ML20236K937

Safety Evaluation Supporting Util Request That RHR Sys Be Inoperable for 24 H in Mode 5 in Order to Repair Common Valve

ML20236K937
Person / Time
Site:	Sequoyah
Issue date:	11/02/1987
From:	NRC OFFICE OF SPECIAL PROJECTS
To:
Shared Package
ML20236K915	List: ML20236K914 ML20236K937
References
NUDOCS 8711100029
Download: ML20236K937 (6)

Text

-,

'.- .,a ,

i a

)

L l~  !

[ b' .i UNITED STATES -

NUCLEAR REGULATORY COMMISSION . .,

l

r. p wAsMNGTON, D C. 20555 - I e

< s

)

'l SAFETY EVALUATION BY THE'0FFICE OF SPECIAL PROJEC1S

.)

p ON THE RHR SYSTEM BEING INOPERABLE FOR 24 HOURS IN MODE 5 3 TENNESSEE VALLEY AUTHORITY' SEQUOYAH NUCLEAR PLANT, UNIT 2-DOCKET NO. 50-328-

1.0 INTRODUCTION

This evaluation documents the staff's review of the request from the Tennessee Valley Authority (TVA, licensee) in its letter dated October 12, 1987,sto make the 2-loop. residual heat removal (RHR) system inoperable in ModeL5 for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, i TVA requested staff approval because, during this time, the.RHR' system will not 1 be in compliance with Limiting Condition for Operation (LCO) 3.4 6 ef thel -

1 Sequoyah Unit 2 Technical Specifications' (TS). : LCO 3.4.1.4 requires inat two-  !

RHR loops shall be operable and at least one RHR loop shall .berin operation in Mode 5 (cold shutdown). This is page'3/4 4-5 of the TS. The plant is-in Mode 5

.and would remain in Mode 5 while the.RHR. system is rendered inoperable.

The 1kensee proposes to make both RHR loops inoperable for a period expecte'd-to be no more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.. This would allow'the licensee to repair;a valve (2-FCV-74-2) which is common to both RHR~ loops and, therefore, both' loops-must-be isolated and made inoperable to repair the valve.

4 There were phone calls on October 16 and 19, 1987.to the licensee to-discuss and clarify statements made in its October 12, 1987 submittal.

2.0 BACKGROUND

l The RHR system transfers heat from the Reactor Coolant' System (RCS) to.the ,

Component Cooling System. During normal plant operations, the RHR: system is, used to remove decay heat from the core and reduce.the temperature of the reactor coolant to the cold shutdown temperature, at a controlled rate, during the second phase of plant cooldown and' maintains this-temperature until the -

plant is started up again. The RHR' system:also serves as.part of the Emergency Core Cooling System during the injection and' recirculation phases of a' loss'of coolant accident. The RHR at~Sequoyah is an engineered safety feature system which provides cooling to'the RCS at' low reactor pressure.- '

1 The RHR system is'also used to transfer refueling water between the refueling i

.. water storage tank and the refueling cavity before and after the refueling.

operaticns. . This wou'id not be needed to be done' under the proposed ' action.

8711100029 871102 PDR. ADOCK 05000328 p; .

.PDR

. .i i l

l In its submittal, the licensee stated that the RHR isolation valve 2-FCV-74-2 l

has a packing leak caused by a bent stem. Although the leakage is small j

(drops per second), the leakage can no longer be stopped by tightening the valve packing and the leakage of boric acid is corroding the valve bonnet. l The bent stem is damaging the valve packing each time the valve is operated. I I

Necessary maintenance work to correct the packing leak requires the stem to be replaced and the valve repacked. In addition, the bonnet bolting will be i inspected and replaced as necessary. l The repair work would be done by isolating the RHR system and valve 2-FCV-74-2 .

l from the RCS. This would be done by closing valve 2-FCV-74-1 which is a safety grade, RHR isolation valve, and RHR valves 2-FCV-74-21, 2-FCV-74-3, 2-FCV-63-1, 2-74-507 and 2-74-504. The RHR pumps and heat exchanges are not between these isolation valves. The RHR lines between these valves would be drained of its water and the valve bonnet removed.

The alternative to this proposed action would be to perform the repair with the reactor defueled. This condition can occur during refueling outages. The licensee has concluded it is not prudent to wait until the next refueling outage because of the increasing nature of the leak. The licensee also determined that it is not prudent to defuel the reactor for this repair. It stated that personnel radiation exposure to unload and reload the core would be approximately 80 to 100 man-rem based on exposure data from past operations. This includes exposure from reactor head removal and replacement, expected maintenance work on the fuel transfer system, and other routine and miscellaneous work. The potential for an accident involving heavy loads over the reactor core would be increased by the additional removal and replacement of the missile shield, upper internals, the reactor head, and the handling of the 193 fuel assemblies in the reactor vessel.

The TVA submittal provided a safety analysis, a copy of the operations procedure that would be used to prepare the RHR system for maintenance work, monitoring and controlling the RCS during maintenance, and place RHR back in operation upon completion of work and a copy of the special maintenance instruction that would be used to perform this work.

The licensee stated that the resultant cooldown and possible RCS inventory additions will not violate the intent of technical specification action statement prohibiting reductions in boron concentrations or position reactivity insertions. The boron concentration necessary to maintain a k of 0.95 for unit 2 is 1,807 parts per million (ppm). BoththeRCSandRefbingWater Storage Tanks (RWST) boron concentrations will be maintained greater than 2,000 ppm. The present RCS concentration is slightly greater than that in the RWST. However, sufficient boron shutdown reactivity is present to offset any slight positive reactivity insertions caused by a cooldown after the valve repair or by boron concentration changes during feed and bleed cooling operation.

G 3.0 EVAL.UATION 3.1 Introduction The licensee has a packing leak caused by a bent stem in an RHR valve (number 2-FCV-74-2) which is common to both RHR loops in Sequoyah Unit 2. This is .

shown is Figure 5.5.7-1, Flow Diagram Residual Heat Removal System, in Volume 5 of the Sequoyah Nuclear Plant Final Safety Analysis Report Update (FSAR).

Section 5.5.7 of the FSAR is on the RHR system. The leakage has increased each time this valve has been cycled. Although the leakage is now small, the licensee cannot stop the leakage by tightening down on the valve packing, the leakage (boric acid) is corroding the valve slowly and the leakage will increase with valve operation. This leakage is a loss of coolant from the RHR which is cooling the core in the present Mode 5, cold shutdown. Sufficiently high leakage could cause the RHR to be declared inoperable, i The licensee could defuel the reactor to repair the valve because the RHR is required in Mode 5 and Mode 6 (refueling) but is not required when the reactor is defueled. Sequoyah Unit 2 is presently in Mode 5 with fuel in the reactor.

The licensee considers it imprudent to wait until the next refueling outage and to defuel the reactor for this repair work. The licensee's considerations are there is no need to refuel now, the occupational exposure to workers for removing and then adding back the fuel in an additional refueling, and the potential for an accident involving moving heavy loads in the containment.

These considerations are avoidable because, as discussed below, there may be a basis to safely repair the valve without defueling the reactor.

The staff concludes that this valve should be repaired now and there is i sufficient basis to consider repairing the valve in Mode 5 without defueling the reactor.

3.2 Core Cooling The Sequoyah Unit 2 reactor shut down in August 1985 and has not operated since then. The fuel in the reactor vessel has decayed more than 700 days.

The licensee would not have the core go critical in this proposed action nor would the core generate any power.

In its submittal, the licensee provided calculations of the heatup rate of the reactor coolant for 700 days after shutdown and measurements of the heatup rate of the actual care from a test on October 3, 1987. The calculated heatup rate of the reactor coolant is 3.64 F/hr end the measured rate is 1.5 F/hr. The staff has reviewed the calculations and measurements and considers them acceptable.

This range of heatup rates,1.5 to 3.6 F/hr, is low. The licensee would continually monitor and record the temperature of the RCS at least every 30 minutes (15 minutes in its repair procedure) and the actual heat rise rate would be calculated to verify the rate and to take actions to prevent the RCS temperature from going above 200 F.

The licensee has committed to keep the RCS temperature below 200 F. This is written into its procedure provided in its submittal. The licenset is doing this to prevelt the reactor from entering Mode 4 (RCS temperature is greater than 200*F) The procedure has action statements at RCS temperatures of 150 F, If;0 F and 1.0 F.

• * ,- l j 1

The licensee has evaluated other means to provide cooling to the core while i the RHR is out of service. These are the following: normal letdown and i 3

charging, reactor coolant feed and bleed using the charging pumps, and the steam generators (SG). The RHR could also be returned to service by bolting a temporary cover plate over the valve body and returning the system to  :

service. The licensee has incorporated these methods to provide core cooling ]

in its procedures to repair the valve: (1) below 150 F, by adjusting the let-down/ excess letdown flow; (2) exceeds 160*F, initiate feed and biced using a charging flow from the RWST; (3) exceeds 170"F, install cover plate over valve body or install valve bonnet; and (C exceeds 190 F, use SG. At least one train of motor-driven auxiliary feed ,ter piping and components would be operable and the SG would be ava117 a. The reactor cociant pumps would be used to keep the RCS coolant mixed id, if necessary, for circulating the coolant through the SG.

The licensee calculated that using the conservative 3.64 F/hr heatup rate, the j estimated t.ime to reach 200 F using the methods given above is 46 hours5.324074e-4 days <br />0.0128 hours <br />7.60582e-5 weeks <br />1.7503e-5 months <br />. This J is 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> longer than the licensee expects to take to repair the valve and return the RHR system to service.

In the phone call on October 16, 1987, the licensee explained that the RHR is expected to be tagged out for the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />; however, the time to drain the RHR line is 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, the work on the valve will be 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br />, and the time to refill the RHR is 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> The time to return the RHR to service by installing the valve cover is 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> to install the cover and 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> to refill the RHR for a total of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. This latter 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> would allow a rise of 12 to 26 F in the RCS, and with the licensee installing the plate at 30 F from the administrative limit of 200 F, the temperature should not go above the 200 F.

The staff has considered the available cooling methods as to capability,  ;

availability and safety-grade classification. The charging pump from the RWST j for the feed and bleed method is an engineered safety feature system. The I staff concludes that these are acceptable methods to cool the core and, given the decay of the fuel in the core, should allow the licensee to keep the RCS temperature below 200 F and the reactor in Mode 5 while the repair work is being done.

3.3 Accident Evaluation The licensee considered the following potential accidents: boron dilution of the reactor coolant, control rod withdrawal, control rod ejection positive  :

I reactivity addition during cooldown and overpressure protection. The. staff i l has also considered loss of coolant accidents. These accidents are considered '

because the RHR would be used in these accidents to cool the core to cold shutdown and an accident might significantly increase the decay heat rate of the core.

l The repair procedures require communications between the control room and the work area. The operators would be aware of the condition of the RHR while work is proceeding. l 1

f l

L_______----  !

As required by the TS if the RHR system is inoperable in Mode 5, the available 1 RCS boron dilution paths will be closed and placed under administrative control. l Therefore, no inadvertent boron dilution of the RCS should happen. ]

The power to the reactor trip breakers would be removed and placed under administrative control. Therefore, inadvertent withdrawal of control rods should not happen.

For control rod ejection, the licensee stated that, as discussed by FSAR Section 15.4.6, ejection of. control rod will occur due to mechanical failure of a control rod mechanism pressure housing. Control rod pressure housing assemblies were conservatively designed, assembled, and ship tested to 4100 psi under a thorough quality control program ensuring a significant margin of strength in the elastic range. Thus, ejection of control rod is highly unlikely to occur at tha reduced pressure (maximum 275 psig) existing in the RCS while the RHR is out of service.

To prevent positive reactivity being introduced in the core during cooldown, the boron concentration in the RCS would be kept greater than 2000 ppm. This would be done for the RWST where water may be pumped into the RCS. ' This, as .

i calculated by the licensee, would keep the K effective less than 0.95, as j required by the TS for all RCS temperatures between '100 F. and 200*F. As the RCS temperature rises, the need for bcron decreases'. l For overpressure protection, the licensee stated that pressure relief devices on the RCS (on the pressurizer) comprise the three pressurizer safety valves i and two power-operated relief valves. These pressure relief devices discharge j to the pressurizer relief tank by common header and are designed and installed  !

to accommodate the RCS temperature and pressure attained under all expected modes of plant operation. These pressure relief devices would remain in service.

Thus, overpressurization of RCS should not occur. l The staff also considered loss of coolant accidents from pipe breaks, pipe cracks and inadvertently draining the RCS through the opening in the RHR system. The licensee stated in the October 16, 1987, phone discussion on its submittal that there will not be other work in the containment concurrent with the valve repair work that involves heavy loads. Small breaks as discussed in the FSAR l can be handled by the charging pumps which are available. The RCS pressure will be no more than 275 psig and because the RCS is designed for operating pressures of greater than 2200 psig, RCS pipe breaks and leaks are not considered credible.

In its repair prowdure, the licensee addresses the potential for draining the RCS through the valve being repaired in the RHR system. There will be two independent checks that the RHR' isolation valves are fully closed and power is removed from the operators. Before the valve bonnet is removed, the RHR piping is drained, tygon tubing is att. ached to the drain upstream of the 2-FCV-74-1 valve and an unacceptable leak would be seen should water rise too fast in the tygon tube to the elevation of the valve.

i

s

,. -l-1 1

i i

1 1

If the RHR 2-FCV-74-1 isolation valve is leaking, this coolant will.be drained l out through the drain pipe and what cannot drain out will begin to fill the RHR l pipe to the valve to be repaired. The licensee can determine the leakage, if  !

any, through the isolation valve by the rate of water rise in the tygon tubing  !

and water flowing through the drain line. The acceptable leakage through l the RHR isolation valve would be such that the water level would not reach the 'l valve to be repaired in the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period. The RHR isolation valve is a j safety-grade valve with a history of low leakage at reactor operating pressure-therefore, the licensee expects no problems with the RCS at 275 psig when the work l will be done. The safety-grade charging pump would be able to keep up with this {

leakage. j i

4.0 CONCLUSION

S j l

The licensee has requested that the RHR syttem be inoperable in TS Mode 5 for j 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to repair a common valve in the two loop RHR system. The entire RHR j system would be inoperable for up to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The staff has considered the need for the repair and the safety significance of the repair in Mode 5. These are discussed in Section 3 above.

The licensee is requesting the approval because the RHR can still cool the core with the valve in its present degraded condition. If the RHR system )

is inoperable, TS 3.4.1.4 requires for Made 5 that the licensee suspend all l operations involving a reduction in boren concentration of the RCS and imme-  ;

diately initiate corrective to return the required RHR locp to operation. )

The actions being taken by the licensee as part of the repair of the RHR valve 1 are in compliance with the Action statement for TS 3.4.1.4 l

The staff concludes that the bent stem in the 2-FCV-74-2 valve and the leakage,  !

even if small, of boric acid which can corrode the valve is an acceptable j justification for declaring the RHR system inoperable in Mode 5 because the I core can be adequately cooled by other means. The staff concludes that the ,

valve should be repaired now and the reactor does not need to be defueled.  !

As explained in Section 3, the staff has concluded, based on the actions and l calculations of the licensee explained in its October 12, 1987 submittal and i clarified in the October 16, 1987 phone call, that the repair should be i made now and can be safely done ir. Mode 5 with the RHR system inoperable.

Therefore, the staff concludes that having the RHR inoperable in Mode 5 for l 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is acceptable for Seouoyah Unit 2. '

Principal contributor: Jack Donohew Date: November 2, 1987 i l

l

)

!