ML20115D762

Provides Response to Request for Addl Info Re GL 95-07, Pressure Locking & Thermal Binding of Safety-Related Power-Operated Gate Valves

ML20115D762
Person / Time
Site:	Hope Creek
Issue date:	07/10/1996
From:	Eric Simpson Public Service Enterprise Group
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-95-07, GL-95-7, LR-N96179, NUDOCS 9607150270
Download: ML20115D762 (11)

Text

_ _ _ . _. _

i Public Service Electric and Gas Company E. C. Simpson Public Service Electric and P.O. Box 236, Hancocks Bridge, NJ 08038 609-339-1700 sener vice Preusent Nuclear Engensonng u 10 y[Companf996 l LR-N96179 United States Nuclear Regulatory Commission l Document Control Desk l Washington, D.C. 20555 Gentlemen:

RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION l GENERIC LETTER 95-07 " PRESSURE LOCKING AND THERMAL BINDING OF SAFETY-RELATED POWER-OPERATED GATE VALVES" HOPE CREEK GENERATING STATION FACILITY OPERATING LICENSE NPF-57 DOCKET NO. 50-354 This letter provides the response of Public Service Electric and Gas Company (PSE&G) to the request from the NRC for additional information concerning PSE&G's 180 day response to Generic Letter 95-07. The request for additional information (RAI) was received by PSE&G on May 25, 1996 in a letter dated May 17, 1996 and i requested a response within 30 days of receipt. In a subsequent i discussion with the NRC Project Manager, the requested response date was extended to July 10, 1996.

l The attachment to this letter provides a detailed response to the questions contained in the RAI. Should you have any questions or comments on this transmittal, do not hesitate to contact us.

Sincerely, l

l 1

l Attachment

/06b i 9607150270 960710 PDR ADOCK 05000354 P PDR

@ Printedon Recycled Paper

l Document Control Desk JUL 101996 LR-N96179 C Mr. T. T. Martin, Administrator - Region I U. S. Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA 19406 Mr. D. Jaffe, Licensing Project Manager - Hope Creek U. S. Nuclear Regulatory Commission One White Flint North 11555 Rockville Pike Mail Stop 14E21 Rockville, MD 20852 Mr. R. Summers USNRC Senior Resident Inspector (X24)

Mr. K. Tosch, Manager IV Bureau of Nuclear Engineering 33 Arctic Parkway CN 415 Trenton, NJ 08625 1

95-4933

ATTACHMENT RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION GENERIC LETTER 95-07 HOPE CREEK GENERATING STATION DOCKET NO. 50-354 LR-N96179 l l

l l

I. INTRODUCTION l 1

On February 13, 1996, Public Service Electric and Gas Company (PSE&G) submitted its 180-day response to Generic Letter 95-07 in Letter LR-N96034. The NRC subsequently requested additional information regarding PSE&G's 180 day response in a letter dated May 17, 1996. This attachment provides PSE&G's response to the NRC's request for additional information.

II. RESPONSE TO REOUEST FOR ADDITIONAL INFORMATION The following provides a verbatim quotation of each NRC question along with the associated PSE&G response.

OUESTION 1 "Regarding valve 1FDHV-F001, High Pressure Coolant Injection Turbine Steam Admission, the licensee's submittal states that 1FDHV-F002 and 1FDHV-F003 are cycled for surveillance testing purposes during cold shutdown and that no thermal binding of these valves has been indicated over a broader temperature decrease than applicable to F001. Have these surveillances been performed using diagnostic testing equipment? If so, (1) have these tests indicated an increase in thrust requirement after the valve was shut in a hot condition and later opened following cooldown, and (2) has the licensee performed any analytical calculations to compare this thrust requirement to the actuator capability?

If these analyses or evaluations have been completed, please provide them for our review. In addition, the staff requests further justification regarding the reliance on performance history of valves F002 and F003 as a basis for the resolution of thermal binding concerns for F001."

RESPONSE 1 We have not performed as-found thrust testing of the HPCI system 1FDHV-F002 and 1FDHV-F003 steam supply isolation valves with diagnostic equipment to determine what, if any, increment of added thrust might occur as a result of closing 1 of 9

Attachment LR-N96179 Response to GL 95-07 RAI the valves hot and opening them in a cooled state.

A thernal binding analysis has been performed for Valve 1FDHV-F001 that provides a basis for concluding that the valve will satisfy its design basis requirement to open with the thermal binding influence. These calculations are discussed below. In addition, further details concerning  ;

the comparison of Valve 1FDHV-F001 with Valves 1FDHV-F002  !

and 1FDHV-F003 are provided.

At DISCUSSION OF THERMAL BINDING ANALYSIS FOR 1FDHV-F001 A thermal binding analysis has been performed for 1FDHV-F001 using recent test data for the subject valve, data from the EPRI MOV Performance Prediction Program, and a thermal binding nodel developed by MPR Associates. The recent test data for 1FDHV-F001 that was used in the analysis includes post maintenance diagnostic testing completed following internal maintenance that was performed during RFO6 and ,

additional diagnostic testing performed as part of the HPCI system surveillance on March 20, 1996. The latter test included monitoring motor current and worm displacement during an opening stroke at a differential pressure of 922 psig and at saturated steam conditions (~537*F). Both 1FDHV-F001 and the EPRI Program's Valve #5 are Anchor Darling 10 inch - 900 lb. ANSI class flex wedge gate valves, and data for EPRI Valve #5 were used to provide a determination of valve flexibility associated with temperature induced stem growth. Valve body / disc l contraction and stem growth effects were determined using a thermal binding model that was developed by MPR Associates. ,

The thermal binding model considers the most extreme 1 operating temperature effects for the HPCI turbine steam l supply and incorporates the maximum closing thrust value ,

from the post-maintenance test under ambient static conditions. j i

The most conservative bounding analysis indicates that the l valve has a margin of negative 17%. On a best estimate l basis, the analysis model predicts that the available thrust {

exceeds the required thrust with a positive margin of 72%. I The best estimate analysis still contains conservatisms that could be removed to provide additional positive margin.

l The primary differences between the bounding and best estimate analyses are the assumptions for disc relaxation and stem coefficient of friction. Disc relaxation relates to the required unwedging thrust and the stem coefficient 2 of 9

I.

Attachment LR-N96179 Response to GL 95-07 RAI relates to actuator available thrust. Accounting for disc relaxation contributes a 32% improvement to the required thrust value when compared to the bounding value.

Accounting for test performance based stem friction as opposed to a bounding value yields a 41% improvement in analyzed actuator capability. Either best estimate contributor alone assures a positive margin of capability to overcome thermal binding.

It is acknowledged that the best estimate margin can degrade with time; however, the following provides a basis for concluding that the degradation would not be significant.

Based upon experience, the stem friction is the more susceptible of the two primary factors to degradation. The amount of stem friction degradation is largely dependent on the stem loading, the conditions to which the lubricant is exposed, and the frequency of relubrication. This valve is not so significantly loaded that the stem threads would experience wear. Additionally, the valve is cycled once per quarter in compliance with inservice testing requirements for the HPCI system. Periodic stroking in this manner, besides demonstrating valve performance, causes redistribution of the lubricant on the stem thread that renews the performance. Recurring tasks also result in stem lubrication on a nominal 36 month interval.

With no credit taken for disc relaxation, the stem coefficient could potentially degrade to 0.15 from the current as-left value of 0.10 before the margin would be reduced to zero. Although we believe that some degradation of the factors credited in the best estimate analysis is possible, the degradation will not be to the point that positive margin of capability would be lost to overcome the potential thermal binding influence.

PSE&G therefore concludes that the valve will satisfy its desian basis reauirement to open with the thermal bindina influence.

B2 FURTHER DETAILS ON COMPARISON OF VALVE 1FDHV-F001 WITH VALVES 1FDRV-F002 AND 1FDHV-F003 In our previous submittal, we indicated that extreme hot to cool temperature conditions for the 1FDHV-F002 and 1FDHV-F003 valves formed a basis for demonstrating operability of the 1FDHV-F001 HPCI turbine steam admission valve. The 1FDHV-F002 and 1FDHV-F003 valves are closed at approximately 350 F when cooling the plant. Refueling outage surveillance 3 of 9

Attachment LR-N96179 Response to GL 95-07 RAI testing of Valves 1FDHV-F002 and 1FDHV-F003 is performed at approximately 100'F. This is a temperature differential of 250*F. The nominal 160*F temperature differential for the 1FDHV-F001 valve is based on the saturated properties of steam at the nominal HPCI system steam supply pressure of 1000 psig and the minimum HPCI design operating pressure of 200 psig, 546 F and 388'F, respectively. All three valves are Anchor Darling 10 inch - 900 lb. ANSI class carbon steel flexible wedge gate valves with Limitorque size SMB-1 motor actuators. The valve internals, the stem thread configuration and actuator gearing are identical in design.

It is reasonable to conclude that valve similarity, the bounding thermal conditions, and the satisfactory performance of Valves 1FDHV-F002 and 1FDHV-F003 when surveillance tested cold is a sufficient test based indicator that the less severe differential thermal conditions for Valve 1FDHV-F001 are bounded in terms of valve differential contraction characteristics.

QUESTION 2 "Regarding valve 1BDHV-F031, Reactor Core Isolation Cooling Pump Suction From Suppression Pool, the licensee's submittal states that this valve strokes open at least three hours after an isolation event. Has the licensee evaluated the potential for heat transfer from the suppression pool to cause thermally-induced pressure locking of this valve? If so, please provide this evaluation for our review. If not, please provide a basis fer omitting this evaluation."

RESPONSE 2 We have considered the potential for heat transfer from the suppression pool to Valve 1BDHV-F031. It has been determined that a significant insulated horizontal length of 6 inch Schedule 40 piping in excess of 38 feet in length exists between the suppression pool connection and the valve. Comparison of the minimum suppression pool temperature of 60*F to its maximum of 140 F for the band of RCIC operation results in a relatively low differential temperature increase. In consideration of the long length of horizontal piping, analysis indicates that a significant temperature drop occurs within the first five feet from the source and the temperature at 10 feet decays to room ambient. At the valve location, there is no temperature increase because of its 38 foot pipe length distance from the source. Therefore, we conclude that the valve is not susceptible to thermally induced pressure locking.

4 of 9

Attachment LR-N96179 Response to GL 95-07 RAI In addition to this information, our Generic Letter 89-10 analysis indicates that Valve 1BDHV-F031 has significant motor operator opening direction output capability. Based on the most limiting component part of the valve / operator i assembly under design basis operating conditions, the valve i factor equivalent capability is 5.05 in the open direction.

This provides assurance that any internal valve pressure can be overcome by the operator and the valve will satisfy its safety function to open.

OUESTION 3 "In Attachment 1 to GL 95-07, the NRC staff requested that licensees include consideration of the potential for gate valves to undergo pressure locking or thermal binding during surveillance testing. During workshops on GL 95-07 in each Region, the NRC staff stated that, if closing a safety-related power-operated gate valve for test or surveillance defeats the capability of the safety system or train, the licensee should perform one of the following within the scope of GL 95-07:

1. Verify that the valve is not susceptible to pressure locking or thermal binding while closed, i

2. Follow plant Technical Specifications for the train / system while the valve is closed, j

3. Demonstrate that the actuator has sufficient capacity to overcome these phenomena, or

4. Make appropriate hardware and/or procedural modifications to prevent pressure locking and thermal binding.

The staff stated that normally open, safety-related power-operated gate valves which are closed for test or surveillance but must return to the open position should be evaluated within the scope of GL 95-07. Please discuss if valves which meet this criterion were included in your review, and how potential pressure locking or thermal binding concerns were addressed."

RESPONSE 3 PSE&G has identified the population of normally open, safety-related power-operated gate valves that are closed for test or surveillance but must be returned to the open 5 of 9

.- .. 1 Attachment LR-N96179 Response to GL 95-07 RAI position. In accordance with plant procedures, a portion of this population of valves is required to be declared inoperable along with the associated system since the surveillance test renders the system incapable of performing its intended function. These valves meet Option 2 identified at the Generic Letter 95-07 workshops. The balance of the valves is not declared inoperable because they automatically re-position to their required position on a system initiation signal. This latter population of valves was evaluated and further reduced by eliminating the valves that undergo only tests that are of such short duration (less than a few minutes) that pressure locking and thermal binding are not an issue of concern. These valves therefore meet the requirements of Option 1 from the Generic Letter 95-07 workshops. The only valves remaining as a result of the above described screening process are the four RHR pump minflow valves (1BCHV-F007A, B, C, and D).

Periodic testing of the RHR pump minflow valves are performed on a quarterly frequency as required by Technical Specification 4.0.5. During this testing, the valves are placed in the close position for only a short time (less than a few minutes), and development of the thermal binding or pressure locking condition during this short duration test is not credible. This portion of the surveillance testing of the minflow valves therefore meets the requirements of Option 1 from the Generic Letter 95-07 workshops.

The minflow valves are also automatically stroked to the closed position during periodic testing or operation of the respective RHR pump. In addition, the A and B valves stroke closed to support suppression pool cooling during parallel testing of the RCIC and HPCI systems. During these evolutions, the normally open minflow valves will automatically close when sufficient flow for pump cooling is assured. The valve will remain closed until the operator reduces system flow just prior to removing the pump from service. Opening takes place at the same nominal pressure conditions as closing. Our evaluation of susceptibility of the subject valves to thermal binding and thermally or hydraulically induced pressure locking is described below.

A description of valve capability is also provided along with the performance history of the valves during surveillance testing.

6 of 9

I i

I Attachment LR-N96179 l Response to GL 95-07 RAI l THERMALLY INDUCED PRESSURE LOCKING I A review of the minflow piping configuration for each of the i

RHR loops shows the valves to be located a minimum of 48 feet from the main line of the loop. Two of the minflow

lines (C and D) begin with a vertical drop in elevation from the main trunk line. The A and B minflow lines begin with a i

minimum 21 foot horizontal run of piping. In these instances, there is no fluid thermal conduction transmitted to the valves after they are closed. The valves may be subject to a high ambient temperature associated with a HELB in the torus room; however, RHR is not required to respond l to a HELB in the torus room. Therefore, thermally induced j pressure locking is not a concern. .

i l

HYDRAUI,ICALLY INDUCED PRESSURE LOCKING l l In that the pressure associated with automated opening of l the minflow valve corresponds with that for its closing, no basis exists for hydraulically induced pressure locking. l THERMAL BINDING l l

The pump is not expected to be deadheaded during its I operation and in consideration that closing and opening of the minflow valve is associated with establishing a flow loop, RHR line pressure is expected to decline after the valve is closed. Application of our susceptibility criteria for thermal binding has determined that these valves are not susceptible to thermal binding due to heating under ambient conditions after valve closure.

VALVE CAPABILITY DISCUSSION l Of the four minflow valves, the 1BCHV-F007B valve has the most limiting design basis opening capability based on degraded voltage limits. Through differential pressure testing performed in response to Generic Letter.89-10 for Valve 1BCHV-F007B, a 42% positive margin of capability was demonstrated. The other minflow valve tests indicated j margins in excess of this level. Testing was performed at 85% of the design basis differential pressure. The capability margin was determined by comparing an l extrapolated value of peak opening thrust to the maximum available thrust. The maximum available thrust is based on a bounding value of stem friction established at 0.20.

7 of 9

2 J

Attachment LR-N96179

Response to GL 95-07 RAI i Based on torque and thrust measurements during the test, a l i stem friction of less than 0.10 was indicated. On the basis

of the measured stem friction, margin of capability for the ,

l differential pressure condition would be in excess of 100%.  !

j. PERFORMANCE HISTORY DURING TESTING l

Based upon a review of previous work order history for each i -RHR minflow valve, no previous failure of a valve to reopen has been reported and therefore no apparent susceptibility to any kind of binding is exhibited. ]

CONCLUSION I

}' On the basis of the above reviews, these valves are  !

considered operable and judged not to be susceptible to j pressure locking or' thermal binding. Therefore, this '

portion of the surveillance testing of the minflow valves

! also meets the requirements of Option 1 from the Generic

Letter 95-07 workshops.

f i

QUESTION 4 l

"Through review of operational experience feedback, the NRC 3

staff is aware of instances where licensees have completed design or procedural modifications to preclude pressure e

locking or thermal binding which may have had an adverse impact on plant safety due to incomplete or incorrect

evaluation of the potential effects of these modifications.

Please describe evaluations and training for plant personnel that have been conducted for each design or procedural modification completed to address potential pressure locking j or thermal binding concerns."

! RESPONSE 4 i

) No design modifications have been implemented at Hope Creek j to address gate valve thermal binding.

} Design modifications performed at Hope Creek to preclude pressure locking entailed the installation of a 1/8 inch L diameter weep-hole on the high pressure side of the gate valve flex-wedge. This method for providing a pressure release path was selected based on industry experience, its I technical merits and its simplicity. For all cases, the i hole was placed in the normal higher pressure side of the 8 of 9 i

Attachment LR-N96179 >

Response to GL 95-07 RAI valve disc in accordance with a design provided by the valve manufacturer. The valve manufacturer's design reports were reconciled in consideration of the design change.

Valve specific evaluations were performed with respect to valve and system function and impact of potential for in-leakage to the reactor from the isolated system and the ability of make-up systems to bound this loss during normal plant operation and system surveillance testing as applicable. In-leakage volumes were conservatively determined based on the weep-hole being the limiting flow orifice with no pressure drop across the upstream seating surfaces. No adverse impacts on plant safety were identified for design basis conditions. These reviews have been documented for each valve that has been modified. ,

No operational procedure changes were implemented to resolve potential pressure locking or thermal binding concerns for any of the Hope Creek valves. Reference design documents and appropriate procedures have been revised to reflect the design changes for configuration control and valve maintenance purposes.

Training has been conducted for plant licensed operators as part of requalification training regarding the nature of the i modifications installed, which valves were affected and the potential leakage affects due to the modifications to address potential pressure locking concerns.

9 of 9 e y- -e- , + , - - ~ . - - - - -

r