ML17158C229

Safety Evaluation Accepting Licensee 960506 & 0423 Ltrs Re Revs to IST Programs for Plant,Units 1 & 2,respectively

ML17158C229
Person / Time
Site:	Susquehanna
Issue date:	07/17/1997
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML17158C228	List: ML17158C227 ML17158C229
References
NUDOCS 9707230111
Download: ML17158C229 (19)

Text

gP,R RK00 ti UNITED STATES NUCLEAR REGULATORY COMMISSION g

WASHINGTON, D.C. 20555-000I

>>~~++

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO THE INSERVICE TESTING PROGRAM RELIEF RE UESTS PENNSYLVANIA POWER AND LIGHT COMPANY SUS UEHANNA STEAM ELECTRIC STATION UNITS 1

AND 2 DOCKET NOS.

50-387 AND 50-388

1. 0 INTRODUCTION The Code of Federal Regulations, 10 CFR 50.55a, requires that inservice testing (IST) of certain American Society of Mechanical Engineers (ASME) Code Class 1, 2, and 3 pumps and valves be performed in accordance with Section XI of the ASME Boiler and Pressure Vessel Code (the Code) and applicable

addenda, except where alternatives have been authorized or relief has been requested by the licensee and granted by the Commission pursuant to Sections (a)(3)(i),

(a)(3)(ii), or (f)(6)(i) of 10 CFR 50.55a.

In proposing alternatives or requesting relief, the licensee must demonstrate that:

(1) the proposed alternatives provide an acceptable level of quality and safety; (2) compliance would result i'n hardship or unusual difficulty without a compensating increase in the level of quality and safety; or (3) conformance is impractical for its facility.

Section 50.55a authorizes the Commission to approve alternatives and to grant relief from ASME Code requirements upon making the necessary findings.

Additionally, 50.55a(f)(4)(iv) of Section 50.55a provides that IST of pumps and valves may meet the requirements set forth in subsequent editions and addenda of the Code that are incorporated by reference in paragraph (b) of Section 50.55a, subject to the limitations and modifications listed therein, and subject to Commission approval.

Portions of editions or addenda may be used provided that all related requirements of the respective editions or addenda are met.

Guidance related to the development and implementation of IST programs is given in Generic Letter (GL) 89-04, "Guidance on Developing Acceptable Inservice Testing Programs,"

issued April 3,

1989, and its Supplement 1 issued April 4, 1995.

Also see NUREG-1482, "Guidelines for Inservice Testing at Nuclear Power Plants."

The 1989 Edition of the ASME Code is the latest edition incorporated by reference in paragraph (b) of Section 50.55a.

Subsection IWV of the 1989 Edition, which gives the requirements for IST of valves, references Part 10 of the American National Standards Institute/ASME Operations and Haintenance Standards (OM-10) as the rules for IST of valves.

OM-10 replaces specific requirements in previous editions of Section XI, Subsection IWV, of the ASME Code.

Subsection IWP of the 1989 Edition, which gives the requirements for IST of pumps, references Part 6 of the American National Standards Institute/ASME Operations and Haintenance Standards (OM-6) as the rules for IST of pumps.

OM-6 replaces specific requirements in previous editions of Section XI, Subsection IWP, of the ASME Code.

ENCLOSURE 9707230l'i i 970717 PDR ADQCK 05000887 P

PDR

2.0 BACKGROUND

By letters dated Hay 6,

1996, and April 23,

1996, Pennsylvania Power 8 Light Company submitted revisions to the IST programs for Susquehanna Steam Electric

Station, Units 1 and 2 respectively.

The revised programs include changes made in response to the Safety Evaluation (SE), for the second 120-month interval for both units, dated April 26, 1995.

In a letter dated August 4,

.1995, the licensee responded to certain items related to relief requests that were denied in the SE.

The revised programs include the changes made related to the August 1995 response.

Several relief requests have been revised and are evaluated below.

The licensee also submitted a letter dated Hay 20,

1996, that included revised Relief Request RR-23 for both units, which is also evaluated below.

3.0 EVALUATION OF REVISED RELIEF RE UESTS The IST programs for Susquehanna Steam Electric Station (SSES),

Units 1 and 2, were developed to the requirements of the 1989 Edition of the ASHE Code.

The SSES Units 1 and 2 are boiling-water reactors which began commercial operation on June 8,

1983, and February 12, 1985, respectively.

The second 120-month interval for IST commenced on June 1,

1994, placing both units on concurrent intervals to the same edition of the Code.

In response to action items identified in the referenced SE, the licensee revised a number of relief requests which are evaluated below in conjunction with the discussion of the licensee's actions.

Additionally, several cold shutdown justifications and refueling outage justifications were revised as a result of action items in the SE;

however, these revisions are not evaluated

herein, but are subject to further review during IST inspections.

3.1 Action Item 5.1 This action item provided the results of a review of the IST scope for the emergency service water, nuclear boiler, and residual heat removal systems.

The licensee's response indicates the actions taken for each of the recommendations in Action Item 5. 1.

No new or revised relief requests are related to this action item; therefore, the revisions to the IST program are not evaluated herein but are subject to NRC inspections.

3.2 Action Item 5.2 Relief Re uest RR-03 In the April 1995 SE, for the ASHE Code Class 3 submerged diesel fuel oil transfer pumps OP514 A through D, interim relief was granted for RR-03 for a period of 1 year or until the next refueling outage, whichever was later.

Action Item 5.2 recommended that the licensee revise and resubmit the relief request to indicate an alternative course of action such as the institution of a regular maintenance and spare parts program and inspection of the pump bearings whenever the diesel fuel oil storage tanks are drained.

Item 5.2 also recommended that the licensee evaluate the feasibility of determining pump flowrates and differential pressure in accordance with the Code by observing level changes of the day tanks or by installing flow instrumentation and calculating inlet pressure.

For pump OP514E, the one diesel fuel oil transfer pump external to the tanks interim relief was granted for a period of 1 year or until the next refueling

outage, whichever was later.

Relief Request RR-03 contained insufficient basis for not using vibration instrumentation, calculating differential

pressure, and determining flow rate during quarterly testing.

Item 5.2 recommended that the licensee further evaluate the feasibility of complying with the pump testing requirements of Part 6

(OM-6) for pump OP514E; The licensee revised Relief Request RR-03 to address'he recommendations in Action Item 5.2.

3.2. 1 Licensee's Basis for Relief The licensee states in revised Relief Request RR-03:

Four of these pumps (OP514A thru D) are a sealed-unit submersible type

[pump] with the entire unit submerged in the Diesel Oil Storage Tanks.

None of these pumps include any provisions for flowrate, inlet pressure, bearing temperature, or vibration amplitude indication or measurement.

The Susquehanna Steam Electric Station

[SSES] Technical Specifications presently require at least a monthly functional test of these pumps.

This test verifies fuel oil flow from the storage tank to each diesel's skid-mounted day tank.

Similarly, the diesel oil firing pumps are tested during the diesel functional tests.

These pumps take suction from the day tanks and supply the diesel cylinders.

SSES considers all these pumps an extension of the diesel engine equipment skid, and therefore all are adequately tested per Technical Specifications along with the diesel itself.

In addition, the actual flowrate for the diesel fuel oil transfer pumps is over five times the required diesel engine fuel usage at rated conditions.

Thus, even a large reduction in pump flow will not affect system operability.

Submersion of each pump in the respective Diesel Oil Storage Tank renders.vibration measurement impractical.

Disassembly and inspection of the bearings of each pump whenever the respective Diesel Oil Storage Tank is drained for its 10-year inspection will be considered, based on pump performance trends and total running time since the previous bearing inspection for that pump.

Control of the flows and pressures of these pumps is automatic; no means for manual control has been provided.

This fact precludes compliance with OH Standard

[Part 6] paragraphs 5.2(a) and 5.2(b).

The pump flow in these lines [can] vary widely.

The large variations of flow rate produce large inverse variations of pump differential pressure.

U

The premise of OM Standard

[Part 6] paragraph 5.2(c) is that uncontrollable variation of pump flow'ate and differential pressure will be <<+ lOX, because it required comparison of differential pressure and flow rate measurements with their

~

corresponding reference values (and limits given in Table 3, per paragraph 5.2(d)).

This premise is not applicable to these

pumps, as their uncontrollable variation in flow rate (and accompanying inverse variation in differential pressure) is >+ 10X; thus precluding any meaningful comparison with the limits of Table 3

and precluding any compliance with paragraphs 5.2(c) and 5.2(d).

The range of acceptable flow rate values theoretically possible

[for] these pumps is between 10 gpm and 40 gpm.

3.2.2 Alternative Testin The licensee proposes:

Each pump will continue to be functionally tested at least monthly per Technical Specification 4.8. 1. 1.2.

Additionally, at each 18-month diesel maintenance work period, a special test of the corresponding Diesel Fuel Oil Transfer Pump will be run, from which pump inlet pressure, pump differential pressure, and flow rate will be calculated.

This inservice testing will be conducted at the hydraulic conditions established automatically by the Diesel Fuel Oil Transfer System.

The test parameters shown in OH Standard

[Part 6] Table 2 will be determined and recorded, except for vibration of the four submerged pumps.

Vibration of the only accessible

pump, OP514E, will be determined and compared with reference values by the method of paragraph 5.2(d).

Pump differential pressure and flow rate will be determined and compared with limits generated from the system operating requirements of flow rates from 10 gpm to 50 gpm and appropriate corresponding differential pressures, specific to each pump.

3.2.3 Evaluation Though the fuel oil transfer system is in direct support of the operation of the diesel generators, it is not part of the skid-mounted equipment supplied with the engine skid (i.e., integral unit of equipment on a moveable support structure).

Nevertheless, the monthly testing of the diesel required by plant technical specifications challenges the operation of the diesel fuel oil transfer

system, giving assurance that the system is in working condition able to supply makeup fuel oil to the day tank.

The maintenance and testing performed every 18 months gives an added level of assurance that the diesel fuel oil transfer system will function when conditions require the diesel generator to operate.

The changes in the relief request address the recommendations in the April 1995 NRC SE.

The periodic testing is structured to be as close in compliance with the Code as is practical considering the design of the system and components.

Imposition of full compliance with the Code requirements would be a burden on the licensee in that the system.would

have to be modified extensively, either by relocating the four submerged pumps or installing permanent, submerged vibration monitors, and installing pressure and flow instruments, potentially requiring major piping modifications and introducing a potential unreliability factor if instrumentation is submerged.

The external pump and its associated piping would also require extensive modification to meet the Code testing requirements.

The imposition is unwarranted in that the alternative testing will give an adequate level of assurance of the operational readiness of these pumps.

3.2.4 Conclusion In consideration of the impracticality of complying with the Code requirements and of the burden that would result if the requirements were imposed, relief is granted for IST of the diesel fuel oil transfer pumps pursuant to 10 CFR 50.55a(f)(6)(i).

As a condition of granting this relief, the licensee must perform alternative testing as proposed in the relief request to provide an adequate level of assurance of the operational readiness of these pumps.

3.3 Action Item 5.3 A request to run the high pressure cooling injection pumps for less than the 2

minutes prescribed by the Code was denied.

The original request was based upon the need to limit suppression pool temperature increases;

however, no specific information on the expected temperature rise during IST versus time or on the administrative limits for temperature was included in the request.

In the licensee's letter of August 4, 1995, the relief request was withdrawn.

The run time for these pumps will be in accordance with the Code.

3.4 Action Item 5.4 Relief Re uest RR-15 This relief request is applicable to the ASHE Code Class 3 emergency condenser water circulating pumps, OP-171-A/B.

It also includes the non-Code chilled water loop circulating pumps, OP-162-A/B; however, no NRC approval is required for deviations from the Code for non-Code components not subject to the requirements of 10 CFR 50.55a.

The original request for relief from the Code requirements to set either flow or pressure to a reference value and measure the other parameter was granted for an interim period of one year or to the next refueling outage, whichever was later.

Action Item 5.4 recommended that the licensee evaluate compliance with Part 6, paragraph 5.2(c),

which addresses testing where system resistance cannot be varied, and also investigate the use of pump reference curves for testing.

The revised relief request addresses these recommendations.

3.4. I Licensee's Basis for Relief The licensee states in revised Relief Request RR-15:

~ Control of the flows and pressures of these pumps is automatic;

[that is,] the system has no means for manual control.

This fact precludes compliance with OH Standard

[Part 6] paragraphs 5.2(a)

and 5.2(b).

The temperature of the fluid in each hydraulic loop is maintained by one or more temperature control valves automatically.

These temperature control valves have the capability to vary the flow in each hydraulic loop over wide ranges as they automatically regulate the temperature of the water in the loop.

The large variations of flow rate produce large inverse variations of pump differential pressure.

The premise of OH Standard

[Part 6] paragraph 5.2(c) is that uncontrollable variation of pump flow rate and differential pressure will be

<<+ 10X, because it requires comparison of differential pressure and flow rate measurements with their corresponding reference values (and limits given in Table 3, per paragraph 5.2(d)).

This premise is not applicable to these

pumps, as their uncontrollable variation in flow rate (and accompanying inverse variation in differential pressure) is >+ 10X, thus, precluding any meaningful comparison with the limits of Table 3 and precluding compliance with paragraphs 5.2(c) and 5.2(d).

Although the range of flow rate test values already measured has been between 561 gpm and 715 gpm, the range of acceptable flow rate values theoretically possible is between 552 gpm and 740 gpm.

Additional factors affecting flow rate are ambient temperature and heat loads being

serviced, surface condition of each pipe line in each ESW

[emergency service water]

System supply line and in each recirculation line, and the current balanced flow in each ESW System supply line.

3.4.2 lternative Testin The licensee proposes:

Inservice testing will be conducted at the hydraulic conditions established automatically by the Control Structure Chilled Water System.

The test parameters shown in OH Standard

[Part 6] Table 2

will be determined and recorde'd.

Vibration will be determined and compared with reference values by the method of paragraph 5.2(d).

Pump differential pressure and flow rate will be determined and compared with limits of 552 gpm to 630 gpm, along with 26 to 38 psid [pounds per square inch differential] for both pumps OP162A and OP162B; and with limits of 580 gpm to 760 gpm, along with 17 to 28 psid for both pumps OP171A and OP171B.

These limifs have been generated from the system operating requirements of flowrates

> 552 gpm and

> 740 gpm and appropriate corresponding differential pressures, specific to each type of pump.

3.4.3 Evaluation The licensee has demonstrated the need for relief from the Code re'quirements for differential pressure and flow rate reference values due to the design limitations associated with the emergency condenser water circulating pumps.

Though it is impractical to meet the Code requirements, the proposed

alternative testing could be based on pump curves, which would be consistent with guidance given in NUREG-1482, rather than two distinct expanded ranges for differential pressure and flow rate.

Unless changes in the two parameters are monitored using a curve, degradation could be masked with the large proposed range of acceptable values.

A change in one value proportional to the other (i.e.,

pump curves) will provide a means to monitor for degradation as an alternative to the Code requirements for specific reference values.

Therefore, if the licensee's intent in the proposed alternative is to use pump reference

curves, the alternative testing will provide an adequate level of assurance of the operational readiness of the pumps.

However, the relief request does not clarify that pump curves will be used, or that, as a minimum, the data points will be otherwise integrated.

3.4.4 Conclusion Monitoring these pumps using two ranges would be an acceptable alternative to using single reference values for the two parameters if the licensee were plotting the points on established pump curves that can be used as "reference value" curves.

It is not clear from the description in the request whether the use of pump curves is intended.

If the intent is to use pump curves, and the guidance in Section 5.2, "Use of Variable Reference Values for Flow Rate and Differential Pressure During Pump Testing," of NUREG-1482 is followed, then the licensee should revise and resubmit the relief request to indicate compliance with the guidance.

If the intent is not to use pump curves, then the licensee must revise and resubmit the request to indicate how the use of the two variables will provide an acceptable alternative for monitoring degrading conditions.

If the licensee has no additional information to support the request, consideration should be given to a periodic maintenance program of disassembly and inspection to supplement the proposed testing.

Otherwise the testing should conform to the requirements of paragraph 5.2(c) of ON-6.

The licensee must determine the appropriate actions and notify the NRC as necessary within 120 days from the date of thi's SE.

The request is

denied, but a period of 120 days with IST conforming to the proposed alternative, and paragraph 5.2(c) of ON-6 to the extent practical, should be adequate time for the licensee to further assess the testing and not so long that the pump would experience further degradation prior to the performance of testing in a manner consistent with the discussion herein (i.e.,

pump curves, the proposed alternative supplemented by periodic maintenance, or in accord with paragraph 5.2 of OM-6).

3.5 ction Item 5.5 Various Relief Re uests Item 5.5 recommended that the licensee revise and resubmit Relief Requests RR-Ol, RR-02, RR-08, RR-09, RR-16, RR-17, RR-18, RR-19, and RR-20 to clarify that the proposed alternative of disassembly and inspection of the applicable check valves is conducted consistent with the guidance in Position 2 of GL 89-04.

The listed relief requests and new Relief Requests RR-25, RR-26, RR-27, RR-28, RR-29, RR-30, and RR-31, dealing with disassembly and inspection, clarify that the process is consistent with guidance in Position 2 of GL 89-04.

The NRC stated in GL 89-04 that the alternative given in Position 2 is an

0

acceptable alternative when exercising the check valves is impractical.

Therefore, the revised relief requests continue to be approved as discussed in the April 1995 SE, and the new relief requests are approved as indicated in GL 89-04.

No further NRC review is included herein.

The implementation of these relief requests and compliance with the guidance in Position 2 of GL 89-04 are subject to further review during NRC inspections.

3.6 ction Item 5.6 Relief Re uests. RR-04 and RR-08 Similar to Action Item 5.5, this item recommended that the licensee review and revise RR-04 and RR-08 (also covered by Item 5.5) to show conformance to guidance in Position 2 of GL 89-04 for disassembly and inspection of the main steam isolation valves leakage control system check valves (RR-04) and the core spray check valves (RR-08).

RR-08, as noted in Section 3.5 above, has been revised to clarify conformance with Position 2 guidance.

For RR-04, the leakage control system has been removed from Unit 2 and will be removed from Unit 1 in the 1996 fall outage.

Therefore, RR-04 is no longer needed and is withdrawn, effective upon removal of the Unit 1 leakage control system.

3.7 Aetio Item 5.7 Relief Re uest RR-05 RR-05 was granted in the April 1995 SE with the provision that the licensee establish stroke time acceptance criteria and appropriate corrective action upon exceeding the acceptance criteria.

The revised relief request clarifies

'hat opening stroke time acceptance criteria will be established in conjunction with safety relief valve testing required by OM-10 and OM-I and that corrective action will be established and documented.

The implementation of this provision is subject to NRC inspection.

3.8 Action Item 5.8 Relief Re uest RR-07 For control rod drive hydraulic valves, relief from the Code requirements was denied in the April 1995 SE.

The request has been withdrawn and the licensee indicates that testing is in accordance with the Code.

3.9 ction Item 5.9 Relief Re uest RR-13 Relief for the control structure chilled water temperature control valves (Unit 1) TV-08612A/8 was denied in the April 1995 SE.

Testing for these valves is in compliance with the Code;

however, the non-Code valves remain in Relief Request RR-13 to indicate that testing does not conform with the Code.

Relief for the non-Code valves does not require NRC approval pursuant to Section 50.55a.

3. 10 Action Item 5. 10 Re ief Re uest RR-21 Relief from requirements for establishing reference stroke times for several air-operated emergency service water system isolation valves and containment isolation valves was denied in the April 1995 SE.

The relief request has been deleted and the IST" program documents reflect compliance with Code requirements.

Implementation of the testing is subject to NRC inspection.

3. 11 Action Item 5. 11 Relief Re uest RR-22 I

. Relief from individual valve monitoring requirements for the emergency switchgear room cooling pressure control valves was denied in the April 1995 SE.

The relief request has been deleted and testing is performed in accordance with the Code requirements.

3. 12 Action Items 5. 12 and 5. 13 These action items related to cold shutdown justifications and refueling outage justifications for deferring testing from quarterly during power operations to cold shutdowns or refueling outages.

The ASHE Code (OH-10 by reference) allows such deferrals when the testing is otherwise impractical; therefore, the licensee may make the determination -without NRC prior approval and remain in compliance with the Code.

Thus, the specifics are subject to NRC inspection, but are not further evaluated herein.

3. 13 Relief Re uest RR-23 Relief Request RR-23 applies to the ASHE Code Class 1 excess flow check valves which function as containment isolation valves and, therefore, have a function to close (NOTE:

Seat leakage testing is not the subject of this relief request).

During normal power operations, the valves are open, allowing process fluid to the associated instruments.

The excess flow check valves close upon any pressure decrease on the instrument side of the valves.

The specific valves are listed in a table in the relief request.

In its letter of Hay 20, 1996, the licensee submitted a revision to Relief Request RR-23.

The revision was in response to 'NRC comments in. an SE issued February 23,

1996, related to RR-23.

RR-23 was denied in part and approved in part; The licensee maintained that testing quarterly during power operations was impractical, but proposed to test the valves just prior to shutting down for a refueling outage and place the testing on a "refueling outage" schedule.

RR-23 was approved with the provision that the valves be tested during the refueling outage, not prior to the outage while the plant is still at power operating conditions.

The revised relief request contains additional information to justify testing these valves during power operations but at a

frequency based on refueling outage schedules.

3. 13. I icensee's Basis for elief The licensee states:

Excess flow check valves are installed on instrument lines penetrating containment in accordance with Regulatory Guide l. 11

["Instrument Lines Penetrating Primary Reactor Containment"].

The lines are sized and/or orificed such that off-site doses will be substantially below 10 CFR 100 limits in the event of a rupture.

Therefore, individual leak rate testing of these valves is not required for conformance with 10 CFR 50, Appendix J, requirements.

Functional testing of valves to verify closure can be accomplished by the process of venting the instrument side of the valve while the process side is under pressure.

Such testing is required by Technical Specification 4:6.3.4 at least once per 18 months.

Systems design does not include test taps upstream of the Excess Flow Check Valves (EFCVs).

For this reason, the EFCVs cannot be isolated and tested using a pressure source other than reactor pressure.

Testing on a frequency greater than once per 18 months is not prudent for several reasons.

The testing described above requires the removal of the associated instrument or instruments from service.

Since these instruments are in use during plant operation, removal of any of these instruments from service may cause a spurious signal which could result in a plant trip or an unnecessary challenge to safety systems.

Additionally, process liquid will be contaminated to some degree, requiring special measures to collect flow from the vented instrument side and also will contribute to an increase in personnel radiation exposure.

Testing on a quarterly basis is deemed impractical since the risk of performing the test quarterly outweighs the benefit achieved with a quarterly test and will also increase personnel exposure.

Testing on a cold shutdown frequency is also impractical considering the large number of valves to be tested and the condition that reactor pressure

> 500 psig

[pounds per square inch gauge] is needed for testing.

[OM Part 10] allows test deferrals to refueling outages if it is impractical to test quarterly or during cold shutdowns.

In this instance, considering the large number of valves to be tested and the conditions required for testing (reactor pressure), it is also a hardship to test all these valves during refueling outages.

Recent improvements in refueling outage schedules (i.e., shorter outages) minimized the time that is planned for refueling and testing activities during the outages.

The appropriate time for performing these excess flow check valves tests during refueling outages is in conjunction with vessel hydrostatic testing.

As a result of shorter outages, decay heat levels during hydrostatic tests are higher than in the past.

If the hydrostatic test was extended to test all EFCVs, the vessel could require depressurization several times to avoid exceeding the maximum bulk coolant temperature limit.

This is an evolution which challenges the reactor operators and thermally cycles the reactor vessel and should be avoided if possible.

Also, based on past experience, excess flow check valve testing during hydrostatic testing becomes the outage critical path and could possibly extend the outage by 2 days if all EFCVs were to be tested during this time frame.

A review of the maintenance history for EFCVs has shown that they have been extremely reliable over the life of the plant, showing

< 1X failure rates associated with testing of these valves.

Examples of causes for the failures included alarm problems, indication (limit switch adjustments),

blown fuses, and dirt in the instrument lines.

Only half of the failures required replacements of the valves.

This review of the surveillance test history shows no evidence of time-based failure meehan'isms or common mode failures associated with the excess flow check valves.

A proposed alternative to testing during the refueling outage would be to test certain excess flow check valves immediately preceding the refueling outage while the reactor is at power, while also instituting the appropriate administrative and scheduling controls.

This provides the appropriate conditions for testing (reactor pressure

> 500 psig), while also providing an acceptable level of quality and safety.

Performance of the excess flow check valve testing prior to the outage will be scheduled such that, in the event of a failure, the resulting action statement and limiting condition of operation will encompass the planned shutdown for the refueling outage.

Using this strategy, unplanned, unnecessary plant shutdowns as a result of excess flow check valve testing will be avoided.

In summary, considering the extremely low failure rate, personnel and plant safety concerns, and the hardship of testing during refueling outages, EFCV testing at a frequency greater than once per operating cycle and exclusively during refueling outages is impractical and results in a hardship without a compensating increase in the level of safety.

3. 13.2 lternative Testin The licensee proposes:

Functional testing with verification that flow is checked will be performed at least once per 18 months per Technical Specification 4.6.3.4, immediately preceding a planned refueling outage and with the appropriate administrative and scheduling controls established.

3.13.3 Evaluation The licensee maintains that testing of the EFCVs quarterly at power operating conditions is impractical because instruments are removed from service for

testing, possibly causing a spurious signal and a plant trip.

Yet, the proposed alternative testing would occur while the plant is in an operating condition.

The distinction between the quarterly testing at power operations and the testing just prior to shutdown for a refueling outage is that if, during testing, a spurious signal results in a plant trip, the plant would remain shutdown for the refueling'outage rather than cycle back to power operations, thereby eliminating the concerns associated with increasing the number of plant transient cycles.

Testing is best conducted while the reactor coolant system is at a pressure above'00 psig because there are no design features to enable testing of the EFCVs in a manner other than by applying a differential pressure across the valve disc to seat the valve.

Testing during a refueling outage can be accomplished during a hydrostatic test of the reactor coolant system, but the time required to test all EFCVs could result in thermal cycles of the reactor vessel which is highly undesirable.

Past testing has resulted in extending refueling outages by 2 days, with the EFCV testing being the critical path activity.

Testing at one of the Code-specified frequencies (i.e., quarterly, during cold shutdowns, or during refueling outages) does not appear to be the optimal frequency for testing the EFCVs.

Though there is some risk in testing these valves at power, there is also some risk in testing the valves during the reactor coolant system hydrostatic test performed during refueling outages.

The risk of testing at power can be minimized by ensuring that the administrative controls are adequate to prevent removing from service more instruments and valves than necessary at any one time, and for restoring a set of valves to service following testing before any further sets are tested.

This testing can be more highly controlled than testing during refueling outages because the reactor coolant system will be under stable operations, whereas testing during a hydrostatic test of the reactor coolant system creates potentially unstable conditions (e.g.,

several depressurizations and repressurizations; thermal cycles).

Therefore, the imposition of testing the EFCVs during only refueling outage conditions is a hardship without a compensating increase in the level of quality and safety.

Testing performed under strict administrative controls just prior to shutdown for refueling outages will optimize the test conditions for these valves and the primary system.

Nore frequent testing could result in unnecessary plant cycles, and testing during refueling outages could thermally cycle the reactor vessel.

3.13.4 Conclusion The alternative testing frequency for the excess flow check valves to once each cycle just prior to shutdown for refueling outages (i.e., within approximately 1 week of plant shutdown) is authorized pursuant to 10 CFR 50.55a(a)(3)(ii) based on the hardship or unusual difficulty that would

ensue, without a compensating increase in the level of quality and safety, if the Code requirements for the test frequency were imposed.

3. 14 Relief Re uest RR-24 This new relief request applies to the ASME Code Class 2 core spray system check valves in the keepfill lines which prevent reverse flow out through the keepfill line.

3. 14.1 Licensee's Basis for Relief The licensee states:

The check valves located in keepfill lines for core spray provide condensate transfer system water flow into their respective

headers, while preventing flow of process water in the reverse direction, during operation of the core spray system.

In the core spray system, a single, test connection exists upstream of the two check valves, which are located very close together.

This configuration supports only dual testing of each pair of core spray system check valves in combination.

Using the test connections in core spray to monitor essential restriction of reverse flow involves collecting radioactively contaminated seepage while the process system is pressurized, as during flow testing.

This creates the potential for spills and spread of contamination.

The increase in potential for water hammer in these systems due to isolation of keepfill lines during testing, the increase in personnel radiation exposure required to perform this testing during plant operation, and the increase in potential for contamination of personnel equipment through this testing justify reduced frequency.

The stainless steel construction of each check valve and the series configuration of each pair of check valves reduce the probability of failure to restrict reverse flow through any keepfill line.

The relatively small size of each keepfill line minimized the impact of any such failure.

The combination of these mitigating factors warrant reduction in testing frequency.

3. 14.2 Alternative Testin The licensee proposes:

Demonstrate.closure of the pair of check valves in each of the keepfill lines by monitoring the essential restriction of their reverse flow, through their upstream test connections, once per refueling outage while the process system is pressurized, as during flow testing.

The acceptance criteria will address the check valves as a pair and corrective action will be taken on both valves if the leakage criteria, as specified in the acceptance criteria, is exceeded.

3. 14.3 Evaluation The NRC has recommended that licensees test the closure capability (and leaktightness, if applicable) of in-series check valves as a pair with the acceptance criteria applicable to corrective actions on both valves (see Section 4. 1.1 of NUREG-1482).

However, the recommendation is applicable to installations where two valves in series are not required by the safety analysis of the plant.

The licensee has not indicated that the safety analysis takes credit for only one valve of the pair and that the second valve is an additional measure of safety.

The licensee has indicated that the basis for the relief is the impracticality of testing both valves individually due to limitations in the design.

The keepfill system operates in a manner such

that these valves are normally open.

Testing the valves as a pair during refueling outages is practical;

however, the testing does not verify that both valves are capable of closing, thereby providing redundancy.

Therefore, unless there is further justification that only one valve is needed to meet the safety analysis, the licensee must establish a periodic disassembly and inspection program for the valves to supplement the periodic testing.

Under the guidance of Position 2 of GL 89-04, a sampling of one valve each refueling

outage, with all valves disassembled within a 6-year period, would be an acceptable frequency to verify the closure capability of these valves.

If the licensee determines that only a single, valve is credited in the safety

analysis, there is no need for the supplemental disassembly and inspection and the proposed alternative will provide an adequate level of assurance of the operational readiness of the valves.

The relief request should be revised appropriately and resubmitted.

3. 14.4 Conclusion If the licensee determines that a single valve would meet the plant safety

analysis, the relief request should be revised to state the basis for the determination.

However, if both valves are required for meeting the safety

analysis, the guidance of Position 2 of GL 89-04 for disassembly and inspection under a sampling program for these valves should be followed. If the guidance of Position 2 is followed, GL 89-04 indicates that the alternative is acceptable and no further NRC approval is necessary.

If the licensee later proposes to follow.Position 2 in part, as if for example extending the interval for inspecting all of. the valves, NRC approval will be required.

The relief request is therefore denied and should be revised based upon further review by the licensee and resubmitted within 120 days from the date of this SE.

3.15 Relief Re uest RR-32 This request applies to the ASHE Code Class 1 containment isolation valves in the main steam system (main steam isolation valves).

Relief from the test frequency requirements is requested.

3. 15.1 Licensee's Basis for Relief The licensee states:

During full power operation, full stroke exercising these valves causes an interruption in steam flow which would induce a reactor pressure transient with increased probability of reactor

scram, main steam line isolation, and SRV [safety relief valves]

actuation.

Compliance with the Code requirement [to test the valve once per 92 days] would result in a hardship without a compensating increase in safety.

~

~

15

3. 15.2 Alternative Testin The licensee proposes:

Full stroke testing will be performed in Operation Conditions 1

(<80Ã power),

2, or 3 preceding or following cold shutdown when power level is low enough to prevent the above mentioned transients.

No reduction from high power levels will be made specifically to accomplish this testing.

This testing is typically done in conjunction with Control Rod Sequence Exchanges.

The frequency of testing will be no more frequently than once per 92 days when plant conditions permit testing.

3. 15.3 Evaluation In Section 3. 1. 1.2 of NUREG-1482, the NRC recommended that valves tested during power ascension from either a cold shutdown or a refueling outage be scheduled on a "cold shutdown" or "refueling outage" frequency, even though the plant could be in an "operating" condition.

The licensee's request would be similar, though testing may also occur during power decreases.

The frequency is considered within the allowed test deferral in ON-10.

This situation differs from the testing frequency discussed in Relief Request RR-23 (see Section 3. 13 above) in that power reduction or power ascension is a

prerequisite for testing the main steam isolation valves whereas the excess flow check valves can be tested during full power operations prior'o beginning any power reduction toward shutdown.

Therefore, the relief request is effectively considered a cold shutdown/refueling outage justification and is considered within the allowed deferral of paragraph 4.2.1.2(g) and (h) of ON-10.

The basis for test deferral meets the requirements of these paragraphs of ON-10.

No further NRC approval is necessary.

4.0 CONCLUSION

For Relief Request RR-3, which is approved pursuant to 10 CFR 50.55a(f)(6)(i),

the NRC has determined that relief may be granted and is imposing the proposed alternative as discussed in the evaluation.

The granting of such relief is authorized by law, giving due consideration to the burden upon the licensee that could result if the Code requirements were imposed on the facility.

Relief Request RR-23 authorized pursuant to 10 CFR 50.55a(a)(3)(ii).

For denied Relief Requests RR-15 and RR-24, the licensee must take actions within 120 days.from the date of this SE to comply with the Code requirements or to seek review and approval of revised relief requests.

New Relief Requests RR-25, RR-26, RR-27, RR-28, RR-29, RR-30, and RR-31 have been developed in accordance with the guidance in Position 2 of GL 89-04 for disassembly and inspection of applicable check valves and are therefore acceptable alternatives as discussed in Section 3.5 above.

RR-32 is considered acceptable under provisions in ON-10 and, therefore, relief is not required;

however, the request is effectively a cold shutdown/refueling outage justification.

Principal Contributor:

P.

Campbell Date:

July 17, 1997