ML20214Q733

Bypass Timer Calculation for Ads/Eccs Mod for Pilgrim Station

ML20214Q733
Person / Time
Site:	Pilgrim
Issue date:	12/16/1986
From:	Fortney R, Hamon D, Rogers A BOSTON EDISON CO.
To:
Shared Package
ML20214Q694	List: BECO-87-086, Application for Amend to License DPR-35,consisting of Proposed Change 87-06,revising Table 3.2.B,Sections 4.5.E.1.a & 3.2 to Reflect Mods Made to Facility to Address NUREG-0737,Item II.K.3.18 ML20214Q715 ML20214Q733
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.K.3.18, TASK-TM DRF-668-0003-5, DRF-668-3-5, EAS-154-1286, NUDOCS 8706050136
Download: ML20214Q733 (18)

Text

.. . . . . - - . ..

.- . EAS 154-1286

. DECEteER 16, 1986 DRF 663-0003-5 BYPASS TIMER CALCULATION FOR THE ADS /ECCS MODIFICATION FOR PILGRIM STATION s

Prepared by: h2[f R.U. Fortney

~

Plant Performance Engineering i _.

^

4

' Reviewed by: br -

D.A. Hamon l Plant Performance Engineering l

Approved by: .Va O MEM 9es I.E. Rogers, Manager Plant Performance Engineering.

B706050136 870520 PDR ADOCK 05000293 P PDR

4 DISCLAIMER OF RESPONSIBILITY l l

. . l l

1 This document was prepared by*er for the General Electric Company. l Neither the General Electric Company nor any of the contributors to this documeat:

' l A. Makes any warranty or representation, express or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this document, or that the use of any 4 information disclosed in this document may not infringe privately

' owned rights; or B. Assumes any responsibility for liability or damage of any kind which may result from the use of any information disclosed in this I

document.

l.,

1

< L i~

i 1

m

(

.y e CONTENTS Page 1.0

SUMMARY

OF REPORT 1

2.0 BACKGROUND

2

3.0 DESCRIPTION

OF THE ANALYSIS 5 4.0 DISCUSSION OF RESULTS g g ,

5.0 CONCLUSION

S 13

6.0 REFERENCES

14 I an

1. .

4

=;

i  :

3 4 .

1 l

., '}

i ii 1

l I

l

TABLES AND FIGURES

- Page Table 1 Initial Conditions and Assumptions 7 Figure 1 Option 4-b ADS LOGIC 4 s -

Figure 2 Vessel Respor.se to RWCU Line Break 10 Figure 3 - Clad Heatup Over Time 11

. Figure 4 Bypass Timer Settings vs. PCT 12 f

s r 9 e

i ..

1 1

.l iii

. i

- l 1.0

SUMMARY

I l

In response to NUREG-0737, the BWR Owner's Group proposed modifica,tions to the ADS logic with the intent of extending ADS operation (with a minimum of operator int'ervention) to transients and events which may not result 'in the high drywell pressure or low pressure ECCS pumps running signals necessary for automatic ADS activation. An example of such an 4

event, is a line break that is relatively small and occurs outside of the containment.

Of the various modifications considered, the one which adds a timer 4 bypass on the high drywell pressure permissive and a manual inhibit of automatic ADS initiation was considered the best for Pilgrim. In addition, a modification to the low pressure ECCS pump logic was also j reconnended. A timer that bypasses the high drywell pressure and low RPV pressure signal, thus making the logic compatible with the improved ADS l logic, was elected as the modification to the low pressure ECCS pump logic. These combined modifications are designated by GE as Option 4-b.

. . _ . Figure 1 is a schematic of this logic.

An analysis was performed to detennine an upper time limit on the bypass i timer. The goal was to prevent peak clad temperature (PCT) from exceed-

! ing 1500*F for a limiting break, which was determined to be a Reactor Water Cleanup line break. A total delay time of 18 minutes between the

. low water level initiation of the timer and the heatup of the cladding to within acceptable limits, was found to meet this criterion. Of this 18 minutes, 2 minutes is the standard 120 second delay time for the ADS timer. Therefore, the bypass timer upper limit is 16 minutes, which provides a conservative margin for PCT and allows sufficient time for

! operator intervention if required, i

l 1 4

1

. _ . ,, _ . . . _ . . _ . , . . _ . - _ _ . _ _ . _ . - - - . . , . _ _ _ _ _ . _ . . . _ _ _ . , _ . _ . _ _ _ _ . . _ , . . . ~ . . _ . _ . . . _ _ . . . . - . ..

2.0 BACKGROUND

In the aftermath of TMI, attention was focused on minimizing operator intervcntion. The BWR Owners' Group (BWROG) funded a study (Reference 1) that considered modifications to the Automatic Depressurization System (ADS) with the objective of extending non-manual ADS operation to a class of transients that would slowly uncover the core without depressurizing

the vessel or pressurizing the drywell. This class of transient or event i is analyzed without feedwater, High Pressure Coolant Injection (HPCI) or Reactor Core Isolation (RCIC), consistent with the Worst Single. Failure and Safety Systems classification criteria. Only ADS is considered available to depressurize the vessel, pemitting the operation of the low pressure ECC systems. The ADS is currently activated automatically upon coincident signals of low water level in the reactor pressure vessel, high drywell pressure, and low pressua Emergency Core Coolant System

(ECCS) pump running. A time delay of approximately 2 minutes after receipt of the signals allows the operator to reset the logic and prevent an automatic blowdown if the water level in the RPV is being restored or

.-J if the signals are erroneous. The ADS /SRVs can be manually initiated as well. For the events of concern, which do not directly produce a high  ;

I drywell pressure signal and are degraded by a loss of all high pressure injection systems, adequate core coolir.g is currently assured by manual l

, depressurization of the reactor followed by injection from the low 4

pressure systems. To reduce the dependence on operator action and to satisfy the intent of the Nuclear Regulatory Commission (NRC) requirements, Item II.K.3.18 of NUREG-0737 (Reference 2), several design 1- options for BWR 3 plants were identified as a result of the BWROG's study. The option reconnended for Pilgrim is identified in this report

! as Option 4. The Option 4 modification adds a timer which bypasses the high drywell pressure pemissive of the current logic and adds a manual l

switch which allows the operator to inhibit automatic ADS initiation.

! Inplementation of this option requires that the pump start logic of the f low pressure Emergency Core Conling System (ECCS) be modified for some

BWRs. This is because the ECCS pump start logic for all BWR 3s and i

i ~

,---m-- ,,,-vw-,-,---,-,,- m x-w--,-,-ww-,m. .e.,,w-e--,e,mv-*-mm, err-m. n,w-owo---,----,-~~- -m,----mmvou- e-aa,-w ,s

Vermont Yankee requires either; (1) high drywell pressure or (2) reactor low-low water level and low reactor pressure. Neither low RPV pressure nor high drywell pressure would be expected on a timely basis for transients events that the ADS modification is intended to cover. Thus, the timely start of the low pressure ECCS pump could not be assured without operator action.

To alleviate this situation, it was proposed that Pilgrim add a bypass timer to the pump logic as well as to the ADS logic. These two modifications combined have been designated Option 4-B by GE. In Option 4-B, the bypass timer for the high drywell signal would also be used to bypass the low RPV pressure signal. Low wate'r level would initiate the bypass timer, which would automatically reset if the water level is restored. When the bypass timer times out, the low pressure ECCS pumps would . start running and ADS would initiate after the 120 second delay if there is no operator intervention.

s This logic arrangement maintains the same nriginal design philosophy of minimizing the potential of unnecessary ECCS pump starts at high reactor

pressure. Also, the additional logic would not affect the present high

- drywell pressure-low water level initiation logic necessary for pipe l breaks inside the drywell, and the addition of a manual inhibit switch has no effect on automatic ADS response to isolation or LOCA events.

' The addition of a bypass timer to the ADS and pump start logic requires specifying a maximum time such that the PCT will not exceed 1500'F for the limiting case and that the presently defined limiting cases in the FSAR will not be affected by this change. Specifying this time and the limiting break, is the purpose of this report.

f

~

High Drywell Pressure Low Water Level Seal In ?I (Low Pressure n ECCS Actuation)

Low Water Level (Low Pressure ECCS Actuation)

Bypass Timer

,, - (Nominal Setting-4 f $6 minutes) l 120-Second -

Actuation Timer

<r 4 ,

Low Pressure ECCS Pumps e Low Pressure Running ECCS Pumps Start 9 _

Manual Inhibit Not Activated y

ADS Actuation OPTION 4B (Add Bypass Timer For the High Drywell Pressure Permissive of ADS and to the Low Pressure ECCS Pump)

FIGURE 1 ADS LOGIC MODIFICATION 4-B

l

~

3.0 DESCRIPTION

OF THE ANALYSIS As stated above, the purpose of this analysis is to investigate a class of transients and events that do not pressurize the containment, but do i eventually require AD3 to depressurize the reactor vessel to within the operating range of the low pressure ECC systems because high pressure systems are considered failed. After identifying the. limiting case, the goal was to find the length of time ADS can be delayed before the PCT exceeds 1500*F. Note that a lower PCT limit than that imposed by 10CFR 50.46 (2200*F) has been specified for this class 'of events. This lower s

limit will assure that the transient or outside line break events will' not be the limiting events for the plant safety analyses. The current limiting events are large recirculation line breaks inside the containment, which are not affected by this modification.

The limiting case for this class of events is a Reactor Water Cleanup line (RWCU) break outside of the containment. This had been determined by sensitivity studies for similar plants in previous studies and con-firmed for this pitnt also. This break can result in a significant amount of inventory 70ss before the break isolates. Also analyzed was an

.' outside steamline break, which is of a similar type but has proved not to

! be as severe as a RWCU line break. The RWCU line break would result in a high temperature downstream of the heat exchanger and a low water level signal, either of which would cause the RWCU isolation valves to close, isolating the vessel. For this analysis, the conservative assumption is that valve closure is initiated on low water signal and complete in 40 seconds.

The scenario for this event is as follows. The RWCU line break causes an initial loss of coolant inventory. When the water level reaches the setpoint for MSIV closure, which is also the bypass timer initiation setpoint, the vessel isolates. This occurs at approximately 35 seconds, as can be seen from the plot of water level and vessel pressure (Figure 2). Isolation causes a collapse in the water level and an

l 1

. \

increase in vessel pressure which is relieved by the safety relief valves '

(SRVs) at approximately 90 seconds. Without feedwater, HPCI and RCIC, the SRV actuations gradually reduce the reactor, water inventory to uncover the top of the active fuel which causes the core to heat up.

Once the bypass timer times out, the low pressure ECCS pumps start and the ADS initiates. When the ADS actuates, the rapid depressurization results in bulk flashing with significant water level swell which subsequently covers the core and cools the fuel. As the depressurization continues, the core once again uncovers and further core heatup occurs The fuel cladding heat up decreases when LPCI and core spray begins to s provide core cooling. Eventually core heatup is teminated when the core refloods. Several such cases were done, varying the bypass timer sitting. Figure 3 shows PCT over time for some selected cases.

This analysis was perfomed consistent with, and in accordance to, the Pilgrim LOCA Safety Analysis Report, NED0-21696, August 1977. Table 1 lists the more pertinent initial conditions and assumptions. The results of this analysis are discussed in Section 4.

1

(

~

Table 1 INITIAL CONDITIONS AND ASSUMPTIONS (1) The reactor is operating at 102% of rated power. Yais maximized the fuel cladding heatup and conforms with the requirements of 10CFR50 l

Appendix K.

(2) The initial reactor water level is at the scram level i.e., low g water level. This minimizes the vessel. inventory, resulting in higher decay heat upon core uncovery.

(3) The initial system pressure and steam flow are consistent with the heat balance values for the assumed initial power and water level.

(4) The RWCU line is instantaneously severed by a complete circumferen-

, tial break. The break is physically arranged so that the coolant

~~

i niischarge through the break is unobstructed. These assumptions

! result in the most severe mass loss for the event.

~

. (5) The RWCU isolation valves are assumed to close on the low water level signal (484.5 inches above vessel zero) rather than high temperature trip signal.

(6) The RWCU isolation valves closure time is 40 seconds which is the maximun value allowed by the plant technical specifications. The

^

flow area is assumed to reduce linearly.

(7) Feedwater flow is assumed lost within one second after the break

occurs. This assumption minimizes the amcunt of inventory available to the reactor.

(8) The high pres' coolant injection (HPCI) and the reactor core isolation cooling (RCIC) systems are assumed not available. A 5

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

complete loss of high pressure makeup systems is a condition which may require rapid depressurization with the ADS.

(9) The low pressure coolant injection (LPCI) and the core spray systems are available for inventory makeup when the reactor is depressurized below the shut-off heads of the low pressure pumps.

(10) The decay heat value is 120% of the 1971 ANS standard which confoms to the requirements of 10CFR50 Appendix K.

4 .

i l

1

, O ,

4.0 RESULTS Bypass timer settings of 16, 17, 18, 19, and 21 minutes for a RWCU line break were evaluated for the effect on PCT. A bypass setting of 16 minutes will not exceed the conservative limit of 1500*F, and will provide the operator time to inhibit any unnecesscry ADS action. The plant response is similar for the different settings except that, as expected, the longer ADS is delayed the higher the PCT. Figure 4 is a plot of the PCT versus Timer setting for the various cases.

The PCT resulting from a comparison case for a steamline break outside of the containment with a 16 minute timer is also presented in Figure 4.

There is considerable margin between the PCTs for the RWCU line break and the steamline break, confirming that the PWCU line is the limiting break for this class of accidents.

s i

l

_g.

W r m -,emae-- m e - - - r-- - -v.--- -- .e-m m mm -s -- gy- - y w~aser- w ---- ve.v.-.r,-M-*-r-emyy nw+-w-~---- - nww-w-w-

VESSEL PRESSUFE IPSIR) I a BAF - 17.59 FIEET '

s TAF - 29.59 F EET e WATER LEVEL. F EET 17 a10 t 1 1 9 I s .

ADS TIMER Si TTINGS

(

'~

%0 + 120 sec M m 0.8 T (16 + 2 mi n) c_

~

, i n _s a A i I .

am Nr t.LJ CC j

n. 0.4 ,,

, , , , , , gy, , ,

dO N/

W LLJ i N.

O 'O .~ I ' ' ' I ' O' ' ' . 37 0.74 1.11 1.46s10' TIME (SECONDS)

FIGURE 2 VESSEL RESPONSE TO RWCU LINE BREAK 16 MINUTE BYPASS TINER

~

r e

iPEAMCLADTEdRATURE

HTC BTU /HR-FT3-DEG F PCT LIMIT - 22C0.0

3. l

,10' 4 , _ _ _ , , m,0 p ,r, L G, m. s tim e.v-

- . . . . . . i .

E t.o

2. 2 coe -w C 4

~

nsoo -s /

/\

s a.

g -

• ' \

t.u -

>-- e n ,

s s

@ g, E COC *F j t

. d M

\/

g e i e i i t.u , , , ,

. 1 _

t t i 1 l t t t I b , W 4 Q 'O . O.37 0.74 1.Ii 1.48s10' TIME (SECONDS) l FIGURE 3 CLAD HEATUP OVER TIME FOR 16 and 19 MINUTE BYPASS TIMERS t

6

_-. .. . .- . = . . . . _ .. .-

s

~.

BYPASS TIMER SETTINGS VS PCT -

l 3 '

2.8 - .

2.6 -

2. 4 -

2 2.2 -

1 O

! ^ n 1

k m M

O T) Q Wc O 1.8 -

O r*

5S 1.6 - 0 1500 "F.

,  ? i- t-ov

a. 1.4 h STML

• pBRK PCT 1.2 -

1_

0.8 -

0.6 -

i O.4 i i i i i 16 18 20 .

22

TIME (IN MINUTES)'

. FIGURE 4 .

BYPASS TIMER SETTINGS VS PCT FOR RWCU LINE BREAKS

5.0 CONCLUSION

~

It is concluded that the addition of a 16 minute' bypass timer to the ADS and low pressure ECCS pump initiation logic will allow for adequate core cooling for a class of accidents that may uncover the core without pressurizing the containment or depressurizing the vessel to the present ADS / low pressure ECCS setpoints. It will also provide adequate time for operator intervention. . It is also concluded that within this class ~ of transients, the RWCU line break is the bounding case.

s .

t i

e

-i 1

1 I

6.0 REFERENCES

1. "BWR Owners' Group Evaluation of NUREG-0737 Item II.K.3.18 Modifica-tion of Automatic Depressurization System logic", General Electric Company, NEDE-30045, February 1983.

2. " Clarification of TMI Action Plan Requirements", Office of Nuclear Reactor Regulation, U.S. Nuclear Regulating commission NUREG-0737, November 1980.

3. " Modification of ECCS Pump Start Logic", L.L. Chi, General Electric

' ~

Company, AE-06-0184, January 1984.

4. " General Electric Company Analytical Model For Loss-Of-Coolant Analysis In Accordance With 10CFR50 Appendix X", General Electric company, NED0-20566 January 1976.

W m _