05000318/LER-2008-001

From kanterella
Jump to navigation Jump to search
LER-2008-001, Pressurizer Safety Valve Setnoint Low Due to Different Temperature Profiles
Docket Numbersequential Revmonth Day Year Year Month Day Yearnumber No. Ccnpp, Unit 1 05000 317
Event date:
Report date:
Reporting criterion: 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications
3182008001R00 - NRC Website

A. PRE-EVENT PLANT CONDITIONS

Unit 1 was operating at 100 percent of rated thermal power and Unit 2 was operating at 100 percent of rated thermal power on January 4, 2008, prior to the discovery date of the subject event. Prior to the date of discovery, all of the affected valves had been removed from the plant (Units 1 and 2) and replaced with valves set at the correct temperature profile.

B. EVENT

On January 4, 2008, Calvert Cliffs Nuclear Power Plant (CCNPP) received vendor confirmation that the temperature profile used to establish the As-Found Lift Setting for pressurizer safety valve (PSV) 2RV200 (Serial Number BV02950) in 2006 was not consistent with the temperature profile used to establish the As-Left Lift Setting for the valve in 2003. As a result, the As-Found Lift Setting was outside the limits specified in CCNPPs Technical Specifications. The valve (BV02950) was not installed in the system at the time vendor confirmation was received.

Because 2RV200 (BV02950) had partially lifted (simmered) during the November 16, 2006 Unit 2 reactor trip [reference Licensee Event Report (LER) 318/2006-001, Revision 1 dated February 23, 2007], the valve was removed from the plant and shipped to the vendor laboratory (in December 2006) for As-Found testing and a complete refurbishment. Initially (February 2007), the apparent cause evaluation (ACE IRE-018-411) determined that due to a loose compression screw/valve bonnet thread interface, thermal effects from the November 16, 2006 Unit 2 trip may have caused the lift setting to decrease during the event. Therefore, at that time, firm evidence was not available to determine the actual duration of the low setpoint condition.

However, a subsequent revision of the ACE (IRE-027-718; completed in February 2008) determined that the low lift setting was caused by a difference in the temperature profiles established during As-Left Lift Setting in 2003 and As-Found Lift Setting in 2006. This new information provides verification that the condition existed for a time longer than the Technical Specification allowed completion time. Therefore, this condition is reportable pursuant to the reporting criteria specified in 10 CFR 50.73(a)(2)(i)(B).

Calvert Cliffs owns eight PSVs, four sets of two that are rotated between a specific location.

The revised ACE included an extent of condition review to evaluate the impact of the modified temperature on all of the PSVs. The extent of condition review determined that the modified temperature profile condition affected all PSVs installed in the plant from 2004 through 2006, with their As-Found Lift Settings established at the old temperature profile. Specifically, 1RV200 (BN04373), 1RV201 (BM07948), 2RV200 (BN04375), and 2RV201 (BS03213), installed from 2004-2006 and from 2003-2005 for Unit 1 and Unit 2 respectively, were affected. Although the As-Found Lift Settings for those valves were within the Technical Specification limits, had they been tested at the ambient conditions associated with normal operating pressure, they most likely would have tested low outside the Technical Specification limits. Therefore, this LER is applicable to CCNPP Units 1 and 2.

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION (9-2007) I2REV

C. INOPERABLE STRUCTURES, COMPONENTS, OR SYSTEMS THAT CONTRIBUTED

TO THE EVENT

Unit 2 PSV 2RV200 (BV02950) was determined to be inoperable. That inoperable condition is the bases for this report. The extent of condition review determined that 1RV200 (BN04373, 1RV201 (BM07948), 2RV200 (BN04375), and 2RV201 (BS03213), installed from 2004-2006 and from 2003-2005 for Unit 1 and Unit 2, respectively were also inoperable, but capable of meeting their safety function of overpressure protection.

D. DATES AND APPROXIMATE TIMES OF MAJOR OCCURRENCES:

April 2003, 2RV200 (BV02950) As-Left Lift Setting established at the old temperature profile (320 F +/- 20 F). The average inlet flange temperature and average As-Left set pressure for BV02950 during As-Left testing were 294 F and 2505 psia, respectively (3 lifts each time).

Spring 2004, temperature profile modified to current values (includes Inlet Flange temperature of 390 F +/- 20 F).

November 16, 2006, 2RV200 (BV02950) "simmered" following Unit 2 automatic reactor trip.

December 2006, BV02950 shipped to the vendor lab for As-Found testing and a complete refurbishment. The As-Found testing was performed at the new temperature profile (390 F +/- 20 F). The average inlet flange temperature and average As-Found set pressure for BV02950 during As-Found testing in 2006 were 404 F and 2422 psia respectively (3 lifts each time). The first As-Found set pressure was 2414 psia. This is lower than the minimum Technical Specification allowed value of 2475 psia.

December 2006, BV02950 was completely disassembled and inspected. During this inspection the only unsatisfactory condition identified was that the compression screw/bonnet thread interface was considered loose by the vendor technician. The compression screw and bonnet were both tested with the vendor's go/no-go gauge. Both failed this test.

February 2007, the original revision of the ACE attributed the cause of the low lift and possible simmering of BV02950 to this thread looseness.

October 2007, another detailed inspection of BV02950 was performed. During that inspection, several components were identified as off-normal, including the compression screw/bonnet thread interface. None of the off-normal conditions were considered to be credible causes of a safety valve lifting low.

December 2007, the vendor notified CCNPP that the loose threads could not have caused the low lift or possible simmering of BV02950.

NRC FORM 366A� U.S. NUCLEAR REGULATORY COMMISSION (9-2007) January 4, 2008, the vendor confirmed that a different temperature profile was used for the As- Found Lift Setting (performed in December 2006) than the temperature profile used for the As- Left Lift Setting (performed in April 2003) for BV02950.

February 2008, the revised ACE determined that the low setpoint was due to differences in temperature profiles. The extent of condition review determined that five of the eight PSVs owned by CCNPP had been installed with their setpoint established at the wrong temperature profile in the last three years.

E. OTHER SYSTEMS OR SECONDARY FUNCTIONS AFFECTED

No other systems or secondary functions were affected. As discussed in LER 318/2006-001 Revision 01, the impact of the PSV on the November 16, 2006 event was bounded by the impact from the open power-operated relief valve (PORV). All of the PSVs that were affected by the change in temperature profile are discussed in Section III of this LER.

F. METHOD OF DISCOVERY

On January 4, 2008, the vendor confirmed that a different temperature profile was used for the As-Found Lift Setting (performed in December 2006) than the temperature profile used for the As-Left Lift Setting (performed in April 2003) for BV02950. The average inlet flange temperature and average As-Left set pressure for BV02950 during As-Left testing in 2003 were 294 F and 2505 psia respectively (three lifts each time). The average inlet flange temperature and average As-Found set pressure for BV02950 during As-Found testing in 2006 were 404 F and 2422 psia respectively (three lifts each time). As stated previously, the first As-Found set pressure was 2414 psia.

G. MAJOR OPERATOR ACTION

No major operator actions were taken as a result of this condition.

H. SAFETY SYSTEM RESPONSES

experienced an automatic reactor trip due to high pressure in the pressurizer. Unit 2 was operating at 100 percent power at the time. The high pressurizer pressure was due to a main turbine load reduction that was caused by a tagging error. As a result of the trip, Reactor Coolant System (RCS) pressure rose to approximately 2420 psia causing the two PORVs to open. Both PORVs opened as designed. One PORV (2ERV- 402) remained open approximately 90 seconds causing RCS pressure to reduce to 1500 psia. The PORV should have closed slightly below the 2400 psia reset value. Plant data indicated that one of the two PSVs (2RV200) momentarily "simmered", allowing a slight decrease below the level of accumulation necessary to fully open the valve. The impact of the PSV on the event was bounded by the impact from the open PORV.

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION (9-2007)

II. CAUSE OF EVENT:

The event is NUREG-1022 Cause Code [D], defective procedure.

Lack of detail in the review process for determining which PSVs required testing at each temperature profile led to BV02950 As-Found test outside Technical Specification limits. This also prevented As-Found testing BN04373, BM07948, BN04375, and BS03213 at the correct temperature profile.

On January 4, 2008, the vendor confirmed that a different temperature profile was used for the As-Found Lift Setting (performed in December 2006) than the temperature profile used for the As-Left Lift Setting (performed in 2003) for BV02950.

Nuclear Regulatory Commission Information Notice 89-90, Supplement 1 was issued in 1991 to alert the industry to potential problems resulting from operating PSVs in an environment different from that used to establish the PSV setpoint. Specifically, an increase in the temperature of the valve results in the expansion of the body and elongation of the bonnet. This relieves spring pressure and reduces the lift setpoint of the valve. Conversely, a decrease in temperature of the valve increases the lift setpoint. The Information Notice included data from CCNPP, describing a PSV setpoint shift of 60 to 150 psi for a 120 F change in valve body temperature. Also, CCNPP has records from testing at the vendor lab in March of 1995 indicating that an increase in valve temperature of approximately 150 F resulted in a decrease in valve setpoint of approximately 140 psi. Both of these data points correspond very well to the 2RV200 (BV02950) testing results in that a 110 F change in valve body temperature (inlet flange) led to a setpoint shift of approximately 80 psi. Contributing to the difference between the As-Left and As-Found temperature profiles for BV02950 was an error during the As-Left testing in 2003. The actual valve inlet temperature (294 F) was lower than allowed by the specification (300 F minimum). As-Found testing was also conducted high (404 F) in the band (410 F maximum). This led to greater temperature difference (110 F) than would have been expected using the temperature profile averages (70 F).

The historical temperature profiles for CCNPPs PSVs are provided in the table below.

Historical CCNPP Temperature Profiles Year Upper Bonnet Lower Bonnet Inlet Flange 1994 105 F +/- 10 F 140 F +/- 10 F 300 F +/- 20 F 1998 105 F +/- 10 F 140 F +/- 10 F 320 F +/- 20 F 2004 110 F +/- 10 F 140 F +/- 10 F 390 F +/- 20 F In 1994 CCNPP established a recommended temperature profile for the PSVs based upon field data. In 1998, the inlet flange temperature was modified slightly (due to vendor testing issues) and incorporated into the purchase order used to overhaul and test the valves. In 2004 the temperature profile was modified again based on data obtained during the Unit 1 Refueling Outage The data was obtained to ascertain when the valves reach thermal equilibrium. This data indicated that the Unit 1 PSVs were at a higher temperature than previously determined.

NRC FORM 366A � U.S. NUCLEAR REGULATORY COMMISSION (9-2007) I� REV This information was incorporated into the purchase order in November of 2004 and subsequently used for final As-Left testing. Calvert Cliffs planned to use the old temperature profile for As-Found testing and the new for As-Left testing until all valves had been As-Left and As-Found set at the same "new" temperature profile. This decision was made to meet the requirements specified in the Inservice Testing Program to determine if the valve had degraded.

Changing the temperature profile between As-Found and As-Left could mask the true "health" of the valve internals, and thus not meet the requirements in the lnservice Testing Program.

However, to verify operability, the As-Found Lift Setting should be established for the ambient conditions associated with normal operating pressure. Since the ambient conditions associated with normal operating conditions (new temperature profile developed in 2004) were not established during the As-Found Lift Setting, the valves were inoperable.

Extensive testing has been performed at CCNPP to determine the correct temperature profile.

Based upon a review of the information obtained in 2004, we believe that the current temperature profile is correct. In fact, the Unit 2 trip in November 2006 actually provides an empirical data point supporting the validity of the current temperature profile. During the November 2006 trip, 2RV200 most likely simmered at approximately 2400 psia. At the vendor lab, during As-Found Lift Setting at the new temperature profile, the valve popped (lifted) at 2414 psia. This was slightly higher than the simmer point (2400 psia). If the temperature profile was not correct, the valve would have simmered farther away from the lift point.

III.� ANALYSIS OF THE EVENT:

Each Unit at CCNPP has two PSVs (1/2RV200 and 1/2RV201) designed to limit RCS pressure to a maximum of 110 percent of design pressure (2500 psia). The Technical Specification defined setpoints for these valves are as follows:

� As-Found As-Left � � Valve Lift Setting (psia) Lift Setting (psia) � �1/2RV200 >/= 2475 and � �1/2RV201 >/= 2514 and The Technical Specification Bases state that the As-Found setpoints are the limits for operability, i.e., if a valve lifts outside of those setpoints it is inoperable. Calvert Cliffs owns eight PSVs, four sets of two that are rotated between a specific location. The As-Found Lift Setting for 2RV200 (BV02950) measured in 2006 was 2414 psia. This is lower than the Technical Specification Surveillance Requirement (SR) allowed value of 2475 psia. The low lift setting was caused by a difference in the temperature profiles established during As-Left Lift Setting in 2003 and As-Found Lift Setting in 2006. BV02950 was installed at 2RV200 location in the Spring 2005 and removed from the plant in November 2006. During that time, the lift setting was not within the Technical Specification SR defined setpoint limit. The extent of condition review determined that subsequent to establishing the correct temperature profile in 2004, five of the eight PSVs owned by CCNPP were installed with their As-Left Lift Setting established at the wrong temperature profile. Those valves were inoperable while installed in their respective plant locations during applicable modes. Per 10 CFR 50.73(a), unless otherwise specified, NRC FORM 366A � U.S. NUCLEAR REGULATORY COMMISSION (9-2007) I� REV events shall be reported if they occurred within three years prior to the date of discovery. With one PSV inoperable, the Technical Specification Limiting Condition for Operation (LCO) 3.4.10.A required action is to restore the valve to operable status within a 15 minute completion time. If this required action cannot be met, Technical Specification LCO 3.4.10.B requires the plant to be placed in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and to reduce all RCS cold leg temperatures to recognize and meet the requirements of Technical Specification LCO 3.4.10 also should have required entry into Technical Specification LCO 3.0.3. The subject condition as applicable to 2RV200 (BV02950)existed for a time longer than the allowed completion time. The subject condition as applicable to 1RV200 (BN04373) and 1RV201 (BM07948) existed from 2004-2006.

The subject condition as applicable to 2RV200 (BN04375) and 2RV201 (BS03213) existed from 2003-2005. These conditions existed longer than the Technical Specification completion times for the associated required actions (6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />). Therefore, for the conditions specified, this event is reportable in accordance with 10 CFR 50.73(a)(2)(i)(B).

There were no actual nuclear safety consequences incurred from this event. As previously discussed, the new temperature profile established in 2004 more accurately reflects actual plant conditions. All four PSVs currently installed in the plant have their As-Left Lift Settings established per the new temperature profile. As discussed above, subsequent to 2004, five of CCNPPs PSVs were set and installed with As-Left Lift Settings established per the old temperature profile. With the exception of 2RV200 (BV02950), the As-Found Lift Settings were also established at the old temperature profile. As a result, and because the valves were not degraded, their lift settings were within the Technical Specification limits. However, they should have been considered inoperable while installed during applicable modes. Both of the installed Unit 1 PSVs were susceptible to an early PSV lift from 2004-2006, Unit 2 was vulnerable from 2003-2006 (both valves from 2003-2005 and 2RV200 from 2003-2006). Realizing that the valves could have lifted early if challenged, a probabilistic risk assessment analysis was performed. The estimated increase in conditional core damage probability was less than 5E-07 and the estimated increase in large early release probability was less than 5E-09 per year for the subject condition.

IV.� CORRECTIVE ACTIONS:

A.� ACTION TAKEN TO RETURN AFFECTED SYSTEMS TO PRE-EVENT NORMAL

STATUS:

1. 2RV200 removed and replaced with a valve set at the correct temperature profile.

2. Verified all installed PSVs set at correct temperature profile.

B.� ACTION TAKEN OR PLANNED TO PREVENT RECURRENCE 1.� The valve removed from 2RV200 location (BV02950) will be As-Left tested at the current temperature profile.

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION (9-2007) 2. The procedure will be changed to document that changes to the temperature profile will change the setpoint of the valve.

3. The procedure will be changed to track relief valve testing prior to installation in the plant.

V. ADDITIONAL INFORMATION

A. FAILED COMPONENTS:

The subject valves are American Society of Mechanical Engineers Boiler and Pressure Code approved pressurizer safety valves designed to limit RCS pressure to a maximum of 110 percent of design pressure. The safety valves are totally enclosed, back pressure compensatory, spring-loaded valves. The valves are manufactured by Dresser Consolidated, Inc. (EPIX Identification number D243). The valves affected by the subject condition are 1RV200 (BN04373), 1RV201 (BM07948), 2RV200 (BV02950), 2RV200 (BN04375), and 2RV201 (BS03213).

B. PREVIOUS LERs ON SIMILAR EVENTS

A review of Calvert Cliffs' events over the past several years was performed. No previous LERs on similar events were identified.

As discussed above, prior to 2004, all of the PSVs had their As-Found Lift Setting and their As- Left Lift Setting established at the old temperature profile. Since the As-Found Lift Setting and As-Left Lift Settings were at the same temperature profile, and since the valves were not degraded, the Technical Specification values were met.

C. THE ENERGY INDUSTRY IDENTIFICATION SYSTEM (EIIS) COMPONENT

FUNCTION IDENTIFIER AND SYSTEM NAME OF EACH COMPONENT OR SYSTEM

REFERRED TO IN THIS LER:

IEEE 803 IEEE 805 Component EIIS Function System ID Pressurizer Safety Valves

RV AB

Power Operated Relief Valves 20 AB Pressurizer

PZR AB

D.SPECIAL COMMENTS None