ML20059L458

Sser Supporting Util Changes to Inservice Testing Program

ML20059L458
Person / Time
Site:	Byron
Issue date:	09/14/1990
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20059L457	List: ML20059L453 ML20059L458
References
NUDOCS 9009270021
Download: ML20059L458 (15)

Text

. . . ._ .. _ _ _ _ . _ _ __ . . _ . _ _ _ _

e k

, f( ^

-- [.

hf , UNITED STATES NUCLEAR REGULATORY COMMISSION ENCLOSURE 1-

,[ wassiwoTow, p. c.rosss ,

/ SUPPLEMENTAL SAFETY EVALUATION REPORT BY THE OFFICE OF NUCLEAR REACTOR REGULATION

• RELATED TO THE INSERVICE TESTING PROGRAM AND REQUESTS FOR RELIEF >

BYRON STATION. UNITS'1 AND 2 DOCKET NOS.'STN 50-454 AND STN 50-455 -

1.0 INTRODUCTION

' The Code of Federal Regulations, 10 CFR 50.55a, requires that inservice testing (IST) of 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 and r applicable addenda, except where specific written relief has been re uested by the licensee and granted by the Commission pursut.nt to 10 CFR 50.55a a)(3)(1),

(a)(3) ),-or (g)(6)(i). In requesting relief, the licensee must demonstrate ^

that: the proposed alternative provides an acceptable level of quality and -

safety; (2) compliance would result in hardship or unusual difficulty without a. )

com>ensating increase in the'1evel of quality or safety; or (3) the conformance

. wit) certain requirements of the applicable Code edition and addenda-is 3 impractical for its facility.

TheRegulations,10CFR50.55a(a)(3)(1),(a)(3)(ii),~and(g)(6)(1), authorize the Commission to grant relief from these requirements upon making the

..necessary findings.' The NRC staff's= findings with respect to granting or not

• granting the relief requested as part of'the licensee's IST Program are ,

p contained in the SER dated July 24, 1988 and in this Supplement to the Safety g L

EvaluationReport(SER). This IST program covers the first ten-year inspection 4 interval from September 16.-1985 to September 16, 1995, for Byron Unit 1 and from August ~21, 1987-to August 21, 1997, for Byron Unit 2.

2.0 RESPONSE T0 IST PROGRAM REVISIONS

A. Inservice Testing Program Plan For Pumps. Revision 7 l  !

1. Revision of Relief Request'PR-1 '

1.1 Re' lief Request Number PR-1!

-The licensee has~ requested relief from the requirements of Section XI, Table T h IWP-3100-2, for measurement of pump vibration in units of' displacement and  :

!- proposed to measure vibration in units of velocity using a program patterned t after ANSI /ASME OM-6, Draft 8, as described below for all pumps in the IST  ;

program with,the exception of the essential service water makeup pumps, OSX02PA  ?

and-B, for which specialEacceptance ranges have been developed-and provided. (

1.1.1 Licensee's Basis for Requesting Relief The measurement-of pump vibration is required so that developing problems can  !

be detected ~and repairs initiated prior to a pump becoming inoperable.

Measurement of vibration only in displacement quantities does not take into t

-account frequency which is also an important factor in determining the severity of the; vibration.

L NO k ]

. n . .

7 The'ASME Code minimum standards require measurement of vibration amplitude in mils--(dis placement). Byron Station proposes an alternate program of 7 measuring vbration velocity (inches per second) which-is more comprehensive than that required by Section XI.- This technique is an industry-accepted method which is much more meaningful and sensitive to small changes that are indicative of developing mechanical problems. These velocity measurements '

detect not only high amplitude vibration, that indicate a major mechanical problem, such as misalignment or unbalance, but also the equally harmful low amplitude, high frequency vibration due to bearing wear that usually goes

-undetected by simple displacement measurements.

The allowable ranges of vibration and their associated action levels will l be patterned after the guidelines established in ANSI /ASME OM-6 Draft 8. Table  !

6100-1 and Figure 6100-1. These ranges will be used in whole to assess equip-ment operational readiness for all components except the essential service water makeup pumps, OSX02PA and B.

The acceptable performance range for all components except the essential

_ service water makeup pumps OSX02PA and B will be equal to or less than 2.5 L times the reference value, not to exceed 0.325 inches per second. The alert I range, at which time the testing frequency would be doubled, will be between L 2.5 to 6 times the reference value, not to exceed 0.70 inches per second. Any

vibration velocity greater than 6 times the reference value or greater than u

0.70 inches per second will require corrective actions to be performed on the

! affected component.

The OSX02PA and B pumps, however, are submerged vertical well type pumps utilizing diesel engine prime movers rated at 228 continuous brake horsepower and are coupled directly to these engines through right angle gear drives.

These prime movers and gear drives are elevated to protect them against.the E technical specification postulated flood scenario and the pumps hang vertically L down 41' 4" from the bottom of the right angle gear drive to their respective >

L - pump suction inlet basket strainers. These pumps are water lubricated, 4 hollow line shaft drive, seven stage type and utilize rubber line shaft bearings along their length. ,

L Therefore, these OSX02PA and B pumps, due to their design, will experience p . vibration velocity readings that will normally exceed 0.4 inches per second. A i L . vibration-of this magnitude would fall into the " Alert" range of the ANSI /ASME- '

ll -0M-6, Table 6100-1 and would require " Doubling the Test Frequency." For these ,

reasons, given the inherent operating characteristics of the OSX02PA and B l pumps and the testing constraints of the ANSI /ASME OM-6, Draft 8 document, the i O guidance set forth in Table 6100-1 and Figure 6100-1 will be slightly modified I so as to more accurately assess equipment operational readiness.

l The acceptable performance range of the OSX02PA and B pumps will be equal 1 to or less than 2.5 times the reference value, not to exceed 0.60 inches per seconds.. The alert range, at which point in time the testing frequency would be doubled, will be greater than 2.5 times the reference value, not to exceed j h

0.90. inches per second. Any vibration velocity greater than 6 times the .

reference value or greater than 0.90 inches per second will require corrective actions to be performed on the affected component. This reference value is l

i

.. 3 ,

established based on the guidance set forth in ANSI /ASME OM-6, Draft 8, and.

prior operating history of these pumps. The operating history consist of' approximately 3 years of operation and testing, and provides sufficient data -

base from which to establish the aforementioned criteria. This criteria is based upon a sufficient amount of baseline data to provide for the monitoring pump degradatien to the greatest extent possible given this prime mover / pump ,

configuration.

For all pumps, evaluation of data, to assign equipment to the alert or action ranges, will be done within 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br /> (per IWP-3220 of Section XI). This will ,

be done using industry accepted vibration analysis equipment, such as a full spectrum analyzer. 7 Vibration measurement for all pumps will be obtained and recorded in velocity, inches per second, and will be broadband unfiltered peak measurements. The monitored locatiotn for vibration analysis will be marked so as to' permit subsequent duplication in both location and plane.

The frequency response range of the vibration transducers and their readout

-system shall be capable of frequency responses from one-third minimum pump shaft-rotational speed p at least one thousand hertz. .

The centrifugal pumps in the program will have vibration measurements taken in a plane approximately perpendicular to the rotating shaft in two orthogonal directions on each accessible pump bearing housing and in the axial direction on each acceptable pump thrust bearing housing, The vertical line shaft pumps in the program will have vibration measurements l taken on the upper motor bearing housing in three orthogonal directions, one of ,

which is the-axial direction.

Measurements of vibration in mils displacement are not sensitive to small changes that are indicative of developing mechanical problems. Therefore,-the L proposed alternate method of measuring vibration amplitude in inches / seconds.

L provides added assurance of the~ continued operability of the pumps.

1.1 ~. 2 Evaluation

.Jtilizing vibration velocity measurements rather than vibration displacement measurements has been demonstrated to provide better indication of pump y degradation. The ANSI /ASME OH-6, Draft 8, guidelines for measuring vibration L velocity and determining the allowable ranges and actions levels were published.

I' in OMa-1988, Part 6. These are acceptable to the NRC as an alternative to the l ~ requirements of Section'XI provided the licensee complies with all the Part.6 L -vibration measurement requirements except those for which specific relief-has u

l'

'been requested and granted. The licensee's proposed method for performing pump evaluation using measurement of pump vibration velocity as described above is l equivalent or superior to the Section XI requirements and gives reasonable assurance'of pump operational readiness and provides a reasonable alternative l 'to the Section XI requirements.

i i

4 j

The licensee has also proposed to evaluate pump testing data for assignment of equipment to " Alert" or " Required Action" status within 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br />. The NRC

' staff position-on evaluation test results is presented in NRC Generic Letter (GL) 89-04, Attachment 1, Item 8, " Starting Point for Time Period-in TS ACTION Statements." The licensee's proposal conflicts with the NRC's position.

1Th e licensee is required to follow GL 89-04, Attachment 1, item 8, unless specific relief request.is granted by the NRC. ,

The- essential service water makeup pumps, OSX02PA and B, are submerged vertical well type pumps coupled directly to diesel engine prime movers through right angle gear drives. During normal operation, vibration velocity measurements for these pumps exceed 0.4 inches /second (in/sec), which would place these pumps in the " Alert" range of Part 6 and require doubling the test frequency. .

Therefore, the licensee has proposed modified " Alert" (greater than 2.5 times i the reference value, not to exceed 0.90 in/sec) and " Required Action" ranges )

(greater than 6 times the reference value or greater than 0.90 in/sec) for -l these pumps. -

With normal vibration readings of greater than 0.4 in/see the licensee's I proposed " Alert" limit would be 0.9 in/sec (since greater than 2.5 times the i reference value of 0.4 in/sec equals reater than 1.0 in/sec), which is the '

same as.the absolute " Required Action" limit. This is not considered reasonable since, given the normal high readings, it effectively negates the

" Alert" limit.

l Although the licensee has indicated that these limits are based on three years of, testing experience with these pumps and on the guidance provided in ANSI /ASME OM-6, Draft-8, there is no empirical data provided that supports the ,

=11censee's proposed absolute " Alert" and " Required: Action" limit of 0.9 in/sec.

-The' industry consensus is that pump vibration velocity levels above 0.7 in/sec

-indicate imminent pump failure or severe pump damage. Even though these pumps normally experience nigh vibration readings, the assignment of 0.9 in/sec as- '

o the " Required Action" limit does not seem reasonable since this high limit may .

allow a seriously degraded pump to remain in service without requiring the  !

licensee to take necessary corrective actions.

To ensure increased surveillance of these pumps prior to their measured vibra- '

tion velocity exceeding 0.7 in/sec (" Required Action" limit of Part 6) the licensee should assign an absolute value for the " Alert" range that is appro-priate for these pumps. This value should provide a reasonable margin between the normal' vibration readings and 0.7 in/sec and be consistent with historical pump operating data. Further, the licensee should justify their proposed

" Required Action" limit or propose and justify a more restrictive limit for

'these pumps. The licensee should provide empirical vibration test data,

-possibly from the vendor, that supports the selected " Required Action" limit.

The data should demonstrate that the proposed (0.9 in/sec), or a more restric-tive limit wi11' provide a reasonable assurance of operational readiness.

Based on the determination that it is impractical to take vibration measure-ments on the OSX02PA and B pump housings and that it would be burdensome to require the licensee to comply with the maximum vibration velocity limits

I

,L ,

l

i

• )

specified in Part 6, and considering the proposed alternate testin ' elief may  !

begrantedfromtheCoderequirementspursuantto10CFR50.55a(g)g,'(1)for (6  ;

pumps OSX02PA and B provided the licensee complies with the above provisions 1 within six months of the receipt of this SE. l Based on the determination that the licensee's proposal to measure vibration  ;

velocity for all pumps in the IST program in accordance with the requirements 4 of OMa-1988, Part 6 is equivalent or superior to the Code requirements for pump vibration displacement measurement, relief may be granted from the Code vibra--

• . tion testing requirements pursuant to 10 CFR 50.55a(a)(3)(1) for all pumps in the IST program, provided the licensee evaluates pump test data within the time constraints specified in Gl. 89-04, Attachment 1, Item 8.

.2. Deletion of Relief Request PR-3 from the IST Program 2.1 Evaluation The licensee deleted relief request PR-3 from the IST program. This change-is i acceptable,

3. Addition of Note 5 to-the IST Program 3.1 Evaluation Note 5 was added to the program to describe how inlet pressure will be determined for the essential service water makeup pumps and is acceptable.

4. Boric Acid Transfer Pumps The boric acid transfer pumps are not included in the Byron Station IST.-

program. The staff's evaluation,.however,-indicates that the boric acid-transfer pumps are needed.to maintain the plant in a safe shutdown condition.

following a design basis event involving a tornado causing the failure of the reactor water storage tank. The licensee is, therefore, required to include the boric acid transfer pumps in the IST program unless additional information

is provided which can justify omitting these pumps from the IST program.

B. Inservice Testing Program Plan For Valves, Revision 8

1. Revision of Relief Request VR-1 l

1.1 Relief Request Number VR-1 The licensee has requested relief from leak testing the primary containment L

isolation valves in accordance with requirements of Section XI, Paragraph IWV-3420,-and proposed to leak test these valves in accordance with the require-ments of 10 CFR 50, Appendix J, and to apply the analysis and corrective action L criteria of Section XI, Paragraphs IWV-3426 and -3427.

u b

L u

., ,q ,

i 1.1.1 -Licensee's Basis for Requesting Relief Primary containment isolation valves will he seat leak tested in accordance i with 10 CFR 50, Appendix J. For these valves,Section XI testing requirements are essentially equivalent to those of Appendix J. No additional information concerning valve leakage would be gained by performing separate tests to both Section XI, and Appendix J. Therefore, overall plant safety is not affected.

Alternate Testing Primary containment isolation valves shall be seat leak rate tested in accor- ,

dance with the requirements to 10 CFR 50, Appendix J. The results of such leak rate measurements shall be analyzed and corrected, as necessary, in accordance with the guidance set forth in ASME Code,Section XI, Subsection IWV, paragraphs IWV-3426 and IWV-3427.

1.1.2 Evaluation i

l The licensee's proposal complies with the NRC staff's position of testing containment isolation valves in NRC GL 89-04, Attachment 1, Item 10, therefore, L

pursuant to 10 CFR 50.55a(a)(3)(1) relief is granted from the requirements of Section XI, Paragraphs IWV-3421 through -3425 since the licensee's alternative provides'an acceptable level of safety. '

2. Revision of Relief Requests VR-4, VR-10, and VR-16 1

2.1 Relief Request Number VR-4 j The licensee has requested relief from exercising valves 1(2)CS008A and B, containment spray (CS) rin L CS pump discharge checks (g header Category C), inchecks (Category accordance with theA/C), and 1(2)CS003A requirements of- and B,Section XI, Paragraph IWV-3521, and proposed to verify valve operability by L.

part-stroke exercising valves 1(2)CS003A and B quarterly and by performing either a full-flow test.or check valve disassembly and inspection on all these valves during refueling outages.

2.1.1 Licensee's Basis for Requesting Relief These valves cannot be full flow tested-as a matter of course during unit-o)eration or cold shutdown because water from the CS pumps would be discharged tirough the CS ring headers causing undesirable effects on system components inside containment. Additionally, the full flow testing of these check valves during periods of cold shutdown, using the CS pumps, would fill the reactor refueling cavity with' contaminated water from the refueling water storage tank.

The filling of the cavity, via temporarily installed large bore piping, would i require the removal of the reactor vessel head so as to preclude equipment

-damage from borated ater. <

Currently, full flow recirculation flow paths do not exist from the'ischarged at the CS pumps through these check valves to the ' refueling water storage tank.

L The addition of such flow paths would required extensive plant modifications to existing plant designs, including penetration of containment integrity.

1

~ .- - - . - . . . - --- .

1,. ,

a 7-

' Partial stroking of the 1(2)CS008A, B valves using air does not provide ade -

quate assurance of valve operability and may be detrimental for the following reasons:

a.. There is no correlation between air flow and angle of disc movement, e

b. Venting and draining the appropriate )iping quarterly may cause.

deposition of boric acid residue whic1 could in turn promote binding of the check valve internals.

Alternate Testing The~1(2)CS008A_andBvalveswillbeeitherfull-flowtested,ordismantledto demonstrate operability each refueling outage. The 1(2)CS003A and B valves will be partial stroke tested during the quarterly pump surveillance and full q flow tested, or dismantled, to demonstrate operability, each refueling outage, '

This alternative will adequately maintain the system in a state of operational readiness, while not sacrificing the safety of the plant, by testing the valves as often as safely possible. 1 a

2.1.2 Evaluation L Checkvalves1(2)CS008AandBarecontainmentisolationvalvesinthesupply i lines to the. containment spray headers. It is impractical to full-stroke exercise these check valves quarterly during power operation or during cold shutdowns without making significant system modifications, including recon-figuring 10" diameter piping sections utilizing spoolpieces, since flow through i these valves would spray into the-containment and could damage equipment.

Provisions exist-to verify the full-stroke capability of these valves with flow

(. utilizing piping spool)iece hookups and the-containmcnt spray _ pumps during cold ~

p shutdowns. However, t11s _is a very complicated and time consuming procedure L 'and would result in flooding the reactor refueling cavity, which would be L

burdensome-to the: licensee and is impractical to perform each cold shutdown.  ;

Thelicenseehasproposedtofull-strokeexercise. valves 1(2)CS008AandB open by either disassembly and inspection or full flow test each refueling -v outage. The Minutes of the Public Meetings on Generic letter 89-04 state that disassembly and inspection should-be used only when full flow testing is

impractical. Whereas it presents difficulties to full-stroke exercise these i valves with flow each refueling outage, this is obviously not impractical.  !

L These valves are located inside containment in a section-of dry pipe and are. f not exposed to many of the mechanisms that typically cause check valve '

degradation such as those' associated with unstable flow, improper water chemistry, etc. These and other considerations may suitably justify an extension of the test frequency for the full flow test _from each refueling

' outage to some longer interval.  ;

i I Full-strokeexercisingcheckvalves1(2)CS003AandBwithflowduringpower operation or cold shutdowns would require system modifications, which would be expensive and burdensome to the licensee. A test line is used during quarterly pump testing and allows a significant part-stroke exercise of these valves with

-flow.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ __.~4 ., ..-. . - . .

. 1 ,  :

i The licensee has proposed, to perform a part-stroke exercise of check valves .

1(2)CS003A and B quarterly during pump surveillance testing, utilizing the recirculation' flow path, and further, to perform a full flow test or valve disassembly and inspection each refueling outage, however, disassembly and inspection.should be used only when full flow testing is impractical. If using  !

, non-intrusivediagnostic' techniques (e.g.,ultrasonicsandacoustics)can verify that the valve disk is contaci.0ng the back stop (full-stroking) during.

the quarterly exercise, even temporarily, this might be employed to verify tull-stroke exercising. Full-stroke exercising these valves open each  ;

refueling outage with flow should give reasonable assurance of operational readiness and provide a reasonable alternative to the Code frequency require-ments.

7 1

Based on the determination that it is impractical to test these valves in 4 accordance with the Code frequency requirements, and considering the licensee's proposed alternative, and the burden on the licensee if the Code re wereimposed,reliefmaybegrantedpursuantto10CFR50.55a(g)(6)quirements (1) for valves 1(2)CS008A and B provided the licensee full-stroke exercises these valves with flow each refueling outage in accordance with GL Attachment l 1, Position 1. Relief may be granted pursuant to 10 CFR 50.55a(g)89-04(6)(1)for '

valves 1(2)CS003A and B provided the licensee full-stroke exercises.these valves with flow each refueling outage if >ositive non-intrusive diagnostic '

techniques cannot adequately demonstrate tie full stroke capability of these valves during quarterly testing.

2.2 Relief Request Number VR-10 Thelicenseehasrequestedrelieffromexercisingvalves1(2)lA065and 1(2)lA066, instrument air containment isolation valves, quarterly or during -

cold shutdowns _in accordance with the requirements of Section XI, Paragraph a b IWV-3411, and proposed to full-stroke exercise these valves during refueling outages.

2.2.1 Licensee's Basis for Requesting Relief Stroke testing of these valves during plant operation or cold shutdown would,-

'by design, isolate the air operated instruments and valves inside the contain-ment building. The full-stroke exercising of the instrument air containment 1 isolation valves during unit power operations or cold shutdowns, introduces the possibility of causing mjor operating perturbations and/or personnel safety concerns should these valves fail to reopen during testing activities. The tailure of these valves in the closed position, as a result of testing acti-

-vities during plant operation or cold shutdown, would isolate the air operated instruments and valves inside the containment building, thus, resulting in one or more of the~following scenarios:

a) Loss of Pressurizer Pressure Control The pressurizer spray valves 1(2)RY455B and C and the pressurizer auxiliary spray valve 1(2)CV8145 would fail closed and not be available  !

-_for pressurizer pressure control.

wm - - - - -

.u- =2x_r - n_m _.nn - - -'

~ _:. _ _

t i

i

' b) Loss of Chemical Volume Control System Let Down Flow (Both Normal and Excess) and Charging Flow.

The. loss of instrument air would cause a disruption in the unit letdown flow paths resulting in pressurizer level increases. Such valves as the letdown orifice containment outlet header isolation valve 1(2)CV8160, the-letdownlineisolationvalves1(2)CV459and1(2)CV460,theletdownorifice 4

outletisolationvalves1(2)CV8149A B, and C, the excess letdown heat '

exchangerinletisolationvalves1(2)CV8153AandB,andtheregenheat '

exchangerletdowninletisolationvalves1(2)CV8389AandBwouldgoto ,

their fail closed positions. Additionally, the ability to normally make ,

up reactor coolant inventory and adjust the ' reactor chemical shim (i.e.,

normal boration/ dilution) would also be lost as the regenerative heat exchanger inlet isolation valves 1(2)CV8324A and B would fail to their respective closed positions.

c) Loss of Component Cooling to Containment Penetrations The loss of instrument air supply would cause the penetration cooling I supply flow control valve 1(2)CC053 to go to its fail closed position.-  !

The loss of penetration cooling would result in elevated temperature being imposed on the penetrations being supported by the component cooling system.

d) Loss of Personnel Breathing Air t

The loss of instrument air supply to the service air downstream isolation valve 1(2)SA033wouldcausethisvalvetogotoitsfail-closedposition.

This loss of. service-air in the containment building would eliminate the t- normal source of supplied breathing air needed to support numerous-maintenance-and component inspection activities in a' contaminated ,

L radiological environment.

Alternate Testing These valves will be exercised during refueling outages. This testing period will- be each refueling outage as a minimum, but no more-frequently than once per quarter.

2.2.2 Evaluation

. Exercisin valves 1(2)lA065 and 1(2)lA066 quarterly durin'g power operation is

• impractic 1 because the. loss of instrument air would disrupt normal pressurizer  ;

pressure control valve operations by causing a loss of operating air to several

. control valves and possibly result in a reactor trip. Exercising.these valves 3

. during cold shutdowns-would-cause the service air downstream isolation valve to  !

fail closed and would disrupt the supply of breathing air to workers in contain-ment. Further, loss of instrument. air to containment would inhibit the perforinance of shutdown maintenance activities and inspections and could be burdensome to the licensee.. Therefore, testing these valves quarterly during power operations or cold shutdowns is not practical. Imposition of the Code '

requirements to test' these valves quarterly would require significant system a

s

'-__-.--_L_-.____-----------------.---------

l ' $

t ,

t modifications which would be burdensome to the licensee. The licensee's proposal to exercise these valves during refueling outages should give reason- ,

able assurance of their operational readiness and provides a reasonable alter-

= native to the Code requirements.  !

Based on the impracticability of complying with the Code requirements, the burden on the licensee if the Code requirements were imposed, and considering the licensee's proposed alternate testing frequency, the licensee may be granted'reliefpursuantto10CFR50.55a(g)(6)(1),

m 2.3 Relief Request Number VR-16 >

Thelicenseehasrequestedrelieffromexercisingvalves1(2)SI8811AandB, containment sump outlet isolation valves, quarterly or during cold shutdowns in accordance with the. requirements of Section XI Paragraph IWV-3412, and proposedtofull-strokeexercisethesevalvesdurIngrefuelingoutages.

2.3.1 Licensee's Basis for Requesting Relief The full-stroke exercising of valves not stroked quarterly is required to be performed during cold shutdowns. However, the stroking of the Containment Sump p Outlet Isolation Valves, 1(2)S!8811A and B, requires the suction of the Residual Heat Removal pumps to be drained, thus rendering one train of the <

system inoperable. '

For cold shutdown operations with the Reactor Coolant Loops filled and one-train of Residual Heat Removal declared inoperable, Byron Station's Technical Specifications require two steam generators with secondary side narrow. range ,

water leve_1 greater than 41% (Unit 1) and greater than 18% (Unit 2). However,-

if the cold shutdown condition was necessitated by a problem re of the secondary side of the Steam Generators (i.e. tube ' leaks) , Byron quiring draining Station's Technical Specification 3.4.1.4.1 would )reclude such testing of the

, containment sump cutlet isolation-valves until suc1 time as the affected-steam

(

generators had been refilled.

L L For Cold Shutdown operations with the Reactor Coolant Loops not filled b (i.e. drained down to support Reactor Yessel Incore Seal Table. Loop Stop

. Valve, Reactor Coolant Pump and Seal Maintenance or primary leakage), Byron Station's Technical Specification 3.4.1.4.2 wc:1d preclude the testing of the Containment ~ Sump Outlet Isolation Valves as it mandates that= two residual heat removal (RHR)loopsshallbeoperableandatleastoneRHRloopshallbein

, operation.

L' The full-stroke testing of the 1(2)SI8811A and B valves, in conjunction D

refilling and venting of each train, accounts for an-Ewithsystemdraining[3dayspertrain)ofschedulingrequirementsadi additional six days n ncreased radiation dose to operators and radiological control personnel. Processing of thousands of gallons of contaminated water and subsequent required liquid effluent discharges wtv 1 also result from the draining, refilling and-venting L of the RHR system. This time duration required to perform the surveillance-l~ testing of the Containment Sump Outlet noiation Valves during Cold shutdown j activities,-could, as a result, cause a violation of the action requirements H

i

! V.

% =.i*- ,

.t t

for Byron Station's Technical Specifications 3.4.1.4.1 and 3.4.1.4.2. -The violationswouldoccursincetheseactionstatementsrequire(asnotedintheir respectivefootnotesections)thereturnoftheinoperableresidualheat  ;

removal loop to service within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, if such loop was removed for  !

surveillance testing provided the other RHR Loop is operable and in operation.

In addition, NRC Generic Letter 88-17, Loss of Decay Heat Removal, '

highlights the consequences of a loss of RHR during reduced Reactor Coolant -

System inventory (below three feet below the reactor vessel flange). If the 1 operating RHR pump is lost due to air entrainment, and the other train is ,

inoperable for the stroke test, tlien the " operable" train must be vented to i restore decay heat removal. .Under worst conditions, boiling in the core would 1 occur in approximately 10 minutes, the core would be uncovered in approximately .

30 minutes, and fuel damage would occur in approximately-1 hour. 4 1

Given the apparent disparity between the Technical Specification time  :

requirementsforreturningan.inoperableRHRLooptoservice(2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />)andthe time required to perform surveillance stroke testing of the Containment Sump  ;

OutletIsolationvalves(3 days)-duringColdShutdown,theproposedalternate testing frequency of refueling outage periodicity will adequately maintain the system in a state of operational readiness, while not imposing undue hardships or sacrificing the safety of the plant.

Alternate Testing ,i P

, Byron Station will full stroke exercise the Containment Sump Outlet Isolation  !

Valves, 1(2) SIB 811A and B, during refueling outages vice cold shutdown.

'2.3.2 Evaluation Yalves 1(2)S18811A and B are 24-inch motor operated valves in the common suction line from the co-tainment sump to the RHR and containment spray pumps.

This line is also supplir J by the refueling water storage tank (RWST). Opening i these valves would result in the diversion of RWST water into the containment ,

sump, therefore, exercising these valves requires isolating-the RWST and L draining the common suction line to preventLflooding the containment sump.-

L

'~

Performing this testing'at cold shutdowns would be burdensome to the licensee since this renders at least one train of the RHR system inoperable, may result ,

( in delaying-restart of the reactor from cold shutdown, and creates large b

. amounts of-liquid-radioactive waste, which must be processed. For these reasons, it is impractical to exercise these valves quarterly during power j operation or cold shutdowns. Exercising and stroke timing these valves during I each' refueling outage should give reasonable assurance of operational-readiness l

L and provide a reasonable alternative to the Cade requirements.

L, l L Based on the determination that the Code requirements are impractical and i since the licensee's proposed testing frequency provides a reasonable l alternative to the Code requirements, the licensee ray be granted relief j pursuantto10CFR50.55a(g)(6)(1).

l

3. Revision of Relief Request VR-12 #

3.1 Relief Request Number VR The licensee has' requested relief from comparison of st~r oke times for the following listed valves in accordance with the requirements of Section XI, Paragraph IWV-3417(a), and proposed a fast-acting stroke time limit of 2 '

seconds; upon exceeding 2 seconds, corrective action shall be taken in accordance with Section XI, Paragraph IWV-3417(b).

Valve Identification Yalve Identification 1 MS018A-D 1 RE9159A and B i 1 PS228A and B 1 RE9160A and B-l 1 PS229A and B 1 PR066 t- '

1 PS230A'and B 1(2518871 l 1 RC014A-D 1(2 RY8033 1 RE9157 3.1.1 Licensee Basis for Requesting Relief Minor timing inaccuracies, with small stroke times can lead to substantial increases (percent wise) in stroke times. For example, a valve with a stroke time of I second in an initial test,.and 1.6 seconds in the subsequent test, has experienced an apparent 60% increase in stroke time. If the accuracy  !

requirements of IWV-3413(b) are utilized, it could be argued that stroke times between 1 and 2 seconds could constitute as much as a 100% increase in stroke time when, in f act, only a 0.2 second increase occurred. For instance, if the initial time was 1.4 seconds,-(measured to the nearest second is 1.0 second) andifthenexttimeisthen1.6 seconds,(measuredtothenearestsecond_is 2.0 seconds)thepercentincreaseis100%.

Alternate Testing u First-acting valves can be defined as those valves that normally stroke in 2 l seconds-or less. No trending of stroke time will be required, and upon ,

L

' exceeding 2 seconds, corrective action shall be taken immediately in accordance ;

withIWV-3417(b). For small stroke times, the trending requirements are too '

stringent for the accuracies specified in the Code. The alternative specified

will adequately maintain the system in a state of operational readiness, while not imposing undue hardship-or sacrificing the safety of the plant.

3.1 2 Evaluation The licensee's proposal complies with the provisions of NRC GL 89-04, Attach-ment 1, Item 6, therefore, pursuant to 10 CFR 50.55a(a)(3)(1) relief is granted from the ree.uirements of IWV-3417(a) as requested since the licensee's alter-  :

native provides an acceptable level of safety. '

w _.

t m

E 13-

4. Revision of Relief Request VR-15 4.1 Relief Request Number VR-15 The licensee has requested relief from exercising the following valves quarterly or during each cold shutdown in accordance with the requirements of Section XI, Paragraph IWV-3522, and proposed to full-stroke exercise these valves during cold shutdowns if greater than nine months have passed since the last test.

Valve Identification Function 1(?)CV8481AandB Centrifugal pump discharge check valves 1(2)CV8546 Centrifugal charging pumps' suction check veln from the refueling water storage tank 1 SI8815 Charging pump discharge to cold leg check valve 1 SIB 841A and B Hot leg safety injection check valves  !

.1 SI8900A-D Charging safety injection check valves .

- NOTE: Valves 1(2) SIB 819A-D,1(2)SI8905A-D,and1(2)S!8949A-0,which are included in this relief request are not addresse.i in this i

evaluation. These valves are addressed in the NRC SER oc the inservice testing program for Byron Station, Units l'and 2, dated July 29, 1988, and relief is granted as stated therein.

4.1.1 Licensee's Basis for Requesting Relief '

The full-stroke exercising of check valves not stroked quarterly is required to

be performed during cold shutdowns. However, the stroking of check valves L

1(2)S18815,1(2)SI8900A-D, and 1(2)SI8841A-B, associated with emergency core cooling system, during cold shutdowns will induce thermal stresses on their l'

respective reactor ~ vessel nozzles because the reactor coolant system (maintained approximately 180'F) is injected with water from the refueling

- water storage tank (maintained approximately 65'F). This also applies to the

, - stroking of check valves 1(2)CV8546 and 1(2)CV8481A and B because the full P stroke of these check valves causes stroking of the'1(2)S18815 and L 1(2)SI8900A-Dlocatedinthefullflowpath.

In addition to the stroke test exercise used to verify operational readiness of these check valves, the act of such stroking cause the necessity for Technical Specification required leak rate testing of these valves prior to unit criti- 1 cality. This testing, in conjunction with the stroke exercising of these check valves, adds approximately one week to the duration of any outage and addi- <

l tional radiation exposure to workers who must connect flowmeters and  !

differential pressure gauges directly to pipes containing radioactive fluids. I Strokeexercisingthe1(2)CV8481AandB,1(2)CY8546,and1(2)SI8815, 1(2)SI8900A-D, and 1(2)SI8841A and B check valves on the same schedule as the required Technical Specification Reactor Coolant System Boundary Isolation leak i rate testing will allow the coordination of testing activities without imposing  !

additional check valve leak rate testing requirements. Such activity coordi-l: nation will optimize testing efforts and resources while adequately maintaining the system in a state of operational readiness.

1

. ' k y . .

f.y

==_ a.n==. -

nr1 1 ._ n _. "-

1 Alternate Testing-Byron Station's Technical Specifications require routine leak rate testing to be performed on these Reactor Coolant System Boundary Isolation check valves if ,

the unit is in Cold Shutdown for greater than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and such leak rate  ;

e i

testing has not been performed within the previous nine months. Therefore,

- C ron Station will stroke exercise check valves 1(2)SI8815,1(2)SI8900A-D, and .

1 2) SIB 841A and B on the same schedule. To arevent unnecessary stroking of check valves 1(2)S18815 and 1(2)SI8900A-D, c1eck valves 1(2)CV8546 and  :

1 CV8481A and B will be stroke exercised on the same schedule as check valves 1 S18815,1(2)SI8900A-D,and1(2)S!8841AandB. Full stroke exercising of these check valves will be performed at a minimum frequency of once each refueling outage.

.4.1.2 Evaluation i

a Valves 1(2)SI8815,1(2)SI8841AandB,and1(2)SI8900A-D,arereactorcoolant I system pressure boundary isolation vah n in the ECCS injection system lines. .

Valves 1(2)CV8481A and B are discharp a:eck valves for the centrifugal 1

charging- pumps and receive a part stroke exercise quarterly durIng operation. +

E Valve 1(2)CV8546 is the combined suction check valve for the centrifugal charging The act of opening  ;

valves 1(pumpsfromtherefuelingwaterstoragetank.-2)S18815,1(2)S18841A and B, test per Byron Station Technical Specifications. It is impractical to ,

L full-stroke exercise charg ng system check valves 1(2)CV8546 and 1(2)CV8481A '

h and B each cold shutdown s nce this would result in opening valves 1(2)SI8815 and 1(2) SIB 900A-D, which are located in the only full flow path and would ,

require a leak test prior to returning- the plant to power. It is impractical  !

to full- or part-stroke exercise these valves, with the exception of valves i e

1(2)CV8481A and B which receive part-stroke exercising quarterly during power operations as this would result in charging-cold water through the reactor vessel injection nozzles causing thermal stresses in addition to possible reactivity addition and resultant power fluctuations. Full-stroke exercising F all these valves during each cold shutdown would be burdensome since it would require the leak testing of valves 1(2)S18815 and 1(2)SI8900A-D at each cold shutdown and result in additional radiation exposure to workers connecting and

~ disconnecting the leak testing equipment and may delay the plant's return to power operation.

Thelicenseehasproposedtopart-strokeexercisevalves1(2)CV8481AandB ,

quarterly, in the valve tables section of their IST program. The licensee has "

l also proposed to full-stroke-exercise these valves at the same frequency as

-valves 1(2)SI8815 and 1(2)SI8900A-D are leak tested, per plant Technical Specifications, and at each refueling outage. This testing anr' ' ncy should give reasonable assurance of operational readiness and provide 3 Sie alternativt to the Code requirements.

I Based on the impracticability of complying with the Code reqr .nd considering the burden on the licensee if the Code requiremens rosed,

.suant to

-and the licensee's p(roposed alternative test, relief may be grt.,10 CFR 50.55a(g I

n .

- . - -=-- -

n: .. :.

. . . . - --- : -.. *s p  :.+ -

- 15-

1. 1 exercised quarterly during power operation, and all these valves are full-stroke exercised during each cold shutdown when' valves 1(2)SI8815, '

1(2) SIB 841AandB,and1(2)S!8900A-Dare'leaktestedandduringeachrefueling i outage. ,

3.0: CONCLUSION The Byron Station, Units 1 and 2. IST program changes submitted in letters dated December 21, 1988, and January 3,1989, and discussed above are acceptable

+ for implementation with the exceptions and/or provisions identified in this  ;

Supplemental SER. New or revised relief requests contained in any subsequent revisions-may not be implemented without prior approval by NRC unless they are relief requests meeting the positions in Generii Letter 89-04, Attachment 1, 1

l

~

l 4

l.

b.

u

\

i o

0 e

3

- - -_ _ _ , . - - - - ___ # _,_ ._- - . - ,. , _ , , , , - ,.,- ,_