05000456/LER-2012-001

LER-2012-001, Two Main Steam Safety Valves Failed Pre-outage Setpoint Testing Due to Abnormal Spring Geometry

Docket Number
Event date:	04-12-2012
Report date:	06-11-2012
Reporting criterion:	10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications 10 CFR 50.73(a)(2)(vii), Common Cause Inoperability
4562012001R00 - NRC Website

A. Plant Operating Conditions Before the Event:

Event Date:�April 12, 2012 Unit: 1 MODE: 1� Reactor Power: 94 percent Unit 1 Reactor Coolant System [AB]:� Normal operating temperature and pressure

Description of Event:

No structures, systems or components were inoperable at the start of this event that contributed to the event.

On April 11, 2012, pre-outage testing was initiated for the main steam safety valves (MSSVs) [SB] for their setpoint verification per Technical Specification (TS) 3.7.1, Main Steam Safety Valves. During the testing, two of the MSSVs (1MS015D on April 11, 2012 and 1MS014D on April 12, 2012) failed to meet the as-found set pressure acceptance criteria of +/- three percent, and one MSSV (1MS016D on April 12, 2012) tested in tolerance within the evaluation range. When each valve (1MS015D and 1MS014D) was identified as outside the acceptance criteria, the appropriate Limiting Condition for Operation (LCO) was entered, the valve was returned to its operable status by returning the setpoints within the TS surveillance requirements, and the LCO was exited. For the 1MS016D valve, subsequent evaluation determined the valve met the as-found setpoint requirements.

Discrepancies found in TS surveillance tests are normally assumed to occur at the time of the test unless there is firm evidence, based on a review of relevant information that the discrepancy occurred earlier. However, on April 12, 2012, multiple valves (1MS015D and 1MS014D) were found to lift with setpoints outside of TS limits, which is an indication that the discrepancies may have arisen over a period of time. It is reasonable to assume the condition existed during plant operation in excess of TS LCO completion times. Therefore, on April 12, 2012, with the identification of the second valve outside of TS limits, this event was determined to be reportable in accordance with 10 CFR 50.73(a)(2)(i)(B), any operation or condition prohibited by the plants Technical Specifications.

Additionally, because the discrepancies affected multiple valves being inoperable, this event is reportable in accordance with 10 CFR 50.73(a)(2)(vii), any event where a single cause or condition caused two independent trains or channels to become inoperable in a single system designed to mitigate the consequences of an accident.

B. Cause of Event

The apparent cause of the 1MS015D and 1 MS014D valves failing their as found performance tests was determined to be abnormal spring geometry.

The 1MS014D and 1MS015D MSSVs contained their original springs. The 1MS015D spring had a much thinner coil cross section on the top coil than the bottom coil which is significant and can impact valve performance because of off-center loading of the spindle and disc. The 1 MS014D spring had a gap between the top coil of the spring that was wider than recommended based on a statistical evaluation of the Byron and Braidwood MSSV springs over a 10-year history. The spring as-measured top gap between the inactive and active coil was 0 150 inches which indicates abnormal spring squareness.

NRC FORM 36,..A Contributing causes were determined to be spindle wear and steam leakage across the disc seat area. The 1MS015D valve spindle showed significant vibration fretting damage from main steam system flow resonance.

Additionally, the Inconel disc showed eight separate leak paths across the surface, which lowers the lift setpoint because it increases the huddling chamber pressure which provides lifting force during valve actuation. The 1MS014D valve Inconel disc showed two locations of significant steam leakage across the seat area in two areas 180 degrees apart which can lower the lift setpoint.

Safety Consequences:

There were no actual safety consequences impacting plant or public safety as a result of this event. This event captures a setpoint discrepancy, and not an actual demand for the 1MS015D and 1MS014D valves to lift.

From the surveillance testing:

• 1MS015D - Lifted low at 1143.79 psig (-5.08 c./0)

• 1MS014D - Lifted low at 1179.64 psig (-3.31 %) The MSSVs are credited in the UFSAR Chapter 15 analyses for overpressure protection and small break loss of coolant accidents (SBLOCA). For the overpressure protection cases, the MSSVs are modeled with a +4 percent tolerance, and a +5 percent tolerance is used for SBLOCA — the analyses assume the MSSVs lift at pressures higher than the nominal setpoint. Lower MSSV lift pressure will be beneficial (that is gain margin to the acceptance criteria) for overpressure and the SBLOCA analyses. Since both the 1MS015D and 1MS014D lifted below their nominal setpoints, there is no impact on the overpressure and SBLOCA events.

The MSSVs are also modeled in the UFSAR Chapter 15 analyses as a release path for radiological releases.

From this event, the most limiting as-found setpoint was the 1MS015D valve which lifted at 1143.79 psig. Adding a bounding instrument uncertainty of -6.2 psig, the resultant lift pressure is 1137.59 psig (1143.79 — 6.2 = 1137.59 psig). For the steam generator tube rupture (SGTR) event, the steam generator power operated relief valves (PORVs) as well as the MSSVs are modeled in this analysis. The majority of the release from this event is through the steam generator PORV, which is assumed to be stuck open for 20 minutes. A depressurization occurs as the steam generator PORV becomes stuck open. The steam generator PORV is modeled to open at a pressure of 1099.7 psia. Since this is below the 1MS015D as-found lift pressure, the MSSV will not lift and there is no impact on this accident.

Dose calculations for releases other than SGTRs assume MSSVs as a release path. The current calculation assumes all 20 MSSVs remain open at a pressure of 1022.2 psig. The MSSV test results for 1MS015D show a lift pressure of 1137.59 (1143.79 — 6.2) psig. Assuming 10 % blowdown, this one valve would stay open at a pressure of 1137.59 X 0.9 = 1023.83 psig. This value is higher than the value of 1022.2 psig assumed in the analysis, therefore, there is no impact on this accident.

Based on the discussion above, the safety function of the MSSVs was not lost. Therefore, there was no safety system functional failure due to this event.

Corrective Actions:

Corrective actions included refurbishing the 1MS015D and 1MS014D valves (this action is complete).

F. Previous Occurrences:

No previous, similar Licensee Event Reports were identified at the Braidwood Station:

3. Component Failure Data:

Manufacturer Nomenclature Model Mfg. Part Number Dresser Main Steam Safety Relief Valves 3707R N/A