05000446/LER-2024-001-01, Unit Two, Recurring Reactor Scrams Associated with Main Feedwater System Modifications
| ML24274A331 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 09/30/2024 |
| From: | Sewell S Vistra Operations Company |
| To: | Office of Nuclear Reactor Regulation, Document Control Desk |
| References | |
| CP-202400367, TXX-24070 LER 2024-001-01 | |
| Download: ML24274A331 (1) | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv)(A), System Actuation |
| 4462024001R01 - NRC Website | |
text
COMANCHE PEAK NUCLEAR POWER PLANT CP-202400367 TXX-24070 September 30, 2024 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Subject:
Comanche Peak Nuclear Power Plant (CPNPP)
Docket No. 50-446 Steven K. Sewell Nuclear Site Vice President Vistra Operations Company LLC P.O. Box 1002 6322 North FM 56 Glen Rose, TX 76043 Office: 254.897.6565 Ref 1 0CFR50. 73 Two Recurring Reactor Scrams associated with Main Feedwater System Modifications Licensee Event Report 2-24-001-01 Dear Sir or Madam:
Attached is Licensee Event Report (LER) 2-24-001-01, "Two Recurring Reactor Scrams associated with Main Feedwater System Modifications" for Comanche Peak Nuclear Power Plant (CPNPP) Unit 2.
This Communication contains no new commitments regarding CPNPP Units 1 or 2.
Should you have any questions, please contact Kassie Mandrell at (254) 897-6987 or Kassie.Mandrell@vistracorp.com.
Sincerely, Steven Sewell (Sep 30, 202416:04 CDT)
Steven K. Sewell Attachment:
Abstract
Unit 2 (U2) experienced subsequent reactor trips on 03/12/2024 and 03/17/2024 due to Main Feed Pump (MFP) speed reductions and failure to maintain steam generator (SG) level. The first event occurred on March 12, 2024 at 0816 after alternating the 2A MFP servo valve filters. When the standby filter was placed in service, the 2A MFP experienced speed reductions and lost forward flow. The 2B MFP was unable to effectively recover flow following a manual run back to 700 megawatts electric (MWe) and resulted in an automatic reactor trip due to steam generator (SG) 2-03 Lo-Lo water level. Following the first event, the MFP Turbine Digital Control System was modified to correct operating current. Unit 2 returned to 100% power on March 17, 2024 at 1330. At 1515, the 2B MFP tripped and the 2A MFP was unable to maintain level during the automatic run back to 700 MWe. A manual trip was initiated in anticipation of an automatic trip from SG levels approaching the Lo-Lo trip setpoint. Cause analysis determined that weakness in technical conscience allowed design errors to go unrecognized during the initial and subsequent modification review and site acceptance processes for the MFP system, resulting in consequential events. (All times are in Central Daylight Time)
I. DESCRIPTION OF REPORTABLE EVENT A. REPORTABLE EVENT CLASSIFICATION 050 052
- 2. DOCKET NUMBER
- 3. LER NUMBER I
446 Cl NUMBER NO.
I YEAR SEQUENTIAL REV
~-I 001 1-~
Both events are reportable under 10CFR50.73(a)(2)(iv)(A), "Any event or condition that resulted in manual or automatic actuation of any of the systems listed in paragraph (a)(2)(iv)(B) of this section." The systems that actuated, in both events, were the Unit 2 Reactor Protection System (RPS) and the Unit 2 Auxiliary Feedwater System (AFW).
B. PLANT CONDITION PRIOR TO EVENT Event 1: At 0816 on March 12, 2024, Comanche Peak Nuclear Power Plant (CPNPP) Unit 2 was in MODE 1, operating at 100% power.
Event 2: At 1515 on March 17, 2024, Comanche Peak Nuclear Power Plant (CPNPP) Unit 2 was in MODE 1, operating at 100% power.
C. STATUS OF STRUCTURES, SYSTEMS, OR COMPONENTS THAT WERE INOPERABLE AT THE START OF THE EVENT AND CONTRIBUTED TO THE EVENT There were no structures, systems, or components that were inoperable prior to either event which contributed to the events.
D. NARRATIVE SUMMARY OF THE EVENT, INCLUDING DATES AND APPROXIMATE TIME On 03/08/2024, Operations identified an upward trend in the 2A Main Feed Pump (MFP) servo valve [EIIS: (JK)(PSV)] filter differential pressure. On 03/12/2024, Operators were performing a servo valve oil filter swap for 2A MFP. After the operators placed the standby filter in service at 0814, the 2A MFP began to experience speed reductions and feed flow output degraded but did not automatically shutdown the pump. A manual runback to 700 Megawatts electric (MWe) was initiated at 0815:55. The 28 MFP ramped up speed in response to the runback, but Steam Generator (SG) level continued to decrease. Operators initiated actions for a manual reactor trip in accordance with ODA-102, "Conduct of Operations".
Prior to completion of operator trend verifications, the reactor automatically tripped on 2-03 SG LO-LO level at 0816: 15.
Operations personnel promptly stabilized the plant following the reactor trip.
The 2A MFP speed reduction was attributed to foreign material entering the servo during the servo filter swap. The as found inspection of the servo valve did not indicate significant debris or clogging. The inability to sustain power generation following a main turbine run back due to the MFP trip was determined to be caused by insufficient current to the servo valve
(+/-30 milliampere (mA) vice 240mA) resulting in the valve not fully stroking. This was the result of a design error introduced during the installation of the vendor Turbine Digital Control System [EIIS: (JK)(SCO)]. The reduced valve stroke inhibited the ability for the Servo Valve to pass oil contaminants, resulting in degraded MFP turbine speed control. A new vendor supplied design change was approved to install current amplifiers [EIIS: (JK)(AMP)] into the MFP speed control circuitry to produce an output current of 240mA. During the amplifier modification initial startup testing several system diagnostic alarms were received. The diagnostic alarms cleared after the ground wires were tightened and the plant recovery continued.
Unit 2 commenced a startup on 03/14/24 at 2015. During the power ascension, the system diagnostic alarms were received on the 28 MFP. Power was reduced, and 28 MFP was taken off-line. The alarms were determined to be caused by additional ground connections that needed tightening. The alarms cleared after tightening ground wires. U2 MFP B was placed back in service, and the plant ascension continued. Unit 2 was returned to 100% power on March 17, 2024 at 1330.
At 1410, operators received additional power supply fault alarms for the 28 MFP servo. At 1515, the 2B MFP tripped and the 2A MFP was unable to maintain level during the automatic runback to 700 MWe. A manual trip was initiated at 1515:45 in anticipation of an automatic trip from SG levels approaching the trip setpoint. SG 2-03 water level reached Lo-Lo Trip setpoint at 1515:46.
I
- 2. DOCKET NUMBER
- 3. LER NUMBER IB SEQUENTIAL REV NUMBER NO.
446
- - I I-~
001 E. THE METHOD OF DISCOVERY OF EACH COMPONENT OR SYSTEM FAILURE, OR PROCEDURAL PERSONNEL ERROR For both events, the indication of speed reductions of the MFP and loss of feedwater flow resulted in Unit 2 reactor trips.
II. COMPONENT OR SYSTEM FAILURES A. CAUSE OF EACH COMPONENT OR SYSTEM FAILURE The cause of the event on 03/12/2024 was a design error on the installation of the MFP Turbine Digital Control System.
The Turbine Digital Control System modification installed on both Unit 1 and Unit 2 included a significant design error in which the control current to servo valve was changed from +/- 240mA in the original design to +/- 30mA. A current amplifier included by the original manufacturer in the previous design was not included in the design as part of the Turbine Digital Control System modification. This amplifier converts the output current to the required +/- 240 mA for proper servo operation. This change resulted in limiting the open travel of the valve to only 10% of design. This resulted in the servo valve not fully stroking and inhibiting the ability of the valve to pass oil contaminants and recover from transient conditions.
After the second event on 03/17/2024, it was identified that the amplifier modification on the Turbine Digital Control System from the 03/12/2024 event resulted in significant signal noise and introduced a failure mode where one fault could affect both amplifiers on a single pump. An inadequate design installed a common ground scheme that inadvertently created a single point vulnerability and generated signal interference that resulted in a loss of amplifier output signals to the MFP 28 servo.
In both events, the alternate MFP was unable to compensate for flow following a runback to 700 MWe. This failure was caused by ineffective tuning/testing of the original Turbine Digital Control System installation in combination with the restricted servo performance. Dynamic tuning was not performed to determine appropriate settings during transient conditions that would enable system stabilization and reduce system speed reductions. Evaluation of the MFP runback performance also determined that recovery depended on initial plant conditions.
B. FAILURE MODE, MECHANISM, AND EFFECTS OF EACH FAILED COMPONENT These events were first identified by operations as speed changes on MFP. The MFP later tripped when speed could not be adequately controlled. The alternate MFP could not maintain flow following a runback to 700 MWe and the reactor trip occurred as a result of decreasing steam generator level.
C. SYSTEMS OR SECONDARY FUNCTIONS THAT WERE AFFECTED BY FAILURE OF COMPONENTS WITH MULTIPLE FUNCTIONS These events did not involve systems or secondary functions which were affected by the MFP Turbine Digital Control System failure.
D. FAILED COMPONENT INFORMATION 050 052
- 2. DOCKET NUMBER
- 3. LER NUMBER I
446 Cl NUMBER NO.
I YEAR SEQUENTIAL REV
~-I 001 1-~
Each main feed pump is driven by a dual admission low pressure and high pressure turbine using steam control valves.
The Main Feed Pumps operate at variable speed to follow load conditions from the demand signals. This control system provides demand signals to position the electrohydraulic controls to maintain the feed pump at the necessary speed while providing monitoring and equipment protection to the feed pump. The digital speed control system is designed to be operated primarily in automatic by a varied signal or manually by an operator from a remote or local control station.
Ill. ANALYSIS OF THE EVENT A. SAFETY SYSTEM RESPONSES THAT OCCURRED For each event, the Reactor Protection System and Auxiliary Feedwater System started as designed due to steam generator water level exceeding the trip setpoint.
B. DURATION OF SAFETY SYSTEM TRAIN INOPERABILITY These events did not involve the inoperability of any safety related component or system.
C. SAFETY CONSEQUENCES AND IMPLICATIONS OF THE EVENT A turbine trip is an ANS Condition II event (Faults of Moderate Frequency). In both events, when the reactor tripped the Auxiliary Feedwater System responded by providing feedwater to the steam generators. The reactor trip on lo-lo water level in any steam generator provides the necessary protection against a loss of normal feedwater. During both reactor trips, the Auxiliary Feedwater System started automatically as expected and all plant safety systems responded as designed during the resultant transient. These events had no impact on nuclear safety, reactor safety, radiological safety, environmental safety or the safety of the public. These events have been evaluated as not meeting the definition of a safety system functional failure per 10 CFR 50. 73(a)(2)(v).
IV. CAUSE OF THE EVENT Cause Analysis determined that design errors went unrecognized during the modification review and site acceptance process for both the Turbine Digital Control System and Amplifier Modifications, resulting in consequential events. The Turbine Digital Control System modification lacked detail for the servo valve operating current and the amplifier modification missed a design flaw that allowed one fault to affect both amplifiers. The investigation identified an over reliance on vendor expertise in supplying critical design requirements. Additionally, the operational acceptance testing for the initial or subsequent modification did not reveal design errors or prove functionality. Contributing to this, actions to evaluate and address MFP performance were ineffective and untimely. Analysis of unexpected servo valve performance and plant response did not address all concerns or anomalies.
V. CORRECTIVE ACTIONS
Completed corrective actions
- Replace 2A/2B MFP Servo Valve 050 052
- 2. DOCKET NUMBER
- 3. LER NUMBER I
446 Cl NUMBER NO.
I YEAR SEQUENTIAL REV
~-I 001 1-~
- Correct the Turbine Digital Control System current for servo valve operation to +/- 240mA.
- Revise the amplifier design modification to eliminate common grounding and implement changes.
Planned corrective actions
- Develop and implement a testing/tuning plan to validate MFW Pump system response to various operational conditions.
- Formalize the disposition of issues and comments for design changes affecting plant reliability associated systems and those design changes that receive a third-party review in the appropriate procedure.
- Complete training for Engineering and Station Management based on a case study for these events and missed opportunities based upon these events.
VI. PREVIOUS SIMILAR EVENTS LER-23-001-00 and associated supplemental LER 23-001-01, Automatic Trip Due to Lo-Lo Steam Generator Level. On June 16, 2023, unit 1 was operating at 100% power. At 1822, the 1 B Main Feedwater Pump (MFP) began indicating oscillating speed. The oscillations increased in severity and became unresponsive to manual control input. Due to the loss of control, the 1 B MFP was manually tripped as it approached the overspeed setpoints. Following the manual trip of the 1 B MFP the Unit responded with an automatic runback to ?00MWe. Following the runback, the combined effects of the entire transient resulted in an automatic reactor trip due to Steam Generator #4 Lo-Lo water level.
The direct cause of this event was a degraded servo on the 1 B MFP. Vendor analysis of the component did not identify a specific failure mechanism and the cause is inconclusive. Based on engineering evaluation, the most probable cause is particulate accumulation in the servo valve due to historical water intrusion events into the 1 B lube oil system. Servo valves and filters were replaced. An Adverse Condition Monitoring Plan (ACMP) was initiated for MFPs on both units. A Modification intended to address seal water leakage, which had historically been the primary cause of water intrusion was elevated in priority.
Actions taken as a result of the first event, March 12, 2024, were not effective in preventing a subsequent event on March 17, 2024. Design errors in the servo valve modification and effects on performance were not understood.
Therefore, this event was included in the Cause Analysis for MFP reliability along with the 2023 event described in this LER. Corrective actions for the U2 MFP trips were completed in May of 2024 for Unit 1 MFPs. Page 5
of 5