operating at 99.2 percent power.
C. STATUS OF STRUCTURES, SYSTEMS, OR COMPONENTS THAT
WERE INOPERABLE AT THE START OF THE EVENT AND THAT
CONTRIBUTED TO THE EVENT
There were no inoperable structures, systems, or components that contributed to the event.
D. NARRATIVE SUMMARY OF THE EVENT, INCLUDING DATES AND
APPROXIMATE TIMES
Technical Specification (TS) 3.3.5 covers a number of undervoltage functions depicted in TS Table 3.3.5-1, including preferred and alternate offsite source bus undervoltage, 6.9 kv bus loss of voltage and degraded voltage, and 480 degraded and low grid undervoltage functions. SR 3.3.5.3 specifies the performance of a channel calibration on the Loss of Power (LOP) Diesel Generator (DG) Start Instrumentation listed in TS Table 3.3.5-1 every 18 months. SR 3.3.5.4 requires verification that the LOP DG start Engineering Safety Features (ESF) response times are within limits every 18 months on a staggered test basis. The allowable values for the undervoltage relay setpoints are contained in TS Table 3.3.5-1, and the response time values are contained in Technical Requirements Manual (TRM) table 13.3.5-1.
The TS definition for a channel calibration specifies in part "the CHANNEL CALIBRATION shall encompass all devices in the channel required for channel OPERABILITY." The Bases for SR 3.3.5.3 specifies in part "A CHANNEL CALIBRATION is performed every 18 months, or approximately at every refueling. CHANNEL CALIBRATION is a complete check of the instrument loop, including the sensor. The test verifies that the channel responds to a measured parameter within the necessary range and accuracy.
At 1500 hours0.0174 days <br />0.417 hours <br />0.00248 weeks <br />5.7075e-4 months <br /> on April 2, 2004, while reviewing proposed changes to the response times in the TRM, Regulatory Affairs personnel (utility, non-licensed) discovered that not all components of the undervoltage channels [EIIS:(EB)(CHA)] were being tested to meet channel calibration requirements at a frequency of 18 months as specified in SR 3.3.5.3. It was discovered that the Agastat timing relays [EIIS:(EB)(CHA)(RLY)] in these channels were only tested under the response time testing requirement (18 months Staggered Test Basis) meaning that each train was being tested approximately every 36 months.
Further investigation revealed that in 1994, a Technical Evaluation had been generated to answer a question which had been raised related to SR 3.3.5.3 and SR 3.3.5.4. Specifically, in 1994 Maintenance personnel (utility, non-licensed) questioned whether or not the calibration of just the undervoltage relays every 18 months and a response time test every 18 months on a staggered test basis (alternate trains) satisfied the TS requirements for calibration if the response time test is satisfactory and no actual adjustment of the Agastat relays are required.
Engineering's (utility, non-licensed) response via the Technical Evaluation stated that the undervoltage relays should be calibrated every 18 months and incorrectly indicated that the Agastat timing relays should be tested as one train per outage with a maximum of two refueling cycles between tests. It had previously been determined that satisfactory response time testing methodology also met the requirements for a channel calibration for Agastat relays.
As a result of this incorrect Technical Evaluation, the TS scheduling database was modified on January 1, 1995 to separate the undervoltage relay calibrations from the timing relay time test and also extended the Agastat timing relay test to once every other cycle or at a frequency of 36 months (one train per outage). As a result, since 1995 the Agastat timing relays have not been tested per SR 3.3.5.3 at the required 18 month interval and this constitutes a missed surveillance and a reportable condition prohibited by TS.
E. THE METHOD OF DISCOVERY OF EACH COMPONENT OR SYSTEM
FAILURE, OR PROCEDURAL OR PERSONNEL ERROR
While reviewing proposed changes to response times in the Technical Requirements Manual, Regulatory Affairs personnel (utility, non-licensed) discovered that TS Surveillance Requirement 3.3.5.3 had not been completed within the required frequency for all of the functions specified in TS Table 3.3.5-1.
II. COMPONENT OR SYSTEM FAILURES
A. FAILURE MODE, MECHANISM, AND EFFECTS OF EACH FAILED
COMPONENT
Not applicable — No component or system failures were identified during this event.
B. CAUSE OF EACH COMPONENT OR SYSTEM FAILURE
Not applicable — No component or system failures were identified during this event.
C. SYSTEMS OR SECONDARY FUNCTIONS THAT WERE AFFECTED BY
FAILURE OF COMPONENTS WITH MULTIPLE FUNCTIONS
Not applicable — No component or system failures were identified during this event.
D. FAILED COMPONENT INFORMATION
Not applicable — No component or system failures were identified during this event.
III. ANALYSIS OF THE EVENT
A. SAFETY SYSTEM RESPONSES THAT OCCURRED
Not applicable - no safety system responses occurred as a result of this event.
B. DURATION OF SAFETY SYSTEM TRAIN INOPERABILITY
Not applicable No safety system was rendered inoperable.
C. SAFETY CONSEQUENCES AND IMPLICATIONS
The seven affected functions on Unit 1 were verified successfully by performing the required TS channel calibrations during the recently completed tenth refueling outage. The seven affected functions on Unit 2 will be verified via TS channel calibrations prior to completion of the eighth refueling outage. As required by TS SR 3.0.3, an evaluation was performed that determined that the impact of these missed surveillances on plant risk is very small, thus extending the surveillance to the end of the current cycle for Unit 2 is not risk significant.
All of the affected channels that have been tested demonstrated that the channels would have performed their intended safety function, if required. This is consistent with the historical performance of this type of relay at CPSES, where less than one percent of the relay settings were found to be beyond the allowable value. There were no safety system functional failures associated with this event.
Based on the above, it is concluded that this event did not adversely impact the safe operation of CPSES or the health and safety of public.
IV. CAUSE OF THE EVENT
TXU Energy believes that the cause of the event was less than adequate review of a change to the preventive maintenance database frequency for these Agastat timing relays due to personnel errors in the review process and because of inadequate procedure referencing. As previously discussed, a Technical Evaluation was generated in 1994 to answer a question which had been raised related to SR 3.3.5.3 and SR 3.3.5.4. Personnel involved had less than adequate knowledge of the requirements of the procedure for TS questions and, therefore, the procedure governing review and approval of this type of change was not followed. In addition, the procedure governing Technical Evaluations did not prohibit, address, or reference to the correct procedure for questions regarding TS. As a result, personnel with a more detailed knowledge and understanding of the definition and scope of a "channel calibration" with respect to TS were not formally part of the review. Those involved in the review arrived at the wrong conclusion.
TXU Energy believes that the Technical Evaluation would have received an adequate review, reached the correct conclusion, and this event would not have occurred had the correct review process, as required by the procedure that was in effect at the time, been followed.
V. CORRECTIVE ACTIONS
The seven affected functions on Unit 1 were verified by successfully performing the required TS channel calibrations during the recently completed tenth refueling outage.
The seven affected functions on Unit 2 will be verified via TS channel calibrations prior to completion of the eighth refueling outage. Per TS SR 3.0.3, an evaluation was performed that determined the impact of these missed surveillances on plant risk is very small, thus extending the surveillance to the end of the current cycle for Unit 2 is not risk significant.
In accordance with the CPSES Corrective Action Program, the following actions will be taken:
1.A Lessons Learned describing this event will be issued to Regulatory Affairs, Engineering, Maintenance, and Operations personnel.
2. Training will consider adding this event to the TS training modules operating experience module or other appropriate locations for personnel needing a better understanding of "channel" and "channel calibration" as used within TS.
3.A review will be conducted to ensure that previous changes to the TS scheduling database did not extend the test frequencies for electrical components beyond the TS requirements.
4. The Corrective Action Program procedure will be revised to clarify that any question involving the meaning of TS requirements must be referred to Regulatory Affairs for resolution.
VI. PREVIOUS SIMILAR EVENTS
There has been one other missed surveillance event in the last two years (see LER 446/02-002). However, the details/causes are sufficiently different from the event described in this LER such that the previous corrective actions could not have prevented this event.