ML20237A718

From kanterella
Jump to navigation Jump to search
SER Accepting Util Response to GL 95-07, Pressure Locking & Thermal Binding of Safety-Related Power-Operated Gate Valves
ML20237A718
Person / Time
Site: Summer South Carolina Electric & Gas Company icon.png
Issue date: 08/13/1998
From:
NRC (Affiliation Not Assigned)
To:
Shared Package
ML20237A712 List:
References
GL-95-07, GL-95-7, NUDOCS 9808170048
Download: ML20237A718 (5)


Text

_ _ - _ _ - _ _- - _ _ - -_ _ ____ _ _ _ ____-_. ._

t

.. meo o p k" UNITED STATES I

g NUCLEAR REGULATORY COMMISSION e WASHINGTON, D.C. 20006 0001 SAFETY EVALUATION BY THE OFFICE OF NUCI FAR REACTOR REGULATION l LICENSEE RESPONSE TO GENERIC LETTER 95-07. " PRESSURE LOCKING AND THERMAL BINDING OF SAFETY-RFI ATED POWER-OPERATED GATE VALVES" i

VIRGIL C. SUMMER NUCI FAR STATION l

DOCKET NUMBER 50-395 l

1.0 - INTRODUCTION l Pressure locking and thermal binding represent potential common-cause failure mechanisms l that can render redundant safety systems incapable of performing their safety functions.

L I

Identifying susceptible valves, and determining when the phenomena might occur, require a thorough knowledge of components, systems, and plant operations. Fluid can become

- pressurized inside flexible-wedge and double-disk gate valve bonnets. The pressurized fluid can create differential pressure across both valve disks. Pressure locking occurs when the f actuator becomes unable to overcome the additional thrust requirements produced by the j differential pressure. Thermal binding is generally associated with closing a wedge gate valve while the system is hot, and then allowing the valve to cool before attempting to open it.

Valve design characteristics (wedge and valve body configuration, flexibility, and material thermal coefficients) and specific valve pressures and temperatures encountered during various i-plant operating modes affect pressure locking or thermal binding. Operating experience -

indicates that many plants did not always consider these situations as part of the valve design basis.

i 2.0 REGULATORY REQUIREMENTS 10 CFR Part 50, Appendix A (General Design Criteria 1 and 4) and plant licensing safety analyses require and/or commit licensees to design and test safety-related components and systems to provide adequate assurance that those systems can perform their safety functions.

I Other individual criteria in 10 CFR Part 50, Appendix A apply to specific systems, and 10 CFR Part 50, Appendix B, Criterion XVI contains additional relevant provisions. We expect licensees to ensure that safety-related power-operated gate valves susceptible to pressure locking or thermal binding are capable of performing their required safety functions, in accordance with the regulations and licensing commitments.

p I

The U.S. Nuclear Regulatory Commission (NRC) issued Generic Letter (GL) 95-07, " Pressure Locking and Thermal Binding of Safety-Related Power-Operated Gate Valves" on August 17,1995. The GL requested licensees to ensure that safety-related power-operated Enclosure Y

9808170048 99081335 DR ADOCK 050

i 1

gate valves that are susceptible to pressure locking or thermal binding are capable ofperforming their safety functions within the current facility licensing bases. GL 95-07 requested that each j licensee, within 180 days of the GL date of issuance: (1) evaluate the operational configurations of plant safety-relate.d power-operated gate valves to identify valves that are susceptible to pressure locking or thermal binding, and (2) perform further analyses and take needed corrective actions (orjustify longer schedules) to ensure that the susceptible valves, identified in (1) above, are capable of performing their intended safety functions under all modes of plant operation, including test configuration. In addition, GL 95-07 requested that licensees, within 180 days of the GL date of issuance, provide to the NRC a summary description of (1) the susceptibility evaluation used to determine that valves are, or are not, susceptible to pressure l locking or thermal binding, (2) the results of the susceptibility evaluation, including a listing of l the susceptible valves identified, and (3) the corrective actions, or other dispositioning, for the I-valves identified as susceptible to pressure locking or thermal binding. The NRC issued GL 95-07 as a " compliance backfit" pursuant to 10 CFR 50.109(a)(4)(i) because modification may be necessary to bring facilities into compliance with the regulations referenced above.

In a February 13,1996, letter, South Carolina Electric & Gas Company (SCE&G) submitted its 180-day GL 95-07 response for the Virgil C. Summer Nuclear Station (VCSNS). We reviewed your submittal and requested additional information in our July 3,1996, letter. In your August 2,1996, letter, you provided the additional information. During the period of March 3 through 7,1997, we performed an inspection to review specific aspects of information summarized in your responses to GL 95-07. We documented this inspection in NRC Inspection Report 50-395/97-01. You responded to the inspection report findings in your December 29,1997, letter.

i 3.0 STAFF EVALUATION 3.1 Scope of Licensee's Review GL 95-07 requested that licensees evaluate the operational configurations of safety-related .

power-operated gate valves in their plants to identify valves that are susceptible to pressure I locking or thermal binding. Your February 13,1996, August 2,1996, and December 29,1997, SCE&G letters described the scope of valves evaluated in response to GL 95-07. We have

' reviewed the scope of your susceptibility evaluation performed in response to GL 95-07, and ,

found it complete and acceptable. 1 l

You did not include valves XVG08071 A,B-RH and XVG08072A,B-RH, Shutdown Cooling l Suction, in the scope of GL 95-07 because you use these valves during plant conditions below l Hot Standby. The licensing base for VCSNS is Hot Standby The criteria for determining the 1 scope of power-operated valves for GL 95-07 are consistent with the staff's acceptance of the scope of motor-operated valves associated with GL 89-10, " Safety-Related Motor-Operated Valve Testing and Surveillance."

, 3.2 Corrective Actions l

GL 95-07 requested that licensees, within 180 day., perform further analyses as appropriate, and take appropriate corrective actions (or justify longer schedules), to ensure that the

susceptible valves identified are capable of performing their intended safety function under all modes of plant operation, including test configuration. Your submittats discussed proposed

,k-corrective actions to address potential pressure-locking and thermal-binding problems. We discuss our evaluation of your actions in the following paragraphs:

a. You stated that you evaluated the following valves for pressure locking and modified them to eliminate the potential for pressure locking:

-XVG08811 A,B-SI Recirculation Sump to Low Head Safety injection Pumps A/B

-XVG08812A,B-SI Reckculation Sump to Low Head Safety injection Pumps A/B l We find that physical modification to valves susceptible to pressure locking is an appropriate j corrective action to ensure valve operability, and is thus acceptable.

l b. .You stated that you would modify valve XVG08889-SI, Low Head Safety injection to Hot Legs, by the end of the April 1999 refueling outage, to eliminate the potential for pressure locking. As short-term corrective action, you prepared a safety evaluation that i

demonstrated that one of the redundant chaiging system hot leg injection flow paths can provide sufficient hot leg recirculation flow. We find that the short-term corrective action of using the 10 CFR 50.59 process to change the facility is acceptable, until you complete the modification to XVG08889-Si to eliminate the potential for pressure locking.

I

c. You stated that you used a thrust-prediction methodology developed by Commonwealth Edison Company (Comed) to demonstrate that the followirig valves could open under pressure-locking conditions:

-XVG08000A,B,C-RC Pressurizer Power-Operated Relief Valve Block Valves

- XVG08801 A,B-SI . High Head Safety injection to Cold Legs

- XVG08884-SI High Head Safety injection to Hot Legs

- XVG08885-Si Altemate High Head Safety injection to Hot Legs

- XVG08886-SI High Head Safety injection to Hot Legs You stated that you would modify the actuators for valves XVG08884-SI, XVG08885-SI and XVG08886-Si to increase the margin between actuator thrust capability and the thrust required to overcome pressure locking. You scheduled to complete these modifications by the end of the 1999 spring refueling outage. You used the Comed pressure locking prediction methodology to demonstrate that XVG08884-SI, XVG08885-SI and XVG08886-Si are operable, until you modify these valve actuators. The margins between actuator capability and the thrust required to overcome pressure locking are lower than what is acceptable for long-term corrective action. However, the Comed methodology demonstrates that the valves are' operable, and is satisfactory short-term corrective action until you complete planned modifications as scheduled.

On April 9,1997, we held a public meeting to discuss the technical adequacy of the Comed pressure-locking thrust prediction methodology and its generic use by licensees in their submittals responding to GL 95-07. We issued the minutes of the public meeting on April 25,1997. At the public meeting, Comed recommended that, when using its methodology, minimum margins should be applied between calculated pressure-locking l thrust and actuator capability. Comed indicated that its methodology is undergoing review

l. and may be revised. Calculations used to demonstrate that valves can overcome pressure

!' locking are required to meet the requirements of 10 CFR Part 50, Appendix B, Quality Assurance Criteria for Nuclear Power Plants. Therefore, controls are required to be in place

O S

4_

to ensure that any industry pressure-locking thrust prediction methodology requirements and revisions are properly implemented. Under this condition, we find that the Comed methodology provides a technically sound basis for assuring that valves susceptible to pressure locking are capable of performing their intended safety-related function.

d. You identified several test configurations where systems could be inadvertently pressurized from leakage through closed isolation valves. You stated that you revised procedures to ensure that a train is not inadvertently pressurized when testing the opposite train, to eliminate the potential for the following valves to pressure lock:

- XVG01001A,B-EF Motor-Driven Emergency Feedwater Pump A/B Service Water A/BCross Connect Valves -

- XVG03002A,B-SP Sodium Hydroxide to Reactor Building Spray Pump A/B Suction

- XVG03003A,B-SP A/B Train Reactor Building Spray Header Isolation

- XVG03005A,B-SP Recirculation Sump to Reactor Building Pump A/B Suction We find that your procedural changes to require monitoring the pressure in a train opposite to the train that is being tested provides assurance that pressure locking conditions are eliminated, and are thus acceptable.

e. You identified several configurations where valves could become pressure locked when j performing pump surveillance testing. You stated that you revised procedures to remove
i. the requirement to shut the following valves during testing to eliminate the potential for the valves to pressure lock:

l

- XVG08888A,B-Si Low Head Safety injection to Cold Legs

- XVG03109A,B,C,D-SW Reactor Building Cooling Units 1A/2A/1B/2B Outlet isolation Valves We find that your procedural changes to eliminate the requirement to shut valves during testing provide assurance that pressure locking conditions are eliminated, and are thus acceptable.

f. You stated that you revised procedures to cycle Valves XVG03003A,B-SP, Reactor Building Spray Pump Discharge, following reactor building spray pump testing to eliminate the potential for pressure locking. We find that your procedural changes to require cycling the valves as corrective actions previde assurance that pressure locking conditions are eliminated, and are thus acceptable.
g. You stated that you would revise procedures to cycle XVG08000A,B,C-RC, Pressurizer

. Power-Operated Relief Valve Block Valves, approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after being closed to isolate a leaking power-operated relief valve to verify that the valve is not thermally bound.

These valves are also equipped with spring compensators which help mitigate the potential for thermal binding. At VCSNS, the pressurizer power-operated relief valve block valves are not used for cold overpressure protection. This reduces the required temperature range in which the valves are required to operate. We find that your procedural changes to require cycling the valves provide assurance that thermal binding conditions are adequately identified and eliminated, and are thus acceptable. When historical data demonstrate that I the valves are not susceptible to thermal binding, then cycling the valves will no longer be u

o 1

(

required. However, if actuator setup conditions or valve unwedging forces are significantly altered, then it would be necessary to demonstrate that the valves are not susceptible to thermal binding.

h. . You stated that you evaluated valves within the scope of GL 95-07 for thermal binding. You assumed that thermal binding would not occur below specific temperature thresholds, when evaluating whether valves were susceptible to thermal binding. These assumptions were based on industry experience. You did not consider that gate valves in systems with a L

normal operating temperature less than approximately 200*F were susceptible to thermal binding. Further, you did not consider those flexible wedge gate valves that are shut and that experience a cooldown differential temperature of less than 100*F were susceptible to l thermal binding. You stated that there were no solid wedge gate valves within the scope of GL 95-07.

The screening criteria you used appear to provide a reasonable approach to identify those l valves that might be susceptible to thermal binding. We conclude ; hat your actions to L address thermal binding of gate valves are acceptable, until more definitive industry criteria are developed.

4.0 CONCLUSION

On the basis of this evaluation, we find that you have performed appropriate evaluations of the operational configurations of safety-related power-operated gate valves to identify valves at the VCSNS that are susceptible to pressure locking or thermal binding. In addition, we find that you have taken, or are scheduled to take, appropriate corrective actions to ensure that these valves are capable of performing their intended safety functions. Therefore, we conclude that you have adequately addressed the requested actions discussed in GL 95-07.

Principal Contributor S. Tingen, NRR Date:

I I