05000250/LER-2003-004
| Turkey Point Unit 3 | |
| Event date: | 02-27-2003 |
|---|---|
| Report date: | 04-25-2003 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications |
| 2502003004R00 - NRC Website | |
DOCKET
Description of the Event
On February 27, 2003, Turkey Point Unit 3 was in Mode 1 and holding at approximately 60 percent reactor power for testing of the Main Steam Safety Valves (MSSVs) [SB:RV] during the downpower for the Turkey Point Unit 3 Cycle 20 refueling outage.
At 9:28 AM, MSSV RV-3-1410 was declared Out of Service (00S) when its tested opening pressure was 6.2% above the required setpoint. RV-3-1410 is configured on the 3C Steam Generator (SG). Since the Technical Specification (TS) allowable lift setting is 44-3%, the unit entered the Action Statement for TS 3.7.1.1.b, which requires either valve restoration to operable status or a unit reduction in power to within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. RV-3-1410 was subsequently tested twice to within 4-1% of its required setpoint.
The cause for the failure was determined to be micro-bonding of the nozzle and disc due to differential thermal expansion between the contact surfaces of the nozzle and the disc. Due to the planned unit shutdown scheduled for 2/28/03, RV-3-1410 would not be subjected to a subsequent heatup and the set pressure was satisfactorily demonstrated to be within the limits of TS 3.7.1.1. The valve was returned to service at 12:10 PM and the unit exited from the TS Action Statement.
At 2:38 PM, RV-3-1402 was declared OOS when its tested opening pressure was 4.8% above the required setpoint. RV-3-1402 is configured on the 3A SG. The unit again entered the Action Statement for TS 3.7.1.1.b. RV-3-1402 was subsequently tested twice to within 4-1% of its required setpoint. The cause for the valve failure was again determined to be due to nozzle and disc micro-bonding. RV-3-1402 was returned to service at 5:10 PM and the unit exited from the TS Action Statement.
At 4:31 PM, RV-3-1406 was declared OOS when it tested at 8.2% above the required setpoint. RV-3-1406 is configured on the 3B SG. RV-3-1406 was subsequently adjusted and retested to open within 4-1% of its required setpoint however, the cause for the failure was not determined at that time and the valve remained OOS. Unit reactor power was reduced to 3.7.1.1.b.
At 10:20 PM, RV-3-1411 was declared OOS when its tested opening pressure was 7.4% above the required setpoint. RV-3-1411 is configured on the 3C SG. RV-3-1411 was declared OOS, and subsequently satisfactorily retested. The cause for the valve failure was again determined to be due to nozzle and disc micro-bonding. The unit had already complied with the applicable Action Statement of TS 3.7.1.1.b due to failure of RV-3-1406 and reactor power remained at shutdown on 3/1/03.
The findings of the MSSVs failing setpoint testing are listed in Table 1.
Table 1 As-Found Lift Settings Valve Number Design Setpoint As-Found Setpoint As-Found Percentage RV-3-1402 1115 psig 1169.0 psig 104.84 % RV-3-1406 1100 psig 1190.6 psig 108.24 % RV-3-1410 1085 psig 1152.4 psig 106.21 % RV-3-1411 1100 psig 1181.4 psig 107.40 %
DOCKET
Four safety valves are installed on each of the unit's three main steam lines. The safety valves protect the steam generator and main steam piping from overpressure conditions and serve as a heat sink for the reactor coolant system if the main condenser is unavailable and the atmosphere steam dump valves cannot relieve pressure during a reactor trip or a secondary transient.
The Turkey Point MSSV Inservice Testing Program conforms to the ASME Section XI-1989 requirements, invoking ASME/ANSI OM-1987, Part 1. Paragraph 1.3.3.1 (e) of Part 1 specifies that:
"Any valve exceeding its stamped set pressure by 3% or greater shall be repaired or replaced, the cause of the failure shall be determined and corrected, and the valve shall successfully pass a retest before it is returned to service.
Cause of the Event
RV-3-1402/1410/1411 Inspection of valve internals confirmed the cause of the high lift setpoints for RV- 3-1402, RV-3-1410, and RV-3-1411 to be micro-bonding of the nozzle and disc. Set pressure test results that exhibit a large reduction following the as-found test (typically 30 psig) are symptomatic of this phenomenon. Micro-bonding of the nozzle and disc can occur when the harder disc (410 SS) causes microscopic galling of the softer nozzle (316 SS) during heatup, because of the differential thermal expansion between the contact surfaces. These small gall beads cause the disc and nozzle to fuse to a limited degree. Under these conditions, the first set pressure lift off is often significantly higher than subsequent lifts. The high initial setpoint pressures, followed by subsequent acceptable set pressure tests, are consistent with this phenomenon.
The traditional corrective action for micro-bonding has been to resurface the nozzle and disc (the disc is usually replaced) during the safety valve overhaul. However, it was noted that there had been an increasing trend in occurrences of micro-bonding, which is apparently related to seating surface finish. Previous MSSV overhaul procedures had specified that the valve seating surfaces be lapped to a "mirror finish" without specifying the quality of the finish. This lead to different vendors using different quality finishes. The MSSV surface finishes were subsequently specified to conform to an L4-L8 profile (i.e., lapped to a 4-8 p-in finish) to eliminate nozzle and disc micro-bonding.
RV-3-1406 A review of the history of RV-3-1406 indicated that several valve configuration changes had been implemented during the previous refueling outage. The valve spring had been replaced with a larger coil diameter spring, evaluated as equivalent. The larger spring required that new spring washers, machined to achieve the correct setpoint, be installed to match the diameter of the spring. The new spring washers were thicker, such that installation of approximately 3/8 inch thick "C" shaped spacers between the shoulders on the yoke rods and the yoke were needed. This change FACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (6) PAGE (3) Turkey Point Unit 3 05000250 allowed the spring to be extended sufficiently to reduce the setpoint adjustment to the required range.
The spacers allowed the valve to be initially set to the required setpoint. However, since the "C" shaped spacers did not provide a continuous contact surface between the yoke rod and the yoke nut, some degree of dimensional instability was introduced during Cycle 19 operation. This caused misalignment, and thus, variability of the setpoint.
The surveillance test results of RV-3-1406 showed the largest deviation from the design setpoint and consistently tested high. These test results are not consistent with micro-bonding so dimensional instability, created by the spring washers and yoke rod spacers, was determined to be the cause of the high lift setpoint. Corrective actions were specified to correct the dimensions of the spring washers and remove the yoke rod spacers from the valve.
Analysis of the Event
The existence of similar discrepancies in multiple valves is an indication that the discrepancies may well have arisen over a period of time and that the condition existed during plant operation. Based on the fact that the Unit 3 MSSV surveillance testing Cycle 19 operation, this event is reportable under 10 CFR 50.73(a)(2)(i)(B).
A review of past MSSV test failures going back to 1997 indicates that multiple failures with similar discrepancies did not occur during this time period. However, single failures, which may have existed during the operating cycle prior to surveillance testing may have occurred. The existence of single failures during the period were assessed at the time as not reportable based on no firm evidence that the failures existed during the operating cycle. Also, no occurrence of multiple, similar failures existed to support indication that the discrepancies may have arisen over a period of time. Additionally, at the time of the new MSSV failures, no previous failure experience may have existed for definitive assessment.
The following events, not previously reported, involved failures of MSSVs during setpoint testing:
RV-3-1412 lift setpoint was tested to be 4.8% high due to a clogged body drain. The clogged drain allowed water buildup to occur around the nozzle, causing increased cooling, raising the valve setpoint lifting pressure. Reactor power was reduced to occurred at the time of the test with no generic valve issues.
RV-4-1412 lift setpoint was tested to be 4.2% high due to micro-bonding. Reactor power was FACILITY NAME (1) LER NUMBER (6) RV-3-1406 lift setpoint was tested to be 3.1% high due to micro-bonding. Reactor power was RV-3-1406 lift setpoint was tested to be 4.5% high and RV-3-1412 was tested to be 3.6% high. RV-3-1406's failure is most likely due to a misalignment due to a collapsed spring. RV-3-1412's failure was determined to be due to micro-bonding. Reactor power was at 0% at the time of testing.
Analysis of Safety Significance A review of the Turkey Point Safety Analyses was performed to assess the impact of the combined effect of the four (4) MSSV's "as found" condition on Cycle 19 operation. The analyses in which the MSSVs are modeled include:
- Rod Withdrawal From Power (RWFP) - Feedwater System Malfunction - Increase in Steam Flow - Loss of Flow - Locked Rotor - Loss of External Electrical Load (LOL)/Turbine Trip - Loss of Normal Feedwater (LONF) With/Without Loss of Non-Emergency AC Power - Small Break LOCA The review indicated that only the events that are susceptible to overpressure conditions may be impacted by the "as found" condition of the subject MSSVs. These include:
Loss of External Electrical Load (LOL)/Turbine Trip (Limiting Event) - Rod Withdrawal From Power (RWFP) - Loss of Normal Feedwater (LONF) With/Without Loss of Non-Emergency AC Power The overpressure design criteria to which these transients have to comply, require that the peak pressures reached during the event (for both the Reactor Coolant System (RCS) and the Main Steam System (MSS)) do not exceed 110 percent of the design pressures (2485 psig RCS and 1085 psig MSS). This translates into the following design basis criteria:
2733.5 psig for the RCS and 1193.5 psig for the MSS.
Analysis of the three pressurization safety analyses events resulted in no challenges to the design bases of these events and it is concluded that the "as found" MSSVs performance test data did not result in any past operability concerns for Unit 3's Cycle 19 operation. Thus, this event had no impact on the health and safety of the public.
DOCKET
FACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)
Corrective Actions
1. RV-3-1406 was disassembled, inspected and repaired to correct the conditions leading to unacceptable as-found testing.
2. RV-3-1402, RV-3-1410 and RV-3-1411 were disassembled and the seating surfaces were lapped to an L4 - L8 acceptance criteria.
3. Plant maintenance procedures will be revised to include specific lapping instructions to prevent micro-bonding of the MSSV seats.
4. The MSSV manufacturer has resolved the spring washer dimensional tolerance discrepancy.
5. All Unit 3 MSSVs were tested as satisfactory prior to unit restart as part of the expanded scope testing requirements of ASME/ANSI On-1987, Part 1, paragraph 1.3.3.1(e)(1).
Additional Information
second component function identifier (if appropriate)].