ML090400621

to Final Safety Evaluation for Nuclear Energy Institute (NEI) Addenda 3, 4, 5, 6, and 7 to Electric Power Research Institute (EPRI) Topical Report (TR) 103237, EPRI Mov (Motor-Operated Valve) Performance Prediction Program, Rev. 2.

ML090400621
Person / Time
Site:	Nuclear Energy Institute
Issue date:	02/24/2009
From:	Bone S Division of Policy and Rulemaking
To:	Jeffrey Riley Nuclear Energy Institute
Mensah T, NRR/DPR, 415-3610
Shared Package
ML090430444	List: ML090400610 ML090400621
References
TAC MD3236, TR-103237
Download: ML090400621 (14)

Text

SUPPLEMENT 4 TO FINAL SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION ADDENDA 3, 4, 5, 6, AND 7 TO ELECTRIC POWER RESEARCH INSTITUTE (EPRI)

TOPICAL REPORT (TR)-103237, EPRI MOV PERFORMANCE PREDICTION PROGRAM, REVISION 2 NUCLEAR ENERGY INSTITUTE (NEI)

PROJECT NO. 689

1.0 INTRODUCTION AND BACKGROUND

1.1 Purpose Topical Report (TR)-103237, "EPRI MOV [Motor-Operated Valve] Performance Prediction Program," may be used by licensees to demonstrate the design-basis capability of valves when valve-specific design-basis test data are not available. The EPRI MOV Performance Prediction Methodology (PPM) includes methods for prediction or evaluation of system flow parameters; gate, globe, and butterfly valve performance; and motor-actuator rate-of-loading effects. The EPRI MOV PPM includes computer models, software, and hand calculation models to implement individual valve and system models and methods. The EPRI MOV PPM is not limited to MOVs. It can also be used to determine air-operated valve (AOV) and hydraulic-operated valve (HOV) requirements.

1.2 Background In February 1994, the NEI submitted the EPRI TR-103237, Revision 0 (Reference 1), for NRC review. On November 30, 1995, the NEI submitted the EPRI TR-103237, Revision 1 (Reference 2), for NRC review. On March 15, 1996, the NRC staff issued a safety evaluation (SE) (Reference 3) documenting the NRC staff's acceptance of the EPRI MOV PPM described in the subject TR, with certain conditions and limitations. The SE addressed the EPRI computer model for globe and butterfly valves and various gate valves, and EPRI hand-calculation models for Anchor/Darling double disk gate valves and Westinghouse Electric Company flexible wedge gate valves.

On February 20, 1997, the NRC staff issued Supplement 1 to the SE (Reference 4) dated March 15, 1996. This supplemental SE discussed two additional gate valve (WKM parallel expanding and Aloyco split wedge) hand-calculation methods and highlighted other aspects of the TR.

ENCLOSURE

On September 8, 1999, the NEI submitted for NRC staff review Addendum 1, PPM Version 2.0, and Addendum 2, Thrust Uncertainty Method, (Reference 5) to EPRI TR-103237, Revision 2. Version 2.0 of the EPRI MOV PPM described in Addendum 1 to the TR resolved several previous modeling errors and incorporated other improvements to the modeling software. In Addendum 2 to EPRI TR-103237, Revision 2, the EPRI describe the development of its Thrust Uncertainty Method that takes into account conservatism in the EPRI MOV PPM to predict a more realistic thrust requirement for closing gate valves.

On April 20, 2001, the NRC staff issued Supplement 2 to the SE (Reference 6) dated March 15,1996, concluding that the changes made to the PPM in Addendum 1 to EPRI TR-103237, Revision 2, improve the ability of the EPRI model to predict the applicable thrust or torque required to operate gate, globe, and butterfly valves.

On January 5, 2001 (Reference 7), December 6, 2001 (Reference 8), and June 10, 2002, (Reference 9), the NEI submitted for NRC staff review revisions to Addendum 2 to EPRI TR-103237, Revision 2. These revisions involved the establishment of additional criteria for application of the Thrust Uncertainty Method for gate valves under cold water applications within the scope of the Thrust Uncertainty Method.

By letter dated September 30, 2002, (Reference 10), the NRC staff issued Supplement 3 to the SE dated March 15, 1996. The NRC staff concluded that the Thrust Uncertainty Method developed by the EPRI is acceptable for the prediction of minimum allowable thrust at control switch trip (or flow isolation) for applicable gate valves under cold water applications within the scope of the Thrust Uncertainty Method and subject to the applicable limitations and conditions specified in the Thrust Uncertainty Method.

On June 8, 2004, (Reference 11), the NEI submitted Addendum 5, PPM Version 3.1 Software Changes, Addendum 6, PPM Version 3.2 Software Changes, and recent PPM error and information notices to EPRI TR-103237, Revision 2, for NRC staff review. Addendum 5 provided PPM Software Version 3.1; updated the PPM to a Window-based application; automated certain manual calculations; adjusted the software to allow more ready use for AOV and HOV applications; and incorporated software and documentation improvements.

Addendum 6 provided PPM Software Version 3.2; corrected potential non-conservative effects from upstream disturbances; and prevented incorrect use of best estimate butterfly valve torque predictions.

On January 6, 2006, (Reference 12), the NEI submitted Addendum 3, An Improved and Validated Method for Predicting Gate Valve Unwedging Thrust Requirements, Addendum 4, Use of Static Closure Data for Determining Stem Stem-Nut Coefficients of Friction at Unwedging, and Addendum 7, PPM Version 3.3 Software Changes, to EPRI TR-103237, Revision 2, for NRC staff review. Addenda 3, 4, and 7 document changes made to the EPRI PPM computer code between Versions 3.2 and 3.3, and also provide improved methods for evaluating gate valve undwedging thrust requirements.

On April 23, 2007 (Reference 13), and June 3, 2008 (Reference 14), the NRC requested the NEI to submit additional information to support its request. The NEI submitted the requested information in letters dated September 27, 2007, (Reference 15) and August 7, 2008, (Reference 16).

This SE provides the NRC staff evaluation, based on the review of Addenda 3, 4, 5, 6, and 7 to EPRI TR-103237-R2.

2.0 REGULATORY EVALUATION

The NRC regulations require that MOVs important to safety be treated in a manner that provides assurance of their intended performance. Criterion 1 to Appendix A, General Design Criteria for Nuclear Power Plants, to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR Part 50) states, in part, that structures, systems, and components important to safety shall be designed, fabricated, erected, and tested to quality standards commensurate with the importance of the safety functions to be performed. The quality assurance program to be applied to safety-related components is described in Appendix B, Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants, to 10 CFR Part 50. In Section 55a of 10 CFR Part 50, the NRC requires licensees to establish inservice testing (IST) programs in accordance with the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code and, as IST programs are updated according to 10 CFR 50.55a, the ASME Code for Operation and Maintenance of Nuclear Power Plants.

In response to concerns regarding MOV performance, on June 28, 1989, the NRC staff issued Generic Letter (GL) 89-10, "Safety-Related Motor-Operated Valve Testing and Surveillance,"

(Reference 17) which requested that nuclear power plant licensees and construction permit holders ensure the capability of MOVs in safety-related systems to perform their intended functions by reviewing MOV design bases, verifying MOV switch settings initially and periodically, testing MOVs under design-basis conditions where practicable, improving evaluations of MOV failures and necessary corrective action, and trending MOV problems. The NRC staff requested that licensees complete the GL 89-10 program within approximately three refueling outages or 5 years from the date of issuance of the GL. Permit holders were requested to complete the GL 89-10 program before plant startup or in accordance with the above schedule, whichever was later.

On September 18, 1996, (Reference 18), the NRC staff issued GL 96-05, Periodic Verification of Design-Basis Capability of Safety-Related Motor-Operated Valves, requesting each licensee to establish a program, or ensure the effectiveness of its current program, to verify on a periodic basis that safety-related MOVs continue to be capable of performing their safety functions within the current licensing bases of the facility. In GL 96-05, the NRC staff summarized several industry and regulatory activities and programs related to maintaining long-term capability of safety-related MOVs. For example, the NRC staff discussed the use of the EPRI MOV PPM in establishing an effective MOV program. Many licensees are applying the EPRI MOV PPM as part of their programs to provide assurance of the capability of MOVs to perform their safety functions in response to GL 89-10 and GL 96-05.

3.0 TECHNICAL EVALUATION

3.1 Summary of Proposed Revisions Addendum 3 to EPRI TR-103237, Revision 2, provides an improved method to estimate the unwedging thrust required to open a gate valve from the wedged closed position. The original unwedging methodology contained two terms. The first term estimated that portion of the total

unwedging thrust due to static unwedging, and the second term estimated that portion due to the differential pressure load on the disc prior to opening. The revised method continues to use the second term but has redefined the first term. Specifically, the revised method now allows the measured static unwedging thrust to be used rather than predicting the unwedging thrust from the measured static closing thrust. As with the original unwedging thrust model, the revised model does not apply to valves that are susceptible to either pressure locking or thermal binding.

Addendum 4 to EPRI TR-103237, Revision 2, provides a new method to use static gate valve closure data at wedging to determine the stem to stem nut coefficient of friction at unwedging.

The approach was developed based on measured coefficient of friction (COF) values at torque switch trip (TST) for valves that were statically closed at ambient temperature conditions and on measured COF values at unwedging for valves at ambient as well as at elevated temperature.

Based on the evaluation, the COF at valve unwedging was determined to be lower than or comparable to the COF of a valve during a static closure, provided the nominal thread pressure at unwedging under design basis conditions is consistent with the limitation in Addendum 4 to EPRI TR-10327, Revision 2. In effect, this allows a COF determined during the static closure of a valve to be used as the unwedging COF.

Addendum 5 to EPRI TR-103237, Revision 2, describes changes to Version 3.1 of the MOV PPM. The most significant change in this version of the software is that a Windows based user interface replaces the DOS based interface of the previous versions. The EPRI noted that the main modules used to perform the calculations have not been modified, so the resulting thrust and torque calculations would not be affected. The EPRI has also modified the MOV PPM to automatically perform the following tasks:

• determine whether the stem thrust for a gate valve closing stroke will be unpredictable based on selected design features and design basis conditions,

• estimate the unwedging thrust of a gate valve based on the PPM unwedging equation,

• evaluate the effect of both the positive and negative gate valve body rail offset runs and combine the results from both runs into a single report,

• perform predictions for both the opening and closing strokes and combine the results from both runs into a single report,

• determine whether the gate valve guide material has an effect on valve performance and include a message in the prediction report which states that the results are best available information if the guide material is stainless steel and its temperature exceeds 100 ºFahrenheit,

• automatically adjust selected system operating parameters if the user selects the equivalent resistance method,

• automatically adjust selected system operating parameters if the user selects the single point method when evaluating a globe valve,

• calculate the required torque for a self actuating globe valve,

• automatically adjust selected system operating parameters if the user selects the compressible flow method when evaluating a butterfly valve,

• adjust the stem orientation of a gate valve to either vertical or horizontal based on previous EPRI guidance,

• determine the bearing COF for a butterfly valve based on the user specified bearing material and fluid condition,

• determine the seat torque coefficient for a butterfly valve if the user chooses to use the default PPM seat torque prediction, and

• implement PPM Error Notice 97-01 which specifies that for self-actuating globe valve strokes, the required thrust should not be set below the packing load for opening strokes or below the packing plus stem rejection load for closing strokes.

The EPRI also modified the MOV PPM to allow the user the option of inputting selected valve and system data or allowing the MOV PPM to automatically calculate the data. Items that can now be automatically calculated include the upstream, downstream, and critical pipe inertia based on user input piping lengths and diameters; the weight of the gate valve disk; both the positive and negative gate valve body rail offsets; the stroke length of a gate valve based on the seat ring inside diameter; the stem factor for a gate valve based on user input stem pitch, lead, and thread type; the valve inlet diameter; the stroke time based on the overall actuator gear ratio, the motor speed, and (for butterfly valves) the HBC actuator gear ratio; the minimum flow resistance coefficient of the piping, the equivalent valve factor; and the stem added length.

The EPRI also made minor changes to the MOV PPM, such as disabling certain system flow model configurations that were not needed for typical PPM predictions, automating the use of the minimum and maximum design basis temperatures for the positive and negative gate valve body rail offset runs, automating the application of the MOV PPM to AOVs and HOVs by adjusting the water inertia and torque reaction calculations, and automating the evaluation of whether the actuator opening and closing thrust limits are exceeded when evaluating a gate valve.

Addendum 6 to EPRI TR-103237, Revision 2, describes changes to Version 3.2 of the MOV PPM. In general, the software changes address errors found in Version 3.1 of the MOV PPM and minor improvements. For example, the EPRI corrected an error in the MOV PPM pertaining to an upstream disturbance within eight pipe diameters of a butterfly valve (PPM Error Notice 2003-1). The EPRI also eliminated the best estimate butterfly valve torque prediction and replaced it with a design basis torque prediction as a function of disc angle (PPM Information Notice 2002-1). The EPRI also fixed minor interface errors.

Addendum 7 to EPRI TR-103237, Revision 2, describes changes to Version 3.3 of the MOV PPM. This version of the MOV PPM has added fields to the user interface so that gate valves with inverted guides can be evaluated, enhances the parametric prediction option, and corrects a number of PPM Error Notices. These changes are discussed further below:

• In order to model a gate valve with inverted guides, previous versions of the MOV PPM were required to redefine selected input values in order to simulate an inverted guide

gate valve. All input fields required to directly specify the geometry of an inverted guide gate valve have been added to the user interface.

• Versions 1.0 and 2.0 of the MOV PPM allowed the user to perform parametric predictions. Using this feature, the user could input multiple values for a single parameter and multiple runs would be performed in order to help evaluate the effect of a given parameter. This capability was not available in Versions 3.1 and 3.2 of the MOV PPM, but has been included in this release of the software.

• Implements PPM Error Notice 2003-2 by adjusting the total required torque for a given butterfly valve based on the type of disc, the shaft orientation, the disc aspect ratio, and the stroke direction.

• Implements PPM Error Notice 2004-1 by correcting the calculation of the equivalent valve factor. This error affected all globe valve predictions as well as gate valve predictions when the User Input Differential Pressure system option was specified.

• Implements PPM Error Notice 2005-1 by correcting the calculation and reporting of the stem thrust and torque required for an unbalanced disc globe valve closing with flow over the seat or any globe valve opening.

• Implements PPM Error Notice 2005-2 by correcting the guide friction coefficients that were being used and correcting the body rail misalignment calculation for the positive offset case.

• Implements PPM Error Notice 2005-3 by correcting the way the user input data table handled the data for both gate and globe valves.

• Implements PPM Error Notice 2005-4 by correcting the predictions for a butterfly valve opening stroke.

• Implements PPM Error Notice 2005-5 by correcting the names of two input parameters on the disc tab of the gate valve input form and in the output prediction reports.

3.2 Evaluation of the Proposed Changes The NRC staff, with contract assistance from the Pacific Northwest National Laboratory (PNNL),

evaluated the modeling changes to the EPRI MOV PPM software described in the Addenda 3, 4, 5, 6, and 7 and the error and information notices enclosed in both the June 8, 2004, and January 6, 2006, NEI submittals to the NRC. Several valve and system models were developed to assist in the review of the software revisions. The results of selected prediction runs were compared to Version 2.0 of the MOV PPM software as selected system and valve input parameters were adjusted. To support this review, one globe valve model, one butterfly valve model, and two gate valve models were developed.

3.2.1 Addendum 3 to EPRI TR-103237, Revision 2 The EPRI validated the revised unwedging thrust methodology using the same data that was used to validate the original unwedging thrust methodology. In all, data from 18 MOVs were used and consisted of both solid and flexible wedge gate valves. Test data from the 15 feet/second ambient water flow test sequences were used. Each test sequence consisted of a series of opening and closing strokes representing both static and a variety of dynamic flow and differential pressure conditions. The submittal also included additional detail on the validation of the model in MPR Calculation 140-189-JEM-1, Validation of Refined Gate Valve Unwedging Methodology, Revision 0. This calculation presented more details on the evaluation, including a valve by valve summary of the test data and the results of implementing the refined unwedging model. Data included in the calculation included valve size data, measured thrust results, valve differential pressure data, and a comparison of the predicted thrust versus the measured thrust. The predicted and measured thrust values were also graphically compared for each valve in order to show how the refined unwedging model compared to actual test results.

Based on this assessment, the EPRI determined that the method supported the conclusion that the refined unwedging thrust equation more closely but conservatively approximates the test data. The EPRI also noted that since the measured static unwedging thrust is being used, a margin would need to be applied to the measured thrust in order to account for potential increases in the required unwedging thrust due to increases in the disc to seat friction due to time or valve stroking.

The NRC staff requested that EPRI provide additional discussion on selected aspects of the model, such as verifying selected test methods, test sequences, and the resulting test data used to develop the new model; expand their discussion on the condition of the valves that were used to obtain the test data; highlight the limitations of the model relative to valves that may experience pressure locking or thermal binding; and to clarify any limitations or conditions on the test data a plant obtains and how they can use it.

Based on a review of the methods used to analyze the test data and on the responses EPRI provided to NRC staff questions to obtain additional insight into the proposed unwedging thrust model, the NRC staff, with assistance from PNNL, concludes that the new model will allow a plant to conservatively estimate the unwedging thrust of a gate valve; reducing the amount of over thrust prediction inherent in the previous unwedging thrust model. The proposed unwedging thrust model is applicable to both solid and flexible wedge gate valves and is restricted to valves that are not susceptible to pressure locking or thermal binding.

3.2.2 Addendum 4 to EPRI TR-103237, Revision 2 The method is based on the test data from over 30 valves. Measured COF values at TST were obtained for all static closing strokes at ambient temperature test conditions. Only ambient temperature strokes were considered since this data is intended to represent valve static setup test conditions, which are typically performed under ambient temperature conditions. Next, measured COF values at unwedging were obtained for all opening strokes at ambient and high temperatures. Previous evaluations of stem COF data have shown that at low thread pressures, there is significant scatter in the data. To eliminate this data scatter, the test results obtained during low stem thread pressure conditions are not used. The data is then normalized

such that the test results for a given valve opening or closing are weighted to reflect the individual contribution of each valve to the total number of openings or closings. Based on this presentation, the static closing COF for all valves is observed to be greater than the unwedging COF. This result will apply to groups of valves and allows the static closing COF to be used as the unwedging COF. Next, the difference between the static closing and the unwedging COF were determined. Based on this data, if the static closing COF is increased in accordance with Addendum 4 to EPRI TR-10327, Revision 2, the resulting estimated unwedging COF will be bounded nearly 99 percent of the time.

The resulting method can be used in one of two ways provided the nominal thread pressure at unwedging under design basis conditions is consistent with the limitation in Addendum 4 to EPRI TR-10327, Revision 2. Either the method can be applied to a large group of valves or to valves on an individual basis. For a large group of valves, the COF that bounds 95 percent of the data at TST during static testing can be applied at unwedging with no adjustment. For an individual valve, the measured COF at TST for a static test can be increased in accordance with Addendum 4 to EPRI TR-10327, Revision 2, to obtain a COF that is applicable to unwedging.

The NRC staff asked the EPRI to provide additional discussion on selected aspects of the model, such as the effect either elevated temperature valves or elevated temperature ambient conditions would have; the effect of different stem lubricants; and the effect that lubrication aging, drying, or excessive contaminants from the atmosphere would have on the proposed method.

Based on a review of the methods used to analyze the test data and on the responses that the EPRI provided to NRC staff questions to obtain additional insight into the proposed stem to stem nut COF at unwedging model, the NRC staff, with assistance from PNNL, concludes that the new model will allow a plant to conservatively estimate the unwedging COF of a gate valve.

The proposed method is applicable to valves that are either evaluated as part of a larger group, or to valves that are evaluated individually. The proposed method is also applicable to both solid and flexible wedge gate valves.

3.2.3 Addendum 5 to EPRI TR-103237, Revision 2 The NRC staff, with assistance from PNNL, reviewed the supporting material to determine if the changes and additions to Version 3.1 of the MOV PPM were appropriate. Based on this review, the NRC staff asked EPRI to provide additional discussion on the enhancements to the PPM, such as the uncertainties or limitations associated with evaluating air-operated butterfly valves or when evaluating the unwedging thrust of a gate valve.

The NRC staff, with assistance from PNNL, also evaluated Version 3.1 of the PPM using several MOV models and then compared the results of selected problems to the results calculated by Version 2.0 of the PPM software. Base models were developed for both versions to verify that the results from Version 3.1 were similar to the results obtained from Version 2.0.

Based on the additional features that were added to Version 3.1, the base models were subsequently modified in order to verify that selected features of the MOV PPM had been incorporated correctly. Due to the number of enhancements to the software, only selected features of Version 3.1 were evaluated. Those features that were evaluated included error and information notices, the ability of the software to perform calculations that were previously performed by hand, as well as whether any additional limitations on the use or interpretation of

the software are necessary. Based on using Version 3.1 of the MOV PPM, the NRC staff asked EPRI to provide additional discussion on areas where differences in the predictions were encountered.

Based on a review of the responses the EPRI provided to NRC staff questions to obtain additional insight into the new models that were incorporated into Version 3.1 of the PPM software, the NRC staff, with assistance from PNNL, concludes that the new models will allow a plant to conservatively estimate the torque requirements of air-operated butterfly valves or the unwedging thrust of a gate valve. Based on using Version 3.1 of the PPM software, the NRC staff, with assistance from PNNL, concludes that the error and information notices have been successfully incorporated and that the previous hand calculations are being performed correctly.

Additional issues that were encountered during the review of the software include:

• The maximum allowable closing thrust and the actuator opening capability for a gate valve are grayed out, implying that they are no longer required. However, when the prediction is performed, the previously entered values for these two parameters are displayed in the output. This could be very misleading to anyone reviewing the calculations, as well as allowing input variables that may not have been verified (as they were grayed out) to potentially be used by others. The EPRI agrees that displaying these values in the prediction report for a closing stroke only evaluation could be misleading to users. Accordingly, the EPRI will issue an Information Notice to software users regarding this issue.

• The PPM software appears to display the correct calculated value for the gate valve body rail misalignment calculation in the input section after the prediction has been run; however, the value displayed in the report has reversed the sign of the positive offset term. The EPRI agrees with this observation and will issue an Information Notice to software users for Version 3.1 of the PPM software regarding the sign of the body rail misalignment (positive offset) in the prediction report.

• Implementation of the inverted guide model requires the user to replace selected conventional valve parameters with inverted guide parameter values. The parameters that need to be revised include the body rail angular misalignment (degree), the disk slot angular misalignment (degree), and the disk slot misalignment (inches). While all the parameters that were required to be revised were found in Version 2.0 of the PPM software, these three parameters could not be found in Version 3.1 of the PPM. The EPRI agrees that Version 3.1 does not provide a means for users to adjust the input parameters for analyzing gate valves with inverted guides, and that the method cannot be implemented in Version 3.1. The EPRI will issue a PPM Software Error Notice, clarifying that the inverted guide methodology cannot be implemented in Version 3.1 of the PPM software.

3.2.4 Addendum 6 to EPRI TR-103237, Revision 2 The NRC staff, with assistance from PNNL, reviewed the supporting material to determine if the changes and additions to Version 3.2 of the MOV PPM were appropriate. Based on this review, NRC staff asked EPRI to provide additional discussion on the enhancements to the PPM, such as the uncertainties or limitations associated with evaluating air-operated butterfly valves.

The NRC staff, with assistance from PNNL, also evaluated Version 3.2 of the PPM using several MOV models and then compared the results of selected problems to the results calculated by Versions 2.0 and 3.1 of the PPM software. Base models were developed to ensure that the results from Version 3.2 were similar to the results obtained from earlier versions. Only selected features of Version 3.2 were evaluated. Those features that were evaluated include error and information notices; the revised torque prediction method for a butterfly valve; as well as whether any additional limitations on the use or interpretations of the software are necessary. Based on using Version 3.2 of the MOV PPM, the NRC staff asked EPRI to provide additional discussion on areas where differences in the predictions were encountered.

Based on a review of the responses the EPRI provided to questions that were asked to obtain additional insight into the revised models that were incorporated into Version 3.2 of the PPM software, the NRC staff concludes, with assistance from PNNL, that the revised models will allow a plant to conservatively estimate the torque requirements of air-operated butterfly valves.

Based on using Version 3.2 of the PPM software, the NRC staff concludes that the error and information notices have been successfully incorporated.

Additional issues that were encountered during the review of the software include:

• The maximum allowable closing thrust and the actuator opening capability for a gate valve are grayed out, implying that they are no longer required. However, when the prediction is performed, the previously entered values for these two parameters are displayed in the output. This could be very misleading to anyone reviewing the calculations, as well as allowing input variables that may not have been verified (as they were grayed out) to potentially be used by others. EPRI agrees that displaying these values in the prediction report for a closing stroke only evaluation could be misleading to users. Accordingly, the EPRI will issue an Information Notice to software users regarding this issue.

• Implementation of the inverted guide model requires the user to replace selected conventional valve parameters with inverted guide parameter values. The parameters that need to be revised include the body rail angular misalignment (degree), the disk slot angular misalignment (degree), and the disk slot misalignment (inches). While all the parameters that were required to be revised were found in Version 2.0 of the PPM software, these three parameters could not be found in Version 3.2 of the PPM. EPRI agrees that Version 3.2 does not provide a means for users to adjust the input parameters for analyzing gate valves with inverted guides, and that the method cannot be implemented in Version 3.2. The EPRI will issue a PPM Software Error Notice, clarifying that the inverted guide methodology cannot be implemented in Version 3.2 of the PPM software.

3.2.5 Addendum 7 to EPRI TR-103237, Revision 2 The NRC staff, with assistance from PNNL, reviewed the above addendum and the supporting material to determine if the changes and additions to Version 3.3 of the MOV PPM were appropriate. Based on this review, the NRC staff asked EPRI to provide additional discussion on the enhancements to the PPM, such as the uncertainties or limitations associated with evaluating gate valves with inverted guides. The NRC staff, with assistance from PNNL, also

evaluated Version 3.3 of the PPM using several MOV models and then compared the results of selected problems to the results calculated by Versions 2.0, 3.1, and 3.2 of the PPM software.

Base models were developed to verify that the results from Version 3.3 were similar to the results obtained from the earlier versions. Based on the additional features that were added to Version 3.3, the base models were subsequently modified in order to verify that selected features of the MOV PPM had been incorporated correctly. Due to the number of enhancements to the software, only selected features of Version 3.3 were evaluated. Those features that were evaluated included error and information notices, the ability of the software to model a gate valve with inverted guides, as well as whether any additional limitations on the use or interpretation of the software are necessary. Based on using Version 3.3 of the MOV PPM, the NRC staff asked EPRI to provide additional discussion on areas where differences in the predictions were encountered.

During its review of Version 3.3, the NRC staff, with assistance from PNNL, noted that the maximum allowable closing thrust and the actuator opening capability for a gate valve are grayed out, implying that they are no longer required. However, when the prediction is performed, the previously entered values for these two parameters are displayed in the output.

This could be very misleading to anyone reviewing the calculations, as well as allowing input variables that may not have been verified (as they were grayed out) to potentially be used by others. The EPRI agrees that displaying these values in the prediction report for a closing stroke only evaluation could be misleading to users. Accordingly, the EPRI will issue an Information Notice to software users regarding this issue.

Based on a review of the responses the EPRI provided to questions that were asked to obtain additional insight into the new models that were incorporated into Version 3.3 of the PPM software, the NRC staff, with assistance from PNNL, concludes that the new models will allow a plant to conservatively estimate the thrust requirements of a gate valve with inverted guides.

Based on using Version 3.3 of the PPM software, the NRC staff concludes, with assistance from PNNL, that the error and information notices have been successfully incorporated and that the inverted guide gate valve model is performing correctly.

4.0 CONCLUSION

Based on the review of Addenda 3, 4, 5, 6, and 7 to EPRI TR-103237-R2, the NRC staff concludes, with assistance from PNNL, that the improved gate valve unwedging methodology, the improved stem to stem nut COF methodology, and the changes made to Versions 3.1, 3.2, and 3.3 of the EPRI MOV PPM, improve the ability of licensees to predict the thrust and torque required to operate gate, globe, and butterfly valves and that they are acceptable for referencing by licensees as delineated in Addenda 3, 4, 5, 6, and 7 to EPRI TR-103237-R2.

5.0 REFERENCES

1. EPRI TR-103237-P, Revision 0, EPRI MOV Performance Prediction Program, February 1994 (ADAMS Accession No. 9411160332).

2. EPRI TR-103237-NP, Revision 1, EPRI MOV Performance Prediction Program, November 30, 1995 (ADAMS Accession No. 9604030085).

3. USNRC Safety Evaluation on Electric Power Research Institute Topical Report TR-103237-R1, EPRI Motor-Operated Valve Performance Prediction Program, March 15, 1996 (ADAMS Accession No. 9608070288)

4. Supplement 1 to USNRC Safety Evaluation on Electric Power Research Institute Topical Report TR-103237-R1, EPRI Motor-Operated Valve Performance Prediction Program, February 20,1997 (ADAMS Accession No. 9704300116).

5. Addendum 1, PPM Version 2.0, and Addendum 2, Thrust Uncertainty Method, to EPRI TR-103237, Revision 2, September 8, 1999 (ADAMS Accession No. 9909280135).

6. Supplement 2 to USNRC Safety Evaluation on Electric Power Research Institute Topical Report TR-103237-R2, EPRI Motor-Operated Valve Performance Prediction Program, April 20, 2001 (ADAMS Accession No. ML011100121).

7. Letter from David Modeen, Nuclear Energy Institute, to the US NRC, Response to Questions on Addendum 2 to EPRI Performance Prediction Methodology Software, Version 2.0 to EPRI TR-103237, January 5, 2001 (ADAMS Accession No. ML012950349).

8. Letter from Alexander Marion, Nuclear Energy Institute, to the US NRC, Addendum to Responses to Questions on Addendum 2 to EPRI Performance Prediction Program Topical Report TR-103237-R2 Thrust Uncertainty Method, December 6, 2001 (ADAMS Accession No. ML020160518).

9. Letter from Alexander Marion, Nuclear Energy Institute, to the US NRC, Additional Responses to Questions on EPRI Report 110779 "Addendum 2 to EPRI TR-103237-R2 Thrust Uncertainty Method," June 10, 2002 (ADAMS Accession No. ML021830330).

10. Letter from William H. Ruland, US NRC, to Gary L. Vine, Nuclear Energy Institute, Supplement 3 to Safety Evaluation on Electric Power Research Institute Topical Report TR-103237-R2, EPRI Motor-Operated Valve Performance Prediction Program, Addendum 2, September 30, 2002 (ADAMS Accession No. ML022410364).

11. Letter from Alexander Marion, Nuclear Energy Institute, to the US NRC, Transmittal of EPRI MOV Performance Prediction Methodology Software, June 8, 2004 (ADAMS Accession No. ML041700093).

12. Letter from Alexander Marion to the US NRC, Transmittal of EPRI MOV Performance Prediction Methodology Software, January 6, 2006 (ADAMS Accession No. ML0600604).

13. Letter from Tanya M. Mensah, US NRC, to James H. Riley, Nuclear Energy Institute Request for Additional Information Regarding Topical Report (TR)-103237, EPRI MOV Performance Prediction Program - Topical Report, April 23, 2007 (ADAMS Accession No. ML071070370).

14. Letter from Vanice Perin, US NRC, to James H. Riley, Nuclear Energy Institute, Request for Additional Information Regarding Topical Report (TR)-103237, EPRI MOV Performance Prediction Program - Topical Report, June 3, 2008 (ADAMS Accession No. ML081130133).

15. Letter from James H. Riley, Nuclear Energy Institute, to Tanya M. Mensah, US NRC EPRI TR MOV Performance Prediction Program - Topical Report, Response to Request for Additional Information, September 27, 2007 (ADAMS Accession Nos.

ML072740037 and ML072740038).

16. Letter from Michael A. Melton, Nuclear Energy Institute to US NRC, EPRI TR MOV Performance Prediction Program - Topical Report, Response to Request for Additional Information, August 7, 2008 (ADAMS Accession Nos. ML082980390 and ML082980392).

17. Generic Letter 89-10, "Safety-Related Motor-Operated Valve Testing and Surveillance,"

June 28, 1989.

18. Generic Letter 96-05, Periodic Verification of Design-Basis Capability of Safety-Related Motor-Operated Valves, September 18, 1996.

Attachment:

Resolution of Comments Principle Contributor: Steve Tingen Date: February 24, 2009

RESOLUTION OF COMMENTS ON SUPPLEMENT 4 TO DRAFT SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION ADDENDA 3, 4, 5, 6, AND 7 TO ELECTRIC POWER RESEARCH INSTITUTE (EPRI)

TOPICAL REPORT (TR)-103237, EPRI MOV PERFORMANCE PREDICTION PROGRAM, REVISION 2 NUCLEAR ENERGY INSTITUTE (NEI)

PROJECT NO. 689 By email dated January 30, 2009, the NEI did not identify any factual errors or clarifications that needed to be made to the draft SE. However, they requested that several deletions be made from the draft SE because of the proprietary nature of the information.

NRC Staff Response:

The NRC staff reworded several sentences within the draft SE to remove the proprietary information.

ATTACHMENT