Learning from valve safety experience

Above: Operators did not always ensure that valve-specific factors were used to calculate appropriate thrust and torque

The US Nuclear Regulatory Commission (NRC) has issued a supplement to its May 2021 Information Notice 2021-01, “Lessons Learned from US Nuclear Regulatory Commission Inspections of Design-Basis Capability of Power-Operated Valves at Nuclear Power Plants”.

The US regulator wanted to alert operators to lessons learned from NRC inspections of the design-basis capability of power-operated valves (POVs) at nuclear power plants.

While issuing the Information Notice, NRC was also working on Inspection Programme 71111.21N.02 to assess the reliability, functional capability and design-basis capability of risk-important POVs to determine whether licensees are maintaining POVs’ capability to perform as intended under design-basis conditions.

The inspection programme was completed at the end of 2022 and has been discussed at industry meetings, but participants requested that the NRC provide a complete list of the lessons as soon as possible. The NRC staff suggests that licensees review this information for applicability to their facilities and consider actions, as appropriate, to identify and address similar issues.

The NRC found that:

• Licensees did not always ensure that valves were properly included and categorised within the scope of the in-service testing (IST) programme in Title 10 of the Code of Federal Regulations. The NRC inspectors found limitations, for example, in POVs with leakage limitation safety functions, or with either remotely or manually-operated safety functions.
• Licensees did not always properly determine the operating requirements and actuator capability for POVs to perform their safety functions. Appropriate parameters to be addressed when calculating valve operating requirements or actuator capability may include valve friction coefficients or valve factors, maximum differential pressure conditions, motor torque temperature derating factors, stem friction coefficients, and butterfly valve bearing friction coefficients. Licensees did not always justify the use of POV parameters from outside sources.
• Licensees did not always ensure that valve-specific valve factors were used, if determined to be higher than generic valve factors. For example, for globe valves, there is a potential for increased thrust and torque requirements to operate globe valves under high-flow dynamic conditions. The unwedging load required for valves is part of the evaluation of the capability of POVs to open to perform their safety functions.
• Licensees did not always ensure that all normal operating loads that act simultaneously with seismic loads were addressed.
• Licensees did not always ensure that sufficient information and test data were developed to validate the assumptions for rate-of-loading and load-sensitive behavior for plant-specific motor operated valve (MOV) applications.

Most licensees agreed in 1996 to implement a Joint Owners Group (JOG) Program on motor operated valve Periodic Verification in response to an NRC letter (Periodic Verification of Design-Basis Capability of Safety-Related Motor-Operated Valves).

The programme included tests performed by the participating licensees at their plants over five years, to assess whether there was a potential for degradation of valve friction coefficients for various valve types and applications.

Test results collected by participants are only applicable to the implementation of the JOG programme, because of the limited amount of MOV test data and the different methods used by individual licensees to evaluate the test data.

Above: NRC regulations supplement the testing requirements for MOVs by requiring licensees establish a programme to ensure that MOVs continue to perform their design-basis safety functions

For example, valve friction coefficients determined for MOVs as part of the JOG MOV Program do not represent a database of valve friction coefficients that can be applied in general to calculate the thrust and torque required to operate various MOVs under design-basis conditions. The NRC inspectors found that licensees did not always re-justify the qualifying basis for MOVs following extensive maintenance (such as disassembly) to determine whether the valves were susceptible to performance degradation as part of the JOG MOV Program.

New commitments

The NRC regulations in 10 CFR 50.55a(b)(3)(ii) supplement the testing requirements for MOVs by requiring that licensees establish a programme to ensure that MOVs continue to be capable of performing their design-basis safety functions.

Licensees committed to implementing the JOG MOV Program are expected to follow suit.

For example, the JOG MOV Program does not include grace periods for JOG test intervals. Further, the JOG program schedule is specified in years rather than refueling outages. In addition, a change in the risk ranking of an MOV, or an adjustment to MOV capability margin based on performance data, can result in a different diagnostic testing interval under the JOG MOV Program.

Some MOVs are outside the scope of the JOG MOV Program, and are defined by JOG as Class D valves. Licensees are required to establish methods to periodically demonstrate the design-basis capability of their Class D valves

Testing

Among testing issues, the NRC inspectors found that:

• Licensees did not always ensure that POV tests were properly conducted, acceptance criteria for the POV testing applied the correct assumptions (such as actuator thrust limits), proper evaluations of test data were completed to demonstrate that the POVs can perform their safety functions, and records of evaluations were maintained in accordance with plant procedures.
• Some POVs have specific limitations related to leakage past the valve disk when closed. MOVs can be set to fully close and meet their leakage limitations when controlled by the torque switch. The NRC inspectors found that licensees did not always have a valid test or analysis demonstrating that the limit switch control setting of the MOV under static conditions would achieve the required leaktight performance when the MOV is closed under dynamic conditions.
• Licensees did not always justify the qualification of POVs to perform their design-basis safety functions, including functional, environmental, and seismic capability. Some licensees lacked adequate justification to extend the life of installed POVs.
• The ASME Code requires that valves with remote position indicators be observed locally at least once every two years to verify that valve operation is accurately indicated. The NRC wants licensees to verify that valve operation is accurately indicated by supplementing valve position indicating lights with other indications, such as flow meters or other suitable instrumentation.
• Licensees did not always conduct a detailed evaluation of the effects of backseating on the valve bonnet and stem to verify structural integrity.
• Licensees did not always perform a complete verification and validation of POV computer software prior to implementation. These calculation methodologies need verification and validation for appropriate assumptions and data points. Further, stroke time might be calculated improperly when computer data are used to measure the MOV stroke time.
• Some licensees rely on the actuator handwheel to manually operate MOVs to perform important functions and licensees did not always evaluate the handwheel for proper sizing and good working condition.
• Licensees did not always determine a proper lubrication interval for each MOV stem to address potential lubrication grease degradation, which can adversely affect MOV operation.
• Licensees did not always follow their procedures for maintaining records associated with POV qualification, testing, operation, maintenance, and corrective action.

The NRC staff discussed the above issues in detail with the applicable licensees during then POV inspections. The licensees took action to address any immediate concerns related to these issues identified by the NRC inspectors. In many instances, the issues were determined to be minor, because of the capability margin available for the specific POVs being evaluated at the applicable nuclear power plant.