Event Notification Report for August 22, 2012

TECH SPEC REQUIRED SHUTDOWN DUE TO REACTOR COOLANT SYSTEM PRESSURE BOUNDARY LEAK

"On August 22, 2012, it was determined, based on a remote video camera inspection inside the secondary shield wall, that a primary system leak at a Pressurizer Heater bundle exists and is not isolable. Technical Specification (TS) Limiting Condition for Operation (LCO) 3.1.6.4 does not allow any leakage through a nonisolable RCS strength boundary. The discovery of leakage from the Pressurizer Heater bundle is considered a degradation of a principal safety barrier. This condition does not represent a reduction in the public health and safety.

"This is reportable as a 4-hour ENS notification under 10CFR50.72(b)(2)(i) due to the initiation of a nuclear plant shutdown required by the plant's Technical Specifications.

"The licensee notified the NRC Resident Inspector."

The licensee expects to also notify state and local government agencies of the shutdown. The current leak rate, as calculated by the plant process computer, is approximately .2 gallons per minute.

AUTOMATIC REACTOR TRIP DUE TO HIGH REACTOR COOLANT SYSTEM PRESSURE

"On August 22, 2012, during a planned load reduction, Three Mile Island Unit 1 Reactor automatically tripped at 0801 EDT due to high reactor coolant system pressure as a result of a main feedwater transient. The cause of the main feedwater transient is still under investigation. The Emergency Feedwater System (EFW) actuated at 0801 EDT.

"The electrical grid is stable and Unit 1 is being supplied by offsite power. All control rods fully inserted. Decay heat is being removed via the Emergency Feedwater System (EFW) flow to both Steam Generators that are exhausting via the normal main condenser cooling loop under manual control. Preliminary evaluation indicates all plant systems functioned normally following the reactor trip and EFW actuation except for manual operation of Turbine Bypass Valve control due to failure of the automatic control function. Three Mile Island Unit 1 remains stable in Hot Shutdown mode while conducting the post trip review. No radioactive releases were experienced as a result of this event.

"This event is reportable under 10 CFR 50.72(b)(2)(iv)(B), Reactor Protection System (RPS) actuation, and under 10 CFR 50.72(b)(2)(xi), due to an information release to local officials, both are four (4) hour reports. This event is also reportable under 10 CFR 50.72(b)(3)(iv)(A) and 10 CFR 50.72(b)(3)(iv)(B), due to a valid actuation of the Emergency Feedwater System.

"The licensee notified the NRC Resident Inspector." Licensee notified the State, local and other Government agencies.

PART 21 - DEFECT DUE TO CHANGE IN MANUFACTURING PROCESS CAUSING RELAY FAILURE IN SAFETY RELATED SYSTEMS

The following is summary of the information received from the licensee:

"The basic component is an Eaton-Cutler Hammer Type ARD660UR DC relay that is commercially dedicated by Westinghouse for use in safety related systems at Palo Verde Units 1, 2 and 3. Except for the Palo Verde plants, Westinghouse is not aware of any other plant that uses this relay as a safety-related component in normally energized applications.

"The relay contacts failed to change state when required to do so during postulated events and/or surveillance testing. Westinghouse has identified the kick-out spring as a possible contributing factor for the relay failure due to stress corrosion cracking. Other anomalies such as relay core barrel tolerance and potential material deficiencies are currently under review. Based upon testing at APS, the relay failure rate is low and non-reproducible. This indicates that a combination of factors could be resulting in the failures with different causes for each failure. Results of testing do not identify a common cause for the failures. For ARD660UR relays used in normally de-energized applications, the kick-out spring will be compressed for only a short period of time and exposure to additional heat generated by intermittent coil energization will be minimal. For relays in normally de-energized applications, it is not expected that the force provided by the kick-out spring will decrease significantly over time and the contacts will change position when the relay coil is de-energized. Westinghouse has not received any reports to date of relay contacts failing to properly change position when the relay goes from a de-energized to an energized state. Because of the kick-out spring's limited exposure to compression and heat generated by the relay coil, it is expected that the springs will perform as intended in normally de-energized applications for the qualified life of the relay.

"Identification of the firm constructing the facility or supplying the basic component which fails to comply or contain a defect.

"Westinghouse Electric Company
"1000 Westinghouse Drive
"Cranberry Township, Pennsylvania 16066"

* * * UPDATE AT 1445 EDT ON 04/08/13 FROM JAMES A. GRESHAM TO S. SANDIN * * *

The following update was received via fax and is summarized below:

"During the investigation into the cause of the ARD660UR relay sticking, many physical and performance aspects and components of the relay were analyzed, as well as the entire manufacturing process. This investigation uncovered several issues that contributed, or could contribute, to the failure of the relay to release when de-energized.

"Based on analysis by Westinghouse, with support from Eaton Corporation, it was determined that the primary cause of the relay failure was a change in the manufacturing process in the plastics molding operation of this relay. This manufacturing change caused the moving cores to adhere to the inner diameter of the relay coil spool when a relay was continuously energized during testing by Westinghouse for longer than 21 days. This change in the manufacturing process began in May 2008 and continued until it was terminated in December 2012. Relay coils manufactured during this time may develop an adhesive like residue in the relay coil spools when energized for an extended period of time. This residue was found on the moving cores of relays which stuck during testing at a Westinghouse facility and relays returned from the customer. This residue was determined to be the primary cause of the relay issue.

"Westinghouse shipped Palo Verde a total of 374 potentially affected Eaton-Cutler Hammer Type ARD660UR DC relays.

"As a result of the investigation, Westinghouse recommended several manufacturing process improvements that are designed to prevent the reoccurrence of the issue. Eaton has agreed to implement these improvements prior to restarting the manufacture of these relays. Westinghouse is revising its commercial grade dedication process for these relays. This action ensures that the commercial grade dedication criteria include replacing the relay kick-out spring in each relay and verifies other relay enhancements have been implemented before future relays are shipped to the customer as safety related components.

"Westinghouse recommends that each plant review the application requirements of each affected relay. If an ARD660UR relay is used in a normally energized application or is required to change state after being energized for at least 21 consecutive days and was manufactured between May 1, 2008 and December 31, 2012, Westinghouse recommends replacing the relay at the next convenient opportunity.

"If an ARD relay manufactured during the May 1, 2008 and December 31, 2012 time frame is successfully, periodically cycle tested, this relay may be less susceptible to sticking."

Notified R4DO (Deese) and NRR Part 21 Group via email.

AGREEMENT STATE REPORT - POTENTIAL MEDICAL OVERDOSE

The following report was received via e-mail:

"On May 26, 2013 during a transfer of electronic treatment planning records to a new system, the University of Minnesota (license number 1049-211-27) discovered a medical event that occurred on August 20-22, 2012 at the university of Minnesota Medical Center in Minneapolis with the Nucletron HDR. The licensee reported that dosimetry staff were testing the transfer of information from previously treated patients into a brachytherapy check program, and it was discovered that in this particular case the source position data was entered into the HDR planning system incorrectly. The licensee is calculating the exact doses delivered and it appears as though the dose to unintended regions by greater than 50% for several areas. The Minnesota Department of Health was notified of the potential event on May 27, 2013. A final report will be submitted when the report is received from the licensee."

A Medical Event may indicate potential problems in a medical facility's use of radioactive materials. It does not necessarily result in harm to the patient.

MISSING FLOOD SEAL

"During flooding walkdowns being performed on August 22, 2012, with the unit at 100 percent power, South Texas Project Unit 2 discovered the potential for water intrusion into the 10' Elevation Electrical Auxiliary Building (EAB) via a 2-inch underground conduit that was found to be missing its flood seal. It has been determined that the missing flood seal compromised the external flood design controls for the EAB. If flooding of the 10 [foot] EAB were to occur as a result of the missing flood seal, the operability of the Train A Engineered Safety Features (ESF) switchgear and the ESF Sequencers for all three Standby Diesel Generators could have been affected. Additionally, the Qualified Display Parameter System process cabinets (which control Auxiliary Feedwater flow and Steam Generator PORVs) and the Auxiliary Shutdown Panel are also located on the 10' Elevation.

"Repairs have been made and the 2-inch conduit is sealed.

"The event is being reported under 10 CFR 50.72(b)(3)(ii)(B) for Unit 2 being in an unanalyzed condition that significantly degraded plant safety, and under 10 CFR 50.72(b)(3)(v) as an event or condition that could have prevented the fulfillment of a safety function.

"The NRC Resident Inspector has been notified."

* * * RETRACTION FROM JAMES MORRIS TO JOHN KNOKE AT 1658 EDT ON 09/20/12 * * *

"The purpose of this call is to retract the notification made on 09/05/2012, Event Number 48277. Further analysis indicates that water intrusion resulting from the missing 2-inch conduit seal would not have been sufficient to affect the operability of the equipment located on the 10-foot elevation of the Unit 2 Electrical Auxiliary Building. It has been determined that the maximum water depth would not have exceeded 2 inches in depth and all safety related equipment on the 10-foot elevation is greater than 2 inches above the floor, therefore there would be no impact to any safety-related equipment. Accordingly, this event notification is being retracted."

The licensee will notify the NRC Resident Inspector. Notified the R4DO (Geoffrey Miller).