ML18102A836

Application for Amends to Licenses DPR-70 & DPR-75,revising TS 3/4.4.3 to Ensure That Automatic Capability of Power Operated Relief Valve to Relieve Pressure Is Maintained When Valves Are Isolated by Closure of Block Valves

ML18102A836
Person / Time
Site:	Salem
Issue date:	01/31/1997
From:	Eric Simpson Public Service Enterprise Group
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML18102A837	List: ML18102A836 ML18102A838
References
LCR-S97-03, LCR-S97-3, LR-N97055, NUDOCS 9702190049
Download: ML18102A836 (20)

Text

Public Service i' - I\,_ Electric and Gas 1  : Company E. C. Simpson Public Service Electric and Gas Company P.O. Box 236, Hancocks Bridge, NJ 08038 609-339-1700 Senior Vice President - Nuclear Engineering JAN 311997 LR-N97055 LCR S97-03 United States Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 EXIGENT REQUEST FOR CHANGE TO TECHNICAL SPECIFICATIONS PRESSURIZER POWER OPERATED RELIEF VALVES SALEM GENERATING STATION NOS. 1 AND 2 FACILITY OPERATING LICENSES DPR-70 AND DPR-75 DOCKET NOS. 50-272 AND 50-311 Gentlemen:

In accordance with 10CFR50.90, Public_ Service Electric & Gas (PSE&G) Company requests a revision to the Technical Specifications (TS) for the Salem Generating Station Unit Nos. 1 and 2. In accordance with 10CFR50.91(b) (1), a copy of this submittal has been sent to the State of New Jersey.

The proposed TS changes contained herein represent changes to the Reactor Coolant System Power Operated Relief Valve (PORV) TS to ensure that the automatic capability of the PORV to relieve pressure is maintained when these valves are isolated by closure of the block valves. This change is based on PSE&G's review of a spurious operation of the Safety Injection System at power.

The proposed changes have been evaluated in accordance with 10CFR50.91(a) (1), using the criteria in 10CFR50.92(c), and PSE&G has concluded that this request involves no significant hazards considerations.

The basis for the requested change is provided in Attachment 1.

A 10CFR50.92 evaluation with a determination of no significant hazards consideration is- provided in Attachment 2. The marked up TS pages affected by the proposed changes are provided in Attachment 3. -

(----9702190049 970131 -----~

I PDR ADOCK 05000272 1snnB"~ ~

1 I

Aool '/I/

\ p PDR

~ Printedon

~ Recycled Paper

UAN 311997 Document Control Desk LR-N97055 Upon NRC approval of this proposed change, PSE&G requests that the amendment be made effective on the date of issuance, but provide for implementation prior to entry into Mode 3 from the current outages for Salem Units 1 and 2. Because this change was identified recently and is needed prior-to entry into Mode 3 on Salem Unit 2, PSE&G is requesting that this request be processed on an exigent basis.

Should you have any questions regarding this request, we will be pleased to discuss them with you.

Affidavit Attachments (3)

C Mr. H. J. Miller, Administrator - Region I

u. s. Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA 19406 Mr. L. Olshan, Licensing Project Manager - Salem

u. s. Nuclear Regulatory Commission One White Flint North 11555 Rockville Pike Mail Stop 14E21 Rockville, MD 20852 Mr. c. Marschall (X24)

USNRC Senior Resident Inspector - Salem Mr. K. Tosch, Manager IV Bureau of Nuclear Engineering 33 Arctic Parkway CN 415 Trenton, NJ 08625 95-4933

JAN 311997 Document Control Desk LR-N97055 SRM/mrh BC Senior Vice President - Nuclear Engineering (N19)

General Manager - Salem Operations (S05)

Director - QA/NSR (XOl)

Manager - Joint Owners/Ext Aff Interface (N28)

Manager - Salem Operations (SOl)

Manager - System Engineering - Salem (S02)

Manager - Nuclear Safety Review (N38)

Manager - Licensing & Regulation (X09)

Principal Engineer [Salem] Operational Licensing (X09)

Onsite Safety Review Engineer - Salem (X15)

S. Miranda G. Schwartz M. Danak W. Chrowrnanski General Solicitor, E. Selover (Newark, 5G)

Perry Robinson, Esq.

Records Management (N21)

Microfilm Copy Files Nos. 1.2.1 (Salem), 2.3 (LCR S97-03)

-~ .

REF: LR-N97055 LCR S97-03 STATE OF NEW JERSEY )

) SS.

COUNTY OF SALEM )

E. c. Simpson, being duly sworn according to law deposes and says:

I am Senior Vice President - Nuclear Engineering of Public Service Electric and Gas Company, and as such, I find the matters set forth in the above referenced letter, concerning Salem Generating Station, Units 1 and 2, are true to the best of my knowledge, information and belief.

My Commission expires on 03* e-2000

Document Control LR-N97055 Attachment 1 LCR S97-03 SALEM GENERATING STATION UNIT NOS. 1 AND 2 FACILITY OPERATING I1ICENSES DPR-70 AND DPR-75 DOCKET NOS. 50-272 AND 50-311 CHANGE TO TECHNICAL SPECIFICATIONS PRESSURIZER POWER OPERATED RELIEF VALVES BASIS FOR REQUESTED CHANGE REQUESTED CHANGE AND PURPOSE The proposed Technical Specification (TS) changes contained herein represent changes to Specification 3.4.5 pertaining to the actions associated with an inoperable pressurizer Power Operated Relief Valve(s) (PORV). The current TS actions for continued plant operation with an inoperable PORV state that the valve must be isolated by its associated block valve and also capable of being manually cycled (e.g., remotely from the control room).

The proposed change eliminates the reference to manual operation and adds the statement permitting continued plant operation with an inoperable PORV due to excessive seat leakage only.

In addition, the applicable Bases section is being revised to include clarification of the design functions of the PORVs and block valves.

This change is proposed based upon PSE&G's review of the impact of an Inadvertent Safety Injection (SI) event and the resulting necessity to have the automatic function of the PORVs to reduce challenges to the pressurizer safety relief valves for this event. Since manual operation of the PORVs is also necessary to mitigate the consequences of a Steam Generator Tube Rupture (SGTR) accident, PSE&G is adopting the recommended TS proposed in Generic Letter (GL) 90-06.

BACKGROUND In June of 1990, the NRC issued GL 90-06 entitled "Resolution of Generic Issue 70, 'Power-Operated Valve and Block Valve Reliability,' and Generic Issue 94 'Additional Low-Temperature Overpressurization Protection For Light-Water Reactors. This GL was issued to increase the reliability of the PORVs and block valves to assure that they would function as required for certain transients and accidents including SGTR, low temperature overpressurization protection, and plant cooldown. One of the actions required by the GL was to revise the limiting conditions for operation (LCO) of the PORVs and block valves in TS.

PSE&G complied by submitting a request to change TS, (ref. letter NLR-N93163 dated December 8, 1993) which was incorporated in Salem Unit 1 and 2 licenses via Amendments 150 and 130, respectively. The submitted request and amendments were based on Page 1 of 11

Document Control LR-N97055 LCR S97-03 the' guidance provided in the GL and also later revisions that were made to the LCO under NUREG-1431, "Standard Technical Specifications Westinghouse Plants," Rev. 0, dated September 1992. One of the changes afforded by the NUREG was to allow PORV isolation provided the PORV is capable of manual operation based on the mitigation of a Steam Generator Tube Rupture accident; whereas, the TSs recommended in GL 90-06 addressed isolation only for valves with excessive seat leakage.

In June of 1993, Westinghouse issued Nuclear Safety Advisory letter, NSAL 93-013, which addressed the Inadvertent SI Actuation at Power event and informed plants that potential nonconservative assumptions were used in evaluating the Inadvertent SI analyses.

Westinghouse determined that crediting PORV operation could be a potential solution for the mitigation of this event. The spurious operation of the SI System at power is classified as a Condition II event, a fault of moderate frequency, as referenced in Salem's UFSAR Section 15.2.14. A Condition II event should result in a reactor shutdown with the plant being capable of returning to operation.

PSE&G has determined that an inadvertent SI at power could cause the pressurizer to become water-solid if the resulting injection of borated water is not terminated. In the event that the pressurizer becomes fully water-solid, timely PORV actuation successfully mitigates the event. However, without automatic operation of the PORVs, the Reactor Coolant System (RCS) pressure may increase to the lift setpoint of the pressurizer safety relief valves before the PORVs are manually opened. The Salem pressurizer safety valves are not designed to relieve water. It is postulated, therefore, that one or more of the valves could fail to completely reseat if relieving a water-solid pressurizer.

A resulting unisolable loss of RCS inventory has been analyzed in Salem's UFSAR as a Condition III event.

A review of the current Salem TS indicates that a TS revision is necessary to preclude the possibility of operating with PORVs that can only be cycled manually. PSE&G's re-analysis of the Inadvertent SI at Power performed to support resolution of NSAL 93-013, credits operator action to unblock the PORVs, if necessary. However, once unblocked it is unlikely that operator actions can be readily accomplished to manually cycle the PORVs such that the pressurizer safety valve pressure is not reached.

Therefore, PSE&G is submitting this proposed TS change to the staff for their review to incorporate the results of PSE&G's analysis, (i.e., to credit automatic operation of PORVs for an Inadvertent SI event), into Salem's license.

JUSTIFICATION OF REQUESTED CHANGES PSE&G has evaluated the ability of the PORVs to mitigate an inadvertent SI event. The discussions below provide detail into the various design aspects that have been reviewed. The Page 2 of 11

Document Control LR-N97055 LCR S97-03 evaluations detailed below were performed for Salem Unit 2, but are considered applicable to both Units due to the similarities in configuration. PSE&G will be performing a specific Unit 1 evaluation on the piping downstream of the PORVs and will also be assessing the impact of the replacement Steam Generators prior to its restart from the current outage. It is anticipated that the evaluation will yield similar results to that presented below. The results of those evaluations, however, do not impact the proposed changes to the TS contained in this submittal, though it may potentially result in the need for plant or procedural enhancements.

Failure Modes The credible failure modes, discussed herein, are derived from the application of the pressurizer PORVs to preempt the opening of the pressurizer safety valves with the RCS in a water-solid condition. An analysis of the Inadvertent SI Actuation event, performed by Westinghouse, using the LOFTRAN code, indicates that the opening of one PORV upon reaching its opening setpoint pressure, is sufficient to limit the pressurizer pressure to levels below the opening setpoint pressure of the pressurizer safety valves; even while the PORV is relieving water.

PSE&G's evaluation concludes that the pressurizer PORVs, block valves, the associated downstream piping, and the Pressurizer Relief Tank can be relied upon to accommodate the water discharged from a water-filled pressurizer.

Salem's PORVs are supplied with air accumulators should the normal supply of air be unavailable. Exhaustion of the accumulator air required to operate the PORVs before the operators can terminate the accident has been reviewed. It has been concluded that sufficient reserve of accumulator air is available to operate the PORVs for about 45 minutes, enough time for the operators to manually terminate the transient. More details on this analysis and the cycling of the PORVs is discussed later.

The PORVs will open when two of two pressurizer pressure channels reach the setpoint pressure. For each PORV, one of the two channels in the circuit is selectable with all channels powered from vital instrument buses. It has been determined that a failure of the "C" vital instrument bus, with the PORVs selected to either selectable channel, or "A" vital instrument bus, with the PORV selected to the alternate channel, would:

(1) prevent automatic operation of both pressurizer PORVs, (2) generate an SI actuation signal in one protection channel, (3) prevent operation of pressurizer spray, and (4) prevent automatic startup of a centrifugal charging pump.

Page 3 of 11

Document Control .De. LR-N97055 LCR S97-03 Failure of a vital instrument bus would not, by itself, fulfill the logic required to actuate the SI System. Failure of a vital instrument bus and an Inadvertent SI Actuation are considered to be independent events.

Following the guidance of ANSI 18.2/51.1, PSE&G has determined that the combination of an Inadvertent SI Actuation with a coincidental failure of a vital instrument bus need not be considered for design. ANSI 18. 2/51.1 states, "If the frequency of occurrence of an initiating occurrence, or initiating occurrence plus single failure or coincident occurrences, is shown to be less than 10~/reactor year on a best-estimate basis per 3.2.3, this event shall not be considered for design." Based upon the data used in Salem's Probabilistic Safety Assessment, the probability of an Inadvertent SI Actuation event occurring in coincidence with failure of "A" or "C" vital instrument bus, is less than 10~/reactor year. Therefore, this scenario need not be considered for design.

Thermal Hydraulic Effects and Piping Loads The thermal hydraulic analysis was performed using the RELAP5/MOD3.2 program and VECTRA Company's PREPREF and REFORC programs. RELAP5 calculates the fluid conditions in the piping system. PREPREF is a pre-processor of the RELAP5 output for use by REFORC. REFORC is used to calculate the forces exerted on the pipe segments of the system .. These forces can be determined as elbow loads or axial pipe loads. The output from REFORC is utilized as input to SUPERPIPE. Plots of the axial forces can also be generated from the REFORC output. Use of these computer codes was previously reviewed and approved by the NRC for use at PSE&G to support submitted responses to NUREG-0737, TMI actions, pertaining to PORV reliability.

The pressurizer pressure relief system was previously modeled in PSE&G calculations that were performed to analyze the PORV and safety relief valve piping for the drained loop seal modification. The stroke time of the PORVs in the analysis was conservatively assumed to be 0.5 seconds, (TS acceptance criteria is 2 seconds) . The piping downstream of the PORVs is initialized at 15 psia and 120°F. The transient problem time was 1.0 seconds. This is consistent with previous analyses.

Certain changes to the RELAP5 model were made to reflect the initial conditions as defined by EPRI Report NP-2296. These changes were:

1. Piping from the pressurizer to the PORV was modeled to be water solid at a pressure of 2353 psia and a temperature of 565°F. Westinghouse LOFTRAN analysis shows the temperature to be higher; however, the lower temperature provides more conservative results for structural analysis.

Page 4 of 11

Document Control LR-N97055 LCR S97-03

2. The valve model was modified to ensure control by the coefficient of valve flow, Cy, and not by the throat area.

The Cy was modified as necessary to ensure the flow through the valve met the EPRI test data and Copes-Vulcan information. Because of the water flashing conditions across the valve, the Cy valve model was deemed more appropriate.

The output of this model was.then used as input into the structural model.

Consistent with the current analysis for the PORV piping, the effect of the Inadvertent SI event was evaluated for both the upstream PORV piping and the PORV downstream or discharge piping.

For the discharge piping, the axial pipe forces due to the Inadvertent SI event developed as part of the thermal hydraulic analysis described above were compared to the axial pipe forces resulting from the steam discharge event considered in the analysis of record for the discharge piping. The comparison shows that the magnitude of the forces resulting from the Inadvertent SI event is less than those evaluated for the steam discharge.

Since the magnitude of the forces resulting from the Inadvertent SI event were equal to or greater than those evaluated for the steam discharge case for the PORV upstream piping, a re-analysis was performed for the PORV upstream piping using the SUPERPIPE computer program. The computer model for the upstream piping was taken from the analysis of record. Consistent with the analysis of record, elbow forces developed as part of the thermal hydraulic analysis described above were used as input for the piping analysis. The results of this analysis show the effect on the upstream piping due to an Inadvertent SI event to be similar to the effect from the PORV steam discharge considered in the analysis of record.

The PORV upstream and downstream piping remains within the Code allowable limits for the Inadvertent SI event.

Operator Action Times The conclusions of the re-analyzed Inadvertent SI at Power event are based upon the assumption that the operators, working according to Emergency Operating Procedures, act within approximately ten minutes, (e.g., 610 seconds), after the event occurs to make at least one pressurizer PORV available by opening its associated block valve. This assumption has been validated by simulator test results which indicate that operators have been successful in accomplishing this procedure within seven to nine minutes. Unblocking an isolated PORV within 610 seconds has been incorporated as a time critical step to which the Salem operating Page 5 of 11

Document Control LR-N97055 LCR S97-03 crews are trained and assessed during simulated emergency exercises.

Operator reaction times were also used to determine the maximum time for mitigating the inadvertent SI event, (i.e., the establishing of normal charging and letdown) . This data was based on simulator exercises using the operating procedures as of May 21, 1996 and was used as input into determining the maximum number of PORV cycles.

The response times for the steps in mitigating the event can be conservatively defined as follows:

Time to reach the point of securing two of the three charging pumps: 23 minutes Time to reach the point of Cold Leg injection isolation: 25 minutes Time to reach the point of event mitigation: 45 minutes The method of analysis used for SI termination accounted for uncertainty in the response times and as such no further conservatism need be applied.

Available PORV Cycles This section discusses the performance of the PORV actuators and the Control Air system with respect to the determination of the number of available open and closed PORV cycles that can be accommodated. A full stroke evaluation will first be discussed in which it is determined that each PORV will have sufficient motive force to fully open and fully close on each cycle.

Second, a partial stroke will be discussed. For the partial stroke it is determined that each PORV will have sufficient motive force for the actuator to be between full and half open.

In the partial stroke case as well as the full stroke case, each PORV will have capacity to fully close (spring return close) .

Third, the in line check valves will be discussed as they result in possible back leakage and loss of air.

This evaluation determines the number of available full strokes of the PORVs while mitigating an Inadvertent SI event. For this evaluation, the Control Air System accumulators are the only sources of air to the PORV actuators. Each PORV has two dedicated air accumulators. Conservatism is added in that the air accumulators pressure starts at 85 psig. This 85 psig is the point at which during normal plant operation if the Control Air System malfunctions and system pressure begins to decrease, the Emergency Control Air Compressors start. Both the Control Air System and the Emergency Control Air System have the capacity to maintain the air system at 110 psig.

For this evaluation, in order to take full credit for all available motive forces for opening and closing the PORVs, the Page 6 of 11

Document Control LR-N97055 LCR S97-03 RCS. system pressure was credited in the assisted opening of each PORV. Using the system pressure at the inlet and outlet of the PORV in the full open condition, the differential pressure across the internal trim and cage of the valve was credited in providing an opening force on the valve disc. This force along with the force of the air actuator, opens the valve. Given the inlet and outlet system pressures at the PORVs, Copes-Vulcan evaluated the full open requirements of the PORVs and determined the minimum air pressure in the actuator. Using this information and maintaining the existing air accumulator configuration, each PORV has the capability to fully stroke open and close 305 times.

Expanding upon the same methodology used in the full stroke evaluation, at the end of 305 full strokes, the PORVs will not stop opening and closing, but will no longer be capable of fully opening on each stroke. As the air pressure in the air accumulator continues to decrease on each stroke of the PORVs, the actuator spring pressure will cause the open stroke of the PORVs to decrease.

PSE&G calculated the available strokes from the end of full stroking of the PORV actuator to fifty percent stroking of the PORV actuators. In this calculation, an additional 486 strokes were available for each PORV.

With the PORVs actuator at fifty percent open, the valve Cy is approximately 15. A full open PORV has a Cy of 50. Westinghouse analyzed the RCS to determine the PORV stroking frequency during the mitigation of an Inadvertent SI event. As part of Westinghouse's analysis, fifty percent stroking (Cy of 15) of a PORV was evaluated (with respect to the later portion of the event) to determine the expected stroking frequency. The stroking frequency remained very similar to that of a full stroking PORV possibly due to the overshooting versus undershooting of the RCS pressure with respect to the PORV setpoint. This Westinghouse analysis documented the adequacy of the PORV partially open with a Cy of 15 in relieving the RCS pressure during the later time frame/portion of the Inadvertent SI event.

As the result of the full stroking and partial stroking of the PORVs, the available stroke of the PORVs to mitigate an Inadvertent SI event is 791 strokes.

Check Valve Leakage For each PORV, the air accumulators have two poppet type check valves. Back leakage of these check valves will reduce the air supply of the air accumulators during containment isolation. It is anticipated that the PORVs will be required to function for no more than 60 minutes following the containment isolation resulting from an Inadvertent SI event. The air mass loss of the accumulators for each PORV due to the check valves' leakage for Page 7 of 11

Document Control LR-N97055 LCR S97-03 60 minutes is calculated to be 0.048 lbm. This potential reduction in air mass within the accumulators is considered to have an insignificant impact on the requisite amount of air. The periodic accumulator check valve testing, performed in accordance with Specification 4.4.5.1, ensures that the actual valve leakage will not impact the analyzed amount of air required.

PORV Controls and Motive Air System PSE&G evaluated the PORVs' automatic, high pressure over pressure protection control grade circuit. The evaluation examined the extent to which these circuits satisfy full safety related design basis criteria and their anticipated reliability in actuating the PORVs to reduce RCS pressure following an inadvertent SI System actuation.

The following approach was used to evaluate the reliability of the PORV automatic, high pressure over pressure protection control grade circuit: (1) determine the functional requirements of the circuit to mitigate the consequences of the Inadvertent SI Actuation event, (2) identify the circuit components and define their qualification status, (3) review components performance history, (4) evaluate the control scheme for redundancy, (5) based on the acquired information, make an overall qualitative assessment of reliability.

The PORV automatic, high pressure over pressure circuit design is highly reliable, with all required electronic components being safety related, Class lE, Seismic Category I, and evaluated for the environmental conditions at their installed location. The control circuits themselves cannot be considered Class lE because they are not redundant. The power supplies are derived from Class lE, safety related battery-backed sources. The PORV air accumulators, associated tubing and solenoid valves are safety related, Seismic Category I. The review of the maintenance and surveillance work orders has shown these components to be relatively trouble free.

The controller, PC455K, for valve PR2 is non-safety related, non-lE, and Seismic Category II and was used to provide anticipatory opening of the PORV. Although this feature is no longer functional with the present logic, the controller remains in the circuitry. Failure of the controller affects PR2 only and would be readily detectable at power since its failure also impacts the pressure controls of the pressurizer heaters and sprays. Based upon the performance history of the controller itself, (only one corrective maintenance work order identified for 2PC455K and no unacceptable as-found readings noted for the channel calibration testing results reviewed) and of the control circuits as a whole, the controller is considered reliable under the expected normal conditions. Failure of this controller would result in an Page 8 of 11

Document Control LR-N97055 LCR S97-03 inoperable valve which would result in complying with the actions of TS 3.4.5b, return the PORV to an operable condition within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> or initiate a plant shutdown.

This evaluation has determined that the PORV associated control circuits and motive air systems are sufficiently reliable for the application and can support the associated operational and design functions.

The Endurance of the PORVs and Actuators The purpose of this section is to address concerns with regards to the functioning of the PORVs and associated air actuators following numerous strokes (approximately 220 strokes) .

Endurance testing of different trim material combinations for the PORV was performed at Wyle Laboratories. A spare 2" Copes-Vulcan Model D-100-160, carbon steel valve and actuator from PSE&G was used to host five different trim combinations. The test PORV was installed on a steam header test stand. A tent-like environmental chamber was installed to duplicate the typical environmental conditions for the PORVs in the pressurizer enclosure. The test PORV was opened at a steam pressure of 2335 psig with 90 psig of air on the actuator and was immediately closed following full open. A two second maximum time from closed-to-open-to-close was targeted. This test was performed 2,000 times over a six hour period for each test (one cycle approximately every ten seconds) .

The following summarizes the results of the testing performed at Wyle Laboratories:

All five trim combinations were full stroke tested 2,000 times without loss of function or stroke time. Each test cycle was completed within a two second time frame.

There were no packing leaks or packing gland adjustments required during any of the five tests.

There were no diaphragm failures during any of the five tests.

The same solenoid valve was utilized for the entire 10,000 cycles without loss of function.

In summary, all five trim material combinations performed within the specified operational requirements. One trim material combination was selected over the others based upon optimum performance. Based upon the above test, the PORVs will function adequately in the mitigation of an Inadvertent SI Event.

Page 9 of 11

Document Control LR-N97055 LCR S97-03 Overall Performance of PORVs and Associated Systems For this section, the operator action times and the available PORV cycles along with a pressurizer PORV cycling analysis is used as input to determine the performance of the PORVs in mitigating an Inadvertent SI event.

The pressurizer PORV opening/closing characteristics with respect to the RCS system, thus predicting the PORV cycling during the Inadvertent SI event, was modeled.

The base model assumed SI flow to be from one Positive Displacement pump and two Charging/SI pumps and discharge flow through one PORV. The analysis was performed in two steps; the first step being with the plant initially at full power and with the pressurizer water level at normal level (steam discharge)

The second step is with the plant at zero power and with the pressurizer filled with water (water discharge) . In both of these steps, several sensitivity cases were evaluated in predicting the worst case scenario.

For the steam discharge, the worst case cycling frequency occurred with a delayed reactor trip and beginning of life core kinetics parameters resulting in an average cycle time of 7.3 seconds. For the water discharge, the worst case cycling frequency occurred with 50% of full flow through the PORV (potential reduction in flow due to flashing) resulting in an average cycle time of 7.3 seconds as well.

As discussed previously, the Positive Displacement pump and one of the Charging pumps will be secured approximately twenty-three minutes into the event. For liquid discharge, the average cycle time is predicted to be 11.6 seconds.

Once the high head injection flow path is isolated at approximately 25 minutes into the accident, the SI flow is reduced to only the 40 gpm of seal injection flow resulting in an average cycle time of 62.0 seconds. At this time, pressurizer fill is minimal and the event is essentially over.

With all three charging pumps in operation, the PORVs cycle at a rate of 8.2 strokes/minute. Statistically this condition could exist for twenty-three minutes. Therefore, 189 strokes of the PORVs would occur. At this point, two charging pumps are secured and the PORVs cycle at a rate of 5.2 strokes/minute. This condition could occur for an additional two minutes and would cause an additional eleven strokes. Once the high head injection flow path is isolated, the cycling of the PORVs is further reduced to 0.97 strokes/minute. This condition could occur for twenty minutes more, at which point the event is terminated, causing the PORVs to stroke twenty more times. This is a total of 220 full strokes of the PORVs. The available air can provide a total of 305 full strokes with an additional 486 partial Page 10 of 11

Document Control LR-N97055 LCR 897-03 strokes until the valves will not open more than 50% of full stroke.

This submittal does not impact the ability for the PORVs to operate for low-temperature overpressurization protection.

CONCLUSIONS The changes submitted by this request ensure that the PORVs will perform their design function for an inadvertent safety injection at power. PSE&G has performed analyses to demonstrate the ability of the PORVs to mitigate the consequences of this accident.

Page 11 of 11

.. Document Control Attachment 2

• LR-N97055 LCR S97-03 SALEM GENERATING STATION UNIT NOS. 1 AND 2 FACILITY OPERATING LICENSES DPR-70 AND DPR-75 DOCKET NOS. 50-272 AND 50-311 CHANGE TO TECHNICAL SPECIFICATIONS PRESSURIZER POWER OPERATED RELIEF VALVES 10CFR50.92 EVALUATION Public Service Electric & Gas (PSE&G) has concluded that the proposed changes to the Salem Generating Station Unit Nos. 1 and 2 Technical Specifications (TS) do not involve a significant hazards consideration. In support of this determination, an evaluation of each of the three standards set forth in 10CFR50.92 is provided below.

REQUESTED CHANGE The proposed TS changes contained herein represent changes to Specification 3.4.5 pertaining to the actions associated with an inoperable Power Operated Relief Valve(s) (PORV). The current TS actions for continued plant operation with an inoperable PORV state that the valve must be isolated by its associated block valve and capable of being manually cycled (e.g., remotely from the control room) . The proposed change will revise the action to limit continued plant operation with an inoperable PORV due to seat leakage.

In addition, the applicable Bases section is being revised to include clarification of the design functions of the PORVs and block valves.

BASIS

1. The proposed change does not involve a significant increase in the probability or consequences of an accident previously evaluated.

The proposal does not involve any physical changes to plant systems or components. No new protection system logic is proposed, and therefore, there is no additional signal that can spuriously actuate the Safety Injection (SI) system.

Consequently, there would be no change in the probability of occurrence of the accident, as previously evaluated in the UFSAR. The proposal is based upon a reanalysis of the Inadvertent SI event to include a case that demonstrates that the postulated event would not be likely to lead to a more serious event.

Page 1 of 3

Document Control LR-N97055 LCR S97-03 Sustained water relief through a PORV can result in a release of reactor coolant into containment from the Pressurizer Relief Tank. The release is limited, however, since (1) it is the result of the SI System addition and consequently cannot exceed the SI flow rate at the PORV setpoint pressure, and (2) the SI flow will eventually be terminated by the operators. The dose consequences for an Inadvertent SI is bounded by that which is calculated for the spurious opening of a pressurizer safety valve, Accidental RCS Depressurization event, which is also a Condition II event.

2. The proposed change does not create the possibility of a new or different kind of accident from any accident previously evaluated.

The Inadvertent Operation of the SI System at Power analysis cases, reported in the UFSAR, are analyzed to challenge fuel integrity. Accordingly, the UFSAR analysis cases produce transients that lead to a reduction in pressurizer pressure, in order to reduce the thermal margin. The results indicate that no fuel damage is predicted. The UFSAR analysis is revised in order to evaluate the effects of an increase in pressurizer pressure and other conditions that could lead to water relief through the pressurizer safety valves.

Allowing water relief from the pressurizer would not affect the likelihood of fuel damage occurring during this event.

The results of the accident reanalysis indicate that the pressurizer safety valves would not discharge water, (a condition for which they are not designed), and consequently this event will not result in the failure of a pressurizer safety valve due to the discharge of water through the pressurizer safety valves.

An evaluation of the effects of water relief through the PORVs and downstream piping have also been conducted. The results of the accident reanalysis and the associated evaluation indicate that a different type of malfunction (e.g., a stuck open pressurizer safety valve or failure of downstream piping or components) would not be expected to result from the analyzed event. Therefore, a different type of accident would not be expected to occur as a result of implementation of this proposal.

3. The proposed change does not involve a significant reduction in a margin of safety.

For this proposed change, the safety analysis criterion, which the analysis of Inadvertent SI Actuation at Power event is required to satisfy, is to show that the pressurizer safety valves would not open and discharge water Page 2 of 3

Document Control LR-N97055 LCR S97-03 at any time during the event. Satisfaction of this criterion indicates that the safety margin is preserved by preventing the Inadvertent Operation of the SI System at Power event (a Condition II event) from escalating into a more serious event, (a Condition III event).

The proposal does not reduce the margin of safety, since the results of the reanalysis indicate that the applicable safety analysis acceptance criterion, which is established to protect the margin of safety, is satisfied.

The conclusions of the reanalyzed Inadvertent Operation of the SI System at Power event are based upon the assumption that the operators, working according to Emergency Operating Procedures, act within ten minutes after the event occurs to make at least one pressurizer PORV available by opening its associated block valve. This is a justifiable assumption, since simulator test results indicate that operators have been successful in accomplishing this procedure within seven to nine minutes and this requirement has been incorporated into the procedures as a time critical step.

Therefore, relief capability is assured prior to the pressurizer achieving a solid water condition.

The PORV surveillance requirements that are currently contained in the Salem TSs ensure that the automatic operation of the PORVs is periodically tested.

CONCLUSION Based on the above, PSE&G has determined that the proposed changes do not involve a significant hazards consideration.

Page 3 of 3

Document Control LR-N97055 LCR S97-03 SALEM GENERATING STATION UNIT NOS. l AND 2 FACILITY OPERATING LICENSES DPR-70 AND DPR-75 DOCKET NOS. 50-272 AND 50-311 CHANGE TO TECHNICAL SPECIFICATIONS PRESSURIZER POWER OPERATED RELIEF VALVES TECHNICAL SPECIFICATION PAGES WITH PROPOSED CHANGES The following Technical Specifications for Facility Operating License No. DPR-70 are affected by this supplement:

Technical Specification Page 3.4.5 3/4 4-5 Bases for 3/4.4.5 B 3/4 4-la The following Technical Specifications for Facility Operating License No. DPR-75 are affected by this supplement:

Technical Specification Page 3.4.5 3/4 4-8 Bases for 3/4.4.5 B 3/4 4-2 B 3/4 4-3 Page 1 of 7

Document Control Del LR-N97055 Attachment 3 LCR S97-03 Insert "A" The OPERABILITY of the PORVs and block valves is determined on the basis of their being capable of performing the following functions:

A. Manual control of PORVs to control reactor coolant system pressure. This is a function that is used for the steam generator tube rupture accident and for plant shutdown.

B. Automatic control of PORVs to control reactor coolant system pressure. This is a function that reduces challenges to the code safety valves for overpressurization events, including an inadvertent actuation of the Safety Injection System.

c. Maintaining the integrity of the reactor coolant pressure boundary. This is a function that is related to controlling identified leakage and ensuring the ability to detect unidentified reactor coolant pressure boundary leakage.

D. Manual control of the block valve to: (1) unblock an isolated PORV to allow it to be used for manual and automatic control of Reactor Coolant System pressure (Items A & B), and (2) isolate a PORV with excessive seat leakage (Item C) .

E. Manual control of a block valve to isolate a stuck-open PORV.

Page 2 of 7