Text

Safety Analysis Input to Startup Team Safety Assessment Rept
	ML20148K796
Person / Time
Site:	Crystal River
Issue date:	06/12/1997
From:	Carlton J, Seals J; FRAMATOME
To:	;
Shared Package
ML20148K774	List: 3F0697-10, TS Change Request 210 to License DPR-72,supporting Operation W/Hardware Changes Primarily Involving Efw,Hpi,Emergency Feedwater Initiation & Control Sys & Edgs,As Well as Associated Licensing & Design Bases Changes; ML20148K781; ML20148K796;
References
	51-1266138-01, 51-1266138-1, NUDOCS 9706180418
	Download: ML20148K796 (78)
	v • d • e

.

t l

(

~

i Safety Analysis input to Startup Team Safety Assessment Report i

i t

l l

l Prepared For:

i Florida Power Corporation l

FTl Document Number 51-1266138-01 l

(

Prepared By:

Framatome Technologies, Inc.

June 12,1997 i.

9706180418 970614 PDR ADOCK 05000302 p

PDR

20440-7(12/95)

ENGINEERING INFORMATION RECORD fM.TM.W.T Documentidentifier 51 12Asisa.01 Title Safety Analysis input to Startup Team Safety Assessment i

PREPARED BY:

REVIEWED BY:

l Name JC Seals Name JD Carlton h!//-/G Signature 2/,[

Signature Date Date [p 1 Technical Ma gers Statement: Initials

$$ kr 3~,T c.

R1 viewer is Independent.

l R: marks:

l l

This revision incorporates additional comments and completely supercedes the original release.

Iri a review of emergency diesel generator (EDG) capacities and emergency feedwater (EFW) dependencies, l

Crystal River Unit-3 (CR-3) determined that insufficient margins could exist for some Emergency Safeguards equipment in mitigation of small break loss-of-coolant accidents (SBLOCAs) subject to coincident loss of offsite power and specific assumed single failures.

The CR-3 final safety analysis report (FSAR), the improved technical specifications, and various supporting design bases documents were reviewed to identify the limiting transient (s) and single failures that could challenge the EDGs and the ability to preserve / maintain EFW flow to the OTSGs. The limiting transients were narrowed to a single event:, a SBLOCA with a coincident loss of offsite power (LOOP) occurring on reactor trip (Figure 1). The single failures of concern were identified as (1) the loss of battery "A" (LOBA), (2) loss of l

battery *R"(LOBB), and (3) failure of the turbine-driven EFW pump.

A successful mitigation path was identified for each of the three single failures. These " solution sets" are depicted in Eigures 2 through 4.

In each of the s51ution sets, certain challenges were identified. This document presents the Framatome Technologies, Inc. (FTI) safety analyses and evaluations addressing those challenges.

Record Of Revisions Rev.

Description Affected Paaes 00 Original Release All 01 Addition Comments All 4

l Page 1 of 77 l

l

\\

l 51-1266138-01 t

Table of Contents

?

L Section Description Paae No.

l r

1.0 Background

8 L

l 2.0 Solution Set Challenges and Resolution Summary 9

l l

3.0 EFW Requirements for SBLOCA 13 4.0 Mission Time for Emergency Feedwater

-15 5.0 EFP-2 Operation 30 6.0 EFP-2 Failure: Cooldown Restrictions 37 7.0 Isolation of HPl Line Breaks 39 8.0 EFW Inventory Requirements 47 9.0 References 50 1

i l

l t

I I

i 1

I k

Framatome Technologies, Inc.

2

4 51-1266138-01 i

)

List of Tables T able No.

Descrict!en

._ ' Pace No.

~

1. Comparison of 1979 and 100 percent of the 1971 ANS 5.1 Standard 16 4

2. HPl Flow Rates into the RCS and Total Flow for One-Pump j

Operation (after 20 minutes with 10% equivalent head degradation) 18

)

i

3. HPI Flow Rates into The RCS and Total Flow For Different Pinch Break Areas (after 20 minutes with 10% equivalent head degradation) 19

4. HPl Flow Rates into the RCS and Total Flow for Two-Pump Operation (after 20 minutes with 10% equivalent head degradation) 20

5. Error-Adjusted RB Spray Actuation Time for 0.01 and 0.04 ft2 CLPD Breaks 24 C. BWST Drain Times 26

7. HPl/ Core Decay Heat Matchup Time (EFW Mission Time) 29

.l i

1 Framatome Technologies, Inc.

3

51-1266138-01 List of Figures i.

L Fiaure No.

Description Paae No.

l l

1.

Solution Set Flow Path 53 l

2.

LOBA Single Failure Solution Set 54 l

3.

LOBB Single Failure Solution Set 55 4.

EFP-2 Single Failure Solution Set 56 2

5.

CR-3 0.0025 ft Break - Pressure Versus Time 57 8

6.

CR-3 0.003 ft Break - Pressure Versus Time 58 1

7. CR-3 0.0035 ft: Break - Pressure Versus Time 59 2

8.

RCS and Secondary Pressures,0.04 ft Break at RC Pump Discharge 60 9.

CR-3 Maximum RB Pressure for 0.01 ft: CLPD Break - One Cooler 61 2

10. CR-3 Mex! mum RB Pressure for 0.01 ft CLPD Break - Two Coolers 62 2

11. CR-3 Maximum RB Pressure for 0.04 ft CLPD Break - One Cooler 63

12. CR-3 Maximum RB Pressure for 0.01 ft: CLPD Break - Two Coolers 64 l

13. EFW Cavitation Ffow Versus Inlet Pressure 65

14. Florida Power Corp. Feed Pump Characteristic Curve for 1080 RPM 66

15. EFW Flow to Match Core Decay Heat 67 2

16. CR-3 0.002 ft Break - Pressure Versus Time 68 2

17. CR-3 0.002 ft Break - Mass Flow Versus Time 69

18. CR-3 0.002 ft' Break - Temperature (CVs 175/275) Versus Time 70 2

19. CR-3 0.002 ft Break - Temperature (CVs 195/295) Versus Time 71 i

20. CR-3 0.0014 ft Break - Pressure Versus Time 72 2

f Framatome Technologies, htc.

4

51-1266138-01 List of Figures (Continued) 1 i

Fiaure No.

Descriotion Paoe No.

21. CR-3 0.0035 ft: Break - Pressure'Versus Time -

73 2

22. CR-3 0.0025 ft Break (2-HPl Pumps)- Pressure Versus Time 74 2

23. CR-3 0.0025 ft Break (2-HPl Pumps) - Temperature Versus Time 75 l

24. Total EFW Flow Required Post-SBLOCA in the CLPD Region 76 i

I

25. Total EFW Flow Required for Core Decay Heat Removal for i

the "Zero" Break SBLOCA 77 F

l l

t I

I 1

1 I

i Framatome Technologies, Inc.

5

51-1266138-01 List of Acronyms l

ADVs Atmospheric Dump Valves ANS American Nuclear Society BWST Borated Water Storage Tank i

CC Control Complex l

CLPD Cold Leg Pump Discharge CR-3 Crystal River Unit 3 l

ECCS Emergency Core Coolant System l

EDG Emergency Diesel Generator l

EFIC Emergency Feedwater Integrated Control EFP-1 Emergency Feedwater Pump-1 (Motor-driven)

EFP-2 Emergency Feedwater Pump-2 (Turbine-driven)

EFW Emergency Feedwater EOPs Emergency Operating Procedures l

FPC Florida Power Corporation FPS Full Power Seconds ESAS Engineered Safeguards Actuation System FSAR Final Safety Analysis Report i

FTl Framatome Technologies, Inc.

HPl High Pressure injection LOBA Loss of Bauey 'A' l

LOBB Loss of Battery 'B' LOCA Loss-of-Coolant Accident LOOP Loss of Offsite Power LPI Low Pressure injection LSCM Loss of Subcooling Margin MSSV Main Steam Safety Valve

(

MUP Makeup Pump j

MUV Makeup Valve MWt Mega-Watts Thermal OTSG Once-Through Steam Generator PSV Pressurizer Safety Valve Q.

Core Decay Heat Q,

Energy Absorption of HPl Fluid RB Reactor Building Framatome Technologies, Inc.

6

_- _ _. ~____.._._.

51-1266138-01 List of Acronyms (Continued)

RBES Reactor Building Emergency Sump RC Reactor Coolant RCP Reactor Coolant Pump RCS Reactor Coolant System RG Regulation Guide SBLOCA Small Break LOCA SCM Subcooling Margin W,

HPl Mass Flow Rate I

t i

i t

L i

i 4

i I

Framatome Technologies, hic.

7

51-1266138-01

1.0 BACKGROUND

i l

In a review of emergency diesel generator (EDG) capacities and emergency feedwater (EFW) dependencies, Crystal River Unit-3 (CR-3) determined that insufficient margins i

could exist for some Engineered Safeguards equipment in mitigation of small break l

loss-of-coolant accidents (SBLOCAs) subject to coincident loss of offsite power and l

specific assumed single failures.

r Certain SBLOCAs require cooling via EFW to the once-through steam generators l

(OTSGs) for some period of time for successful mitigation. At CR-3, EFW is provided by a motor-driven pump (EFP-1), powered by the 'A'-Train emergency diesel generator, I

and by a turbine-driven pump (EFP-2), powered by steam from the OTSGs. The 'A'-

Train emergency diesel generator (EDG-1A) cannot support concurrent operation of the i

EFP-1 with either the 'A'-Train control complex (CC) cooling system operating, or with the 'A'-Train low pressure injection (LPI) pump operating coincident with reacter building (RB) spray pump operation.

This represents an operational limitation associated with EDG-1A that, in combination with specific single failures, could pose challenges to the mitigation of accident events requiring EFW.

The CR-3 final safety analysis report (FSAR), the improved technical specifications, and various supporting design bases documents were reviewed to identify the limiting transient (s) and single failures that could challenge the EDGs and the ability to preserve or maintain EFW flow to the OTSGs. The limiting transients were narrowed to a SBLOCA with a ccincident loss of offsite power (LOOP) occurring on reactor trip (Figure 1). The single failures of concem were identified as (1) the loss of battery 'A' (LOBA), (2) loss of battery "B"(LOBB), and (3) failure of the turbine 4 riven EFW pump.

(Note that a battery failure will disable the associated EDG and the equipment powered by that EDG.) Specific descriptions of these postulated single failures and accident mitigation challenges are contained in Reference 1.

All event sequences begin with the plant operating at power in Mode 1 when a y

SBLOCA occurs in the reactor coolant cold leg pump discharge (CLPD) piping region, or a high pressere injection (HPI) line breaks between the last check valve and the CLPD piping, or e core flood line breaks between the last check valve and the reactor vessel. The CLPG break region is the most limiting relative to available HPl flow for core cooling because a portion of the emergency core cooling system (ECCS) injection fluid is lost directly through the break. Due to the loss of subcooled liquid, the reactor coolant system (RCS) begins to depressurize. The reactor trips on low RCS pressure j

with a coincident LOOP. Subcooling margin will be lost and, if it is not automatically l

actuated on low RCS pressure, the operators will manually initiate high pressure injection (HPI) flow. The emergency feedwater initiation and control (EFlC) system actuates, and EFW will begin to fill the OTSGs to the required setpoint. These events are consistent with assumptions modeled in the SBLOCA design basis analyses.

l Framatome Technologies, Inc.

1

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

51-1266138-01 A success path was identified for each of the three single failures. These " solution sets" are depicted in Figures 2 through 4 and are described in Reference 1. In each of the solution sets, certain challenges were identified. This document presents the Framatome Technologies, Inc. (FTI) safety analyses and evaluations addressing some of those challenges.

i 2.0 Solution Set Challenges and Resolution Summary The solution sets for successful mitigation of SBLOCA rely upon the operability and capacities of the emergency feedwater and safety injection systems subject to specific single failures. In developing the solution sets, six challenges to successful SBLOCA mitigation were identified. Each of these challenges are discussed below, along with a summary'of the resolution to each challenge. Additional discussion of how each challenge was resolved is provided in the subsequent sections (Sections 3 through 8).

1. With two HPl pumps, is EFW required for SBLOCA mitigation? (Section 3.0)

This challenge is related to the single failure of EFP-2, when both HPl pumps would be available. If the flow from two HPl pumps is not sufficient to provide core cooling for all break sizes, then EFWwill be necessary for some period of time.

As discussed in Section 3.0, for certain small breaks, steam generator cooling would be required for a period of time-depending upon the break type and size-to assure adequate HPI flow and core cooling. These small breaks were identified as those smaller than approximately 0.04 ft with one HPI pump available and smaller than 2

8 roughly 0.01 ft with two HPl pumps operating. If EFW is lost for these break sizes, the RCS could repressurize to the pressurizer safety valve setpoint before HPl alone is capable of removing core decay heat, i.e., break /HPI cooling.

2. Under realistic and Appendix K decay heat assumptions, with one and two operating HPl pumps, how long is emergency feedwater required? What is the EFW mission time? (Section 4.0)

In Section 4.0, EFW mission times are calculated to determine the minimum times-for each of the three s%nificant single failures-that feedwater would be required before break /HPl cooling could match core decay heat. These mission times define how long EFW must be available to ensure adequate core cooling for each of the three single failures. Beyond these EFW mission times, break /HPI cooling alone will

. provide adequate core cooling.

With one HPI pump (for either a LOBA or LOBB single failure), the BWST will be drained before the HPl flow into the RCS is sufficient to remove the core decay l

I heat. Therefore, HPl flow will be switched to recirculation from the RB sump, via the

{

LPl pumps (piggy-back operation) with EFW still required. Accounting for the I

l Framatome Technologies, Inc.

51-1266138-01 l

higher enthalpy of the injection fluid during the recirculation mode, the time for the HPl flow to match the core decay heat is 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months , using conservative decay heat.

l With realistic core decay heat, this time is significantly reduced to 14.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months .

l However, the EFW mission time with realistic decay heat is established by the i

BWST drain time of 18.54 hours6.25e-4 days 0.015 hours 8.928571e-5 weeks 2.0547e-5 months .

j With two HPl pumps operating (for a single failure of EFP-2), the core decay heat can be matched at 1.6 hours6.944444e-5 days 0.00167 hours 9.920635e-6 weeks 2.283e-6 months post-LOCA when in the recirculation mode. For realistic decay heat levels, this time is reduced to less than one hour. For the EFP-2 i

single failure, the HPI flow--based on taking suction from the BWST-will not match i

the core decay heat before the BWST is drained. With the CR-3 HPI flows, the BWST will not reach the sump switchover level until after 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months . Therefore, EFP-1 can be secured at that time because HPl flow alone will provide adequate

)

core cooling at the time of'A'-Train LPI pump startup for recirculation mode, piggy-l back operation. That is, the EFW mission time is 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months for those break sizes that require EFW.

L 1

3. -Will EFP-2 be operable through the required mission time? (Section 5.0) 1 For the LOBA single failure, EFP-1 will be unavailable, and EFP-2 must provide the l

required EFW to ensure adequate core cooling. For the LOBB single failure, EFP-2 must be crosstied to the 'A'-powered EFW control valves to provide the necessary i

EFW before EFP-1 is secured. This is because EFP-1 must be secured to allow l

loading of the CC cooling system, or the 'A' Train LPI pump on the 'A' Train EDG for recirculation, or if the EFW/LPl interlock trips EFP-1 at 500 psig in the RCS during an operator-initiated cooldown. In any case, EFP-2 must be able to provide EFW at the reduced OTSG pressures that may occur over the EFW mission time.

FPC and FTl were advised by the Terry Turbine division of Dresser-Rand that there have been no changes to the operation of the Terry Turbine since the original purchase specification. The purchase specification for the EFP-2 pump and driver I

called for a capability to provide EFW to the steam generators for heat removal and i

cooldown over the NSS steam pressure operating range from the steam generator design pressure of 1050 psig down to about 20 psig. This steam pressure is low enough to cool the RCS to the decay heat removal system operating temperature of 280'F.

Recent testing performed by ingersoll-Dresser (Reference 18) on a pump similar to the turbine-driven pump at CR-3 confirms that operation of EFP-2 down to a turbine inlet pressure (OTSG pressure assumed to be the same as the turbine inlet pressure) of 20 psig will support a turbine speed of 1080 rpm.

Under these conditions, EFP-2 can provide flow rates between 50 and 320 gpm, including the pump recirculation flow.

Based on this information. EFP-2 can perform the j

intended safety function at OTSG pressures as low as 20 psig.

i i

Framatome Technologies, Inc.

10 l-

51-1266138-01 Te demonstrate RCS and OTSG response to the break areas requiring the longest EFW mission times, a set of representative RELAP5 calculations were performed.

In the cases for the smallest break areas, single-phase natural circulation was maintained and the HPl flow was throttled to maintain subcooling margin. Both continuous and cyclic (closing steam admission valves at an OTSG pressure of 185 psig and opening the valves when either the pressure increased to 1000 psig or the SG level decreased below the RCP spillover elevation) operation of EFP-2 were modeled, demonstrating continued RCS cooling throughout the EFW mission time.

For slightly larger breaks, decoupling of the OTSGs and RCS may occur for relatively brief intervals, but break /HPI cooling was sufficient to continue depressurization of the RCS to the matchup time.

4. What are the cooldown restrictions for the EFP-2 single failure? (Section 6.0)

With a single failure of EFP-2, only EFP-1 will be available to provide the EFW required to ensure adequate core cooling. If the RCS pressure decreases to 500 psig during post-LOCA RCS cooldown and depressurization, the EFW/LPI interlock could trip EFP-1 at a time when EFW flow is still required (i.e., flow from the two available HPl pumps is not sufficient to provide core cooling). Also, with EFP-2 unavailable, if the EFW/l.Pl interlock at 500 psig can be reached when EFW is still required to provide adeouate core cooling, restrictions must be placed on the RCS cooldown to ensure that the RCS pressure is not reduced below 500 psig until HPl flow alone is sufficient to provide core cooling.

For an operator initiated cooldown, guidance for the operators should be provided, when both HPI pumps are available, to start the RCS cooldown at a reasonable time, maintain the cooldown rate within technical specification limits, and limit the OTSG pressure to 500 psig. This guarantees that the RCS pressure will remain above the 500 psig RCS pressure EFW/LPI interlock setpoint until an assured long-term source of EFW is available for the break sizes that require EFW for mitigation.

For break sizes that are large enough to depressurize the RCS to the EPN/LPI interlock setpoint without operator action, EFW wili not be required for mitigation because the break area is large enough to pass all of the core steam production, and break /HPl cooling will be sufficient to cool the core.

5. What are the broken HPI line isolation criteria when both HPl pumps are operating and EFW it lost? (Section 7.0)

As indicated in item 1, even with both HPl pumps available, EFW will be required for a period of time for breaks of approximately 0.01 ft and less. The current HPl line 2

isolation criterion are evaluated at only cne time during the event. If all EFW were subsequently lost (e.g., EFP-1 is secured to allow operation of the 'A'-Train LPI pump for recirculation from the RBES, along with a single failure of EFP-2), the available HPl flow may not be adequate to ensure core cooling, depending on the decay heat level. If all EFW is lost and the break is in an HPI injection line, flow to 11

\\

Framatome Technologies, Inc.

l l

l 51-1266138-01 j

i I

l the broken HPI line must be isolated to ensure adequate core cooling. Therefore, revised HPl line isolation criterion were developed to ensure that the broken HPI line l

l is isolated when necessary to ensure adequate core cooling.

i A spectrum of breaks, single failures, and plant configurations were evaluated to develop a revised isolation criterion. Based on the evaluation, the revised isolation l

criterion provides sufficient HPI flow to the core for these breaks and single failures to demonstrate that core cooling will be assured. The new criterion also provides j

additional benefits over the existing criterion. The new criterion is as follows:

i fI any time in an unthrottled condition, 2 the highest-reading HPI line indicates flow > 50 gpm higher than the next highest-reading HPI line, RHgN isolate the high l

tiow HPIline.

This new criterion should be applied when all four injection lines are fully-open, with normal makeup isolated. If normal makeup cannot be isolated, or if an HPl injection 5

valve in the broken line is failed open, operator action to isolate the high flow line I

still needs to be taken (if possible), and additional operator actions to preserve EFW or to split the MUP discharge header may be required in any case, the isolation criterion will identify the line that needs to be isolated. If subcooling margin has been maintained or restored before the criterion is met, and the HPI flow has been throttled, no isolation action should be taken for SBLOCA mitigation. If subcooling margin is subsequently lost, the operators should maximize the HPl flow before l

taking any action to isolate a broken line.

l

6. What inventory of emergency feedwater is required until break /HPl cooling is adequate to cool the core? (Section 8.0)

As discussed in item 1, EFW is required for a period of time until break /HPl cooling is sufficient to remove the core energy. The amount of emergency feedwater inventory available must be sufficient to support EFW operation over the required time period to ensure adequate core cooling. An assessment was made of the inventory of emergency'feedwater required until break /HPl cooling is adequate to cool the core. The required EFW inventory was calculated by integrating the core decay heat over time with an end time based on the calculated mission times for EFW.

i l

Since not all break sizes require emergency feedwater for the same period of time, a series of calculations for different break areas were performed and are presented.

These calculations provide an approximate calculation of the required EFW flow for various break sizes. Break sizes above approximately 0.04 ft effectively do not 2

need EFW to mitigate the transient and the inventory required is that necessary to I

fill both OTSGs (32,093 gallons). For the smallest break area, about 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months are required to reach HP!/ break /PSV cooling. Approximately 339,000 gallons of EFW Framatome Technologies, Inc.

'2

51-1266138-01 inventory are used over the 35-hour period. This is based on the mission time calculated when one HPl pump is available, in the recirculation mode, with the limiting pinch break HPl flow rates.

The 339,000 gallon inventory required over the 35-hour EFW mission time is available on site. The total capacity of the EFW tank is 184,000 gallons, with a 150,000 gallon minimum capacity, as required by the Technical Specifications. In addition, water in the condensate storage tank (200,000 gallon capacity) and in the condenser hotwells (200,000 gallon combined capacity) is also available.

Additional discussions of how each of the above challenges were resolved is provided in the following sections (Sections 3 through 8).

3.0 EFW Requirements for SBLOCA Small break LOCA safety analyses assume that uninterrupted EFW flow is provided to the steam generators until OTSG level reaches the required setpoint (error-adjusted 95 percent on the operating range). Thereafter, EFW flow was modeled in the analyses (as needed) to maintain OTSG level. This configurat5n was relied on until long term cooling (decay heat removal system operation) conditions were established. Mitigation of the transient with acceptable consequences was demonstrated with only one train of ECCS equipment (one HPI pump and one LPI pump) and one EFW pump available.

The SBLOCA spectrum includes breaks where the break flow rate exceeds the capacity of the makeup system at normal RCS pressures, approximately 0.001 ft based on a 2

net makeup flow of 160 gpm into the RCS at 2200 psig, up to and including an area of 0.5 ft'. The larger break sizes, above approximately 0.04 ft', based on conservative core decay heat values, do not require 8EFW flow to mitigate the transient. In these cases, all of the steam generated in the core passes out the break removing decay heat (Reference 2). The break is large enough that the RCS pressure decreases below the OTSG pressure due to the break flow, and additional OTSG cooling is not required.

The RCS pressure will not increase for these breaks, even if EFW is lost at some point during the transient, and ECCS injection flow will still be adequate to remove the core energy.

For smaller break areas, however, EFW flow is required for some period of time. The llPl/ core decay heat matchup time, or the EFW mission time, is when the energy absorption by the HPl fluid is sufficient to match core decay heat when boiled at the maximum expected RCS pressure. This time is based on the number of operating HPl pumps, break location, and the RCS pressure response. Using conservative decay heat with one operating HPl pump, break areas of roughly 0.04 ft and smaller require 2

EFW for mitigation. With two HPl pumps, break sizes less than approximately 0.01 ft' require EFW. If EFW is lost before the HPl injection flow can absorb core decay heat and the break size is small enough, the RCS can repressurize, possibly as high as the l

Framatome Technologies, Inc.

13

51-1266138-01 pressurizer safety valve (PSV) lift setpoint. As the RCS pressure increases, the HPl flow rate will decrease. At these high pressures, depending on the break location, there may be insufficient ECCS injection flow to remove core decay heat.

For both the LOBA and LOBB single failures, the feedwater flow recWd for successful-mitigation must be provided by EFP-2, the turbine-driven emergency feedwater pump.

For the LOBA scenario, EFP-1 is unavailable and EFP-2 alone must provHe the required - EFW.

For the LOBB scenario, both EFP-1 and EFP-2 will i..itially be -

operating. However, EFP-2 must be crosstied to the 'A'-powered EFW control valv'es tu provide the EFW before EFP-1 is secured. EFP-1 can not operate concurrent with either the CC cooling system operating, or the LPI pump when the RB sprays are operating, since this would overload EDG-1A. EFP-1 could also be tripped due to actuation of the 500 psig EFW/LPI interlock during an operator-initiated cooldown.

- With a single failure of EFP-2 (the turbine-driven EFW pump) and loss of offsite power coincident with reactor trip, EFP-1 would be available to supply feedwater flow to the steam generators, and would be automatically loaded on EDG-1 A. EFP-1 may need to be tripped manually in order to start the 'A'-Train LPI pump and align the 'A'-HPI pump for recirculation mode (piggy-back operation). In addition, EFP-1 could also'be tripped due to actuation of the 500 psig EFW/LPI interlock during an operator initiated cooldown. Therefore, EFW flow to the steam generators could be interrupted. The need to initiate CC cooling will not effect EFP-1 operation with a single failure of EFP-2 because CC chillers can be loaded on the 'B'-Train EDG, as part of EDG-1A load management.

Depending on the break size and location, the RCS may repressurize if all EFW flow is

{

terminated. If the RCS repressurizes, timely identification and isolation of a broken HPl j

line is necessary to assure adequate core cooling for certain size HFl line pinch breaks that repressurize (i.e., greater than approximately 1600 psig which is the corresponding saturation pressure for the initial hot leg temperature). This requires a revision to the existing HPI line isolation criterion currently implemented at CR-3, Reference 3.

A discussion of the bases for a new criterion is provided in Section 7.0 of this report.

Based on the number of operating HPl pumps and their respective RCS flow delivery profiles, a time can be calculated for each SBLOCA sequence at which HPl alone can absorb all of the core decay heat. After this time, cooling from the OTSGs-EFW flow--is not necessary, but availability of EFW is desirable to continue to cool down the RCS. This HPl core decay heat matchup time is a function of the core decay 1: eat (10CFR50 Appendix K versus realistic) and the expected RCS pressure response to the postulated S8LOCA transient.

i l

i k

l Framatome Technologies, Inc.

14

51-1266138-01 4.0 Mission Time for Emergency Feedwater The EFW mission time is a function of the available HPI flow that reaches the core, which in turn is based on the source for the HPl flow, i.e., the '8WST or the RBES in recirculation mode, the number of operating HPl pumps, and location of the break.

Once the HPI flow rates are known, the energy absorption of the injection fluid can be calculated. Based on the initial core power level, the core decay heat as a function of time after reactor trip is known and 'can be equated to the HPl absorption energy. This defines the EFW mission time, the time at which HPl flow matches decay heat, without the need for OTSG cooling.

Before the HPl/ core decay heat matchup time can be determined, several parameters need to be calculated:

Section 4.1 contains a list of assumptions for boundary conditions. Section 4.2 provides the CR-3 specific HPI flow rates for different breaks.

In Section 4.3 the BWST drain time is calculated. Finally, the EFW mission time is calculated in Section 4.4, and a summary is provided in Section 4.5.

4.1 Assumptions in order to determine a conservative time at which HPl alone can match core decay heat, a number of assumptions are made:

1. Temperature of fluid in the BWST is 120*F, Reference 4 (conservatively greater than the 100*F maximum design basis temperature).

2. Temperature of the fluid at the exit of the decay heat cooler (s) is 140*F, Peference 5.

3. EFtN flow will be lost at some time and the break is small enough to cause the RCS to instantaneously repressurize to the PSV lift setpoint.

4. A +3 percent lift tolerance on the nominal PSV lift setpoint of 2500 psig will be included, i.e. 2575 psig.

I

5. HPl pump stop-check valves in their current position 1

6. A 10 percent equivalent head degradation on the head-flow curves for the HPI pumps.

Calculations are performed using both realistic (90 percent of the 1971 ANS 5.1 decay heat standard for infinite irradiation) and Appendix K (120 percent of the ANS standard) decay heats. Since it is expected that the HPl core decay heat matchup time will be after the BWST is drained, calculations for both one-and two-operating HPl pumps Framatome Technologies, Inc.

15 i

4 51-1266138-01 taking suction from the RBES will be performed. The HPl injection temperature in the recirculation mode through the RBES corresponds to the exit fluid temperature of the i

- decay heat coolers.

- Decay Heat Calculation f

i L

Appendix.K to 10CFR50 specifies the sources of heat that must be considered i

i following a LOCA. The regulation states that, "...it must be assumed that the reactor i

has been operating continuously at a power level of 1.02 times the licensed power level (to allow for such uncertainties as instrumentation error)." In addition, the radioactive decay of actinides and fission product must be considered. An extra 20 percent-for l

conservatism-is added to the fission product decay contribution. A curve fit to this data is contained in Reference 6.

For non-Appendix K analyses, or best-estimate calculations, that support operator actions, a conservative, but more realistic core decay heat model is used (1979 ANS 5.1 decay heat standard, Reference 26). A comparison of the 1979 to 100 percent of the 1971 data is presented in Table 1.

A ratio of the decay heat fractions is also t

included. From this data, it is reasonable to conclude that using 90 percent of the 1971 data is appropriate to use for non-Appendix K applications.

1 Table 1. Comparison of 1979 and 100 percent of the 1971 ANS 5.1 Standard Time Decay Heat Fraction Ratio (sec) 1979 1.0 *1971 1979/1971 1

0.062505 0.065822 0.9496 10 0.048125 0.053011 0.9078 50 0.036168 0.041356 0.8745 100 -

0.031623 0.036734 0.8609 500 0.022319 0.024864 0.8976 1000 0.019001 0.021664 0.8771 1500 0.016951 0.019726 0.8593 2000 0.015484 0.018197 0.8509 2500 0.014361 0.016965 0.8465 3000 0.013469 0.015965 0.8437 4000 0.012155 0.014480 0.8395 5000 0.011259 0.013464 0.8362 7500 0.009988 0.011990 0 8331 10000 0.009338 0.011177 0.8355 20000 0.008052 0.009370 0.8593 30000 0.007258 0.008264.

0.8783 40000 0.006709 0.007526 0.8914 50000 0.006328 0.007016 0.9019 Framatome Technologies, Inc.

'6

51-1266138-01 4.2 HPl Flow Rates The HPl flow versus RCS pressure is provided by Florida Power Corporation (FPC) and is taken from a CR-3-specific hydraulics model of the makeup (MU) system (Reference

5) using the PIPF-PC code. A BWST drain time calculation using this data is performed for the LOBA and LOBB single failures in Section 4.3. The cases are similar in that only one HPl pump will be available. Another calculation is performed for the EFP-2 single failure case with two HPl pumps available. In each instance, the calculations are l

for CLPD breaks and for HPl line pinch breaks. The flow rates for a double-ended

]

break of an HPl line are included, but the operator can recognize these breaks and will i

take action to isolate the broken line. The double-ended break of an HPl line is bounded by the CLPD breaks because, when the broken line is isolated, all of the HPI pump flow will be available to cool the core (all ECCS flow is to the intact legs).

The HPl flow rates at different RCS pressures for use in the LOBA and LOBB evaluations are taken from Reference 5 and are listed in Table 2. Table 3 contains a comparison of the HPl flow rates for recognizable pinch breaks at 1100,1800,and 2575 psig. HPl flow rates for the EFP-2 single failure case for two HPl pumps flowing are included in Table 4.

k i

Framatome Technologies, htc.

17 j

l

51-1266138-01 Table 2. HPl Flow Rates into the RCS and Total Flow for One-Pump Operation (after 20 minutes with 10% equivalent head degradation)

~

RCS CLPD Break Double Ended HPl Line Break HPI Line Pinch Break' Pressure From BWST From RBES From BWST From RBES From BWST From RBES (psig)

(RCS/ total gpm)2 (RCS/ total gpm)

(RCS/ total gpm)

(Ref. 5, Tables 1 & 8)

(Ref. 5, Table 7)

(Ref. 5, Table 13)

(Ref. 5. Table 13)

(Ref. 5, Table 9)

(Ref. 5, Table 6) 0 399.3/540.5 200 388.0/534.0 400 375.9 /517.4 600 363.3/500.0 800 349.5/481.0 1000 334.9/460.9 1100 327.0 /450.05 338.4 /465.6 0.0 /459.4 12.3/473.8 370.4 /447.7 380.3/462.4 1200 319.1 /439.1 330.7 /455.1 356.7/436.9 367.6/452.8 1400 301.2 /414.5 313.7/431.7 328.1 /413.9 340.1/431.2 1600 283.0/389.4 296.3/407.7 296.8/387.8 310.7/407.3 1800 262.0 /360.6 277.0/381.

263.7/359.7 278.3 /380.2 2000 233.2/327.9 255.2 /351.2 228.0/328.7 244.7 /351.6 2200 211.1 /290.5 230.4/317.1 186.9 /292.1 206.9 /318.6 2400 178.9/246.2 201.7/277.5 139.0/248.6 163.5/279.8 2575 144.3 /198.6 171.8/236.4 88.4 /201.7 119.4 /239.0

1. The pinch break is assumed to be in the B1 cold leg. The isolation criterion is based on a 50 gpm difference between the highest and next highest flowing HPI lines at an RCS pressere of 2575 psig. Flow measurement uncertainties are taken from References 7 and 8.

2. The flow into the RCS is equal to the sum of the three lowest-flowing lines. The total flow is the summation of all four injection lines.

3. Data point was interpolated.

'8 Framatome Technologies, Inc.

51-1266138-01 Table 3. HPl Fiow Rates into The RCS and Total Flow For Different Pinch Break Areas' (after 20 minutes with 10% equivalent head degradation)

RCS Pinch Recognizable at 1100 psig Pinch Recognizable at 1800 psig Pinch Recognizable at 2575 psig Pressure From BWST From RBES From BWST From RBES From BWST From RBES (psig)

(RCS/totalgpm)

(RCS/ total gpm)

(RCShotal gpm)

(RCS/ total gpm)

(Ref. 5. Table 9)

(Ref. 5, Table 6)

(Ref. 5. Table 9)

(Ref. 5, Table 6)

(Ref. 5, Table 9)

(Ref. 5. Table 6) 0 200 400 600 800 1000 1100 278.8/452.3 289.7/467.2 370.4 /447.7 380.3 /462.4 1200 261.7/442.1 273.0/457.5 356.7/436.9 367.6 /452.8 1400 225.9/419.3 238.6/436.4 328.1 /413.9 340.1 /431.2 1600 188.4 /354.1 202.4 /412.7 296.8/387.8 310.7/407.3 1800 148.6 /366.0 164.3 /386.5 212.4 /362.4 227.6 /383.0 263.7/359.7-278.3/380.2 2000 105.4 /334.0 123.0/356.6 174.0 /331.5 191.0/354.2 228.0/328.7 244.7 /351.6 2200 57.3 /296.8 78.8/323.4 130.1 /294.9 150.6/321.3 186.9/292.1 206.9/318.6 2400 2.7/252.6 28.9 /284.1 79.4 /251.2 104.5/282.4 139.0/248.6 163.5/279.8 2575 0.0 /250.3 0.0 /264.2 25.6/203.4 57.5/241.5 88.4 /201.7 119.4 /239.6

1. The pinch break is assumed to be in the B1 cold leg. The isolation criterion is based on a 50 gpm difference between the highest and next highest flowing HPl lines at an RCS pressure of 2575 psig. Aow measurement uncertainties are taken from References 7 and 8.

Framatome Technologies, Inc.

'9

51-1266138-01

-Table 4. HPI Flow Rates into the RCS and Total Flow for Two-Pump Operation (after 20 minutes with 10% equivalent head degradaton)

RCS CLPD Break Double Ended HPI Line Break HPl Line Piiich Break' Pressure From BWST From RBES From BWST From RBES From BWST From RBES (psig)

(RCS/ total gpm)2 (RCS/ total gpm)

(RCS/ total gpm)

(RCS/totsl gpm)

(Ref. 5, Table 12)

(Ref. 5, Table 10)

(Ref. 5, Table 13)

(Ref. 5, Table 11)

(Ref. 5, Table 10) 0 759.7 /1048.0 200 737.4 /1017.2 400 713.4 / 984.1 600 687.4 /948.2 800 661.1 /912.0 1000 632.9/873.1 1100 617.8/852.3 639.7/882.4 404.8/892.2 429.0/ 920.1 710.8 /829.1 727.5/860.3 1200 602.2 /830.7 624.4 /861.2 370.2 /873.4 396.4 / 903.2 687.9/809.2 705.1 /841.3 1400 568.4 /784.1 593.1 /818.1 298.6 /832.8 327.7 /865.1 639.7 /766.9 659.8 /802.5 1600 531.6/733.4 559.6 /771.8 221.7 /786.1 255.0 /822.3 587.0/719.7 610.1 /759.0 1800 493.0/680.1 522.2 /720.3 138.7 /733.2 177.2 /774.0 531.8/669.9 556.5 /711.4 2000 448.5/618.7

-480.8/663.2 0.0/625.0 91.7 /718.1 469.7/613.0 498.0/658.7 2200 397.7 / 548.6 434.2/ 598.9 0.0 /655.5 399.7/ 548.0 432.9/599.2 2400 337.6/465.8 380.5 /524.9 317.9 /471.2 358.8 /530.6 2575 273.1/376.8 324.9/448.2 230.3 /387.8 282.5/459.0

1. The pinch break is assumed to be in the B1 cold leg. The isolation criterion is based on a 50 gpm difference between the highest and next highest flowing HPI lines at an RCS pressure of 2575 psig. Flow measurement uncertainties are taken from References 7 and 8.

2.

The flow into the RCS is equal to the sum of the three lowest-flowing lines. The total flow is the summaton of all four injecton lines Fromatome Technologies, Inc.

20

l 51-1266138-01

?

i The flow rates for the pinch breaks that are presented in Table 3 are based on the break being in.the B1 HPl line with the makeup pump 1A (MUP-1A) i operating. The B1 break is also taken in Tables 2 And 4. MUP1A and MUP1C are assumed operating in Table 4. Calculations performed in Reference 5 have l

demonstrated that breaks in this line and this configuration are slightly more limiting than breaks in the other HPI lines.

l Relative to EFW mission time, the limiting pinch break is one that just meets the isolation criterion (see Section 7.0, which discusses the isolation criterion) at the highest expected.RCS pressure, i.e. 2575 psig, or the longest time before the operators will take action to isolate the line. _ This results in the least HPl flow to the core and the minimum HPI energy absorption capability. This maximizes the EFW mission time. For pinch' breaks that can be recognized at lower RCS pressures, operator action will be taken to isolate the pinched line, which will i

result in a larger amount of HPI flow available for core cooling, once the pinched line is isolated, due to the lower RCS pressure. The higher HPI flow' rates result in a shorter EFW mission time.

q Larger effective pinch areas or full-size HPl line breaks-on the HPl pump side-l are more limiting in terms of core cooling if EFW is lost and the broken HPl line is not isolated. These breaks are larger than breaks which are just recognizable at higher RCS pressures and result in less HPI flow to the RCS due to increased i

i flow out the break. If EFW is lost for those break sizes that EFW is required to l

provide core cooling, the RCS may still repressurize, and the HPl flow to the RCS will be further reduced. If the broken line is not isolated, there will be inadequate HPl flow to cool the core.

The repressurization increases the difference in flow rates between the broken and intact HPl lines, and as the RCS repressurizes, the isolation criterion will be met. If the operators isolate the high j

flow HPl line, core cooling will be assured.

4.3 BWST Drain Time The time required for the BWST to drain is a function of the initial BWST inventory, the HPI flow, and if actuated, the RB spray flow rate.

By j

conservatively modeling these flow rates and initiation times for the HPl and RB spray, the limiting time required to drain the BWST can be calculated.

Tables 2 through 4 represent HPl flow rates for various break types and include 10 percent equivalent head degradation on the pump head versus flow curve. In the cases where an early BWST drain time is conservative, the nominal HPl pump H-Q curve was used, with no pump degradation, to determine the total pump flow (Reference 5). The flow rates are higher and result in an earlier BWST drain time.

For the cases where it is conservative-to maximize the BWST drain time, the flow rates in Table 2 are used with no adjustment.

Framatome Technologies, Inc.

2' I',

51-1266138-01 l

LOBA Sinale Failure in the LOBA single failure, EFP-2 is relied on to supply EFW throughout the

' transient. The mission time for EFP-2 is defined by the time at which HPl flow can match core decay heat without the need for continued EFW. This mission time will be the longest when the HPI flow is at a minimum. Since the HPl flow j

rates will be smaller when taking suction from the BWST, it is conservative to L

maximize the time over which HPi is taking suction from the BWST by assuming that RB sprays are not actuated, and using the maximum BWST inventory available.

in order to calculate conservatively the time when the BWST is drained, the maximum technical specification inventory, less the inventory at the level at l

which RBES swapover must be accomplished (5.5 ft from Reference 9), must be determined. From Reference 9, the maximum BWST inventory defined in the CR-3 Improved Technical Specification is 449,000 gallons which corresponds to a level of 47.72 ft. There is an uncertainty in the level indication of 0.26 ft. The maximum level could be as high as 47.98 ft. After conservatively rounding to 48 ft., the maximum inventory is 451,533 gallons, Reference 10. The volume at the minimum swapover level of 5.5 ft is 51,702 gallons, Reference 10. The net i

inventory that can be drained from the BWST is :

BWST Volume

= (451,533 - 51,702) gallons

= 399,831 gallons The RCS pressure may not always decrease to the low RCS pressure engineered safeguards actuation system (ESAS) setpoint and therefore an automatic actuation is not expected for all of the break sizes of interest.

However, on loss of subcooling margin (SCM), the operators manually initiate HPl. This will occu'r early in the transient, but it will be conservatively assumed that HPl is not started by the operator until 20 minutes after the break opens.

The HPI flow rate will be taken from Table 2. The RCS pressure response for 0.0025,0.003, and 0.0035 ft' break cases are presented in Figures 5,6, and 7, Reference 11. The RCS pressure is less than approximately 1800 psig for most of the transient because EFW was available. The minimum total HPI pump flow when injection fluid is supplied from the BWST at 1800 psig is approximately 360 gpm. If the RB sprays are not actuated, the equivalent time into the transient that the BWST will be emptied for the LOBA is:

Drain Time after LOCA 399,831 gallons / 360 gpm + 20 min

=

1130.6 minutes (or 18.84 hours9.722222e-4 days 0.0233 hours 1.388889e-4 weeks 3.1962e-5 months )

=

i Framatome Technologies, Inc.

22

I 51-1266138-01 1

^

LOSB Sinole Failure in the LOBB single failure, EFP-1 may be relied on to supply EFW flow for a i

portion of the transient. EFP-2, however, must be crosstied to the 'A'-powered l

EFW control valves to provide the required EFW before EFP-1 is secured. EFP-1 may be secured due to one of three operational limita:. ions: First, the CC l

cooling system must be loaded on the A-EDG at approximately one hour (Reference 12), and EFP-1, the motor-driven EFW pump, must be secured to prevent exceeding the allowable EDG loads. Second, if the BWST is drained,

}

EFP-1 must be secured to allow the decay heat pump to be started so that l

recirculation from the RBES can be initiated. (This issue is addressed below.)

Third, an operator-initiated cooldown below the 500 psig RCS pressure EFW/LPI interlock would cause EFP-1 to be tripped, and the decay heat pump to be started. In this single failure only one OTSG atmospheric dump valve (ADV) is available. With a restricted relief capacity, the RCS can not be cooled to the 500 psig RCS pressure EFW/LPI interlock setpoint before CC cooling needs to be established.

In any case, EFP-2 must also be able to provide EFW at the reduced OTSG pressures that may occur over the EFW mission time. If EFW is lost due to any of these operational limitations before break /HPl cooling is adequate to remove core decay heat, the RCS can repressurize, possibly resulting in inadequate core cooling.

In the LOBB single failure case, it is conservative to calculate an early drain time for the BWST since this will result in the earliest loss of EFP-1 due to the need to initiate recirculation from the RBES. To calculate how long it takes for the BWST to drain, the minimum technical specification inventory, less the inventory at the RBES swapover alarm, will be used. From Reference 13, the minimum inventory in the CR-3 Improved Technical Specification minus the measurement uncertainty is 413,976 gallons. The RBES swapover alarm occurs at a BWST level of 15 feet, Reference 9.

Accounting for the uncertainty in level measurement of one foot, Reference 13, the minimum swapover inventory is 150,440 gallons. The net minimum inventory that can be drained from the BWSTis:

BWST Minimum Volume = (413,976 - 150,440) gallons

= 263,536 gallons The largest break size that will require EFW, with one operating HPl pump, is 8

approximately 0.04 ft. This is based on the RCS pressure response for the 0.04 2

ft cold leg break contained in Reference 2, shown as Figure 8. Larger break areas are capable of depressurizing the RCS to below the OTSGs and are large enough to pass all of the steam generated in the core. The RCS will not Framatome Technologies Inc.

23

,..m

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

51-1266138-01 i

i repressurize for these larger breaks and the ECCS injection flow will be sufficient l

to remove the core decay heat. From Figure 8, the minimum expected RCS l

pressure will be greater than 600 psig. For this case, the flow rate, based on a j

nominal H-Q curve for one operating pump, should be used.

Reference 5 l

provides a nominal flow rate for one pump and is 518.2 gpm at 600 psig, rounded to 520 gpm.

In addition to the HPI, RB sprays will also increase the rate that the BWST is drained. For the limiting break size with this single failure, the RB spray is l

expected to actuate. Actuation of the RB sprays increase the rate that the BWST is drained which is conservative for this single failure evaluation. From Reference 14, the maximum RB spray flow rate during the initial drawdown of l

the BWST is throttled to 1500 gpm per pump. Once the transition to recirculation from the RBES is made, the flow is further throttled to 1200 gpm per pump.

Corresponding conservative values, which include allowance for uncertainty, are 1600 and 1300 gpm (Reference 16). Since the time of interest is before the BWST is drained, the RB spray flow rate for suction from the BWST-1600 gpm-will be used.

Revised analyses for the 0.01 and 0.04 ft (the approximate break sizes requiring 2

EFW for 2 HPI pumps available and 1 HPI pump available, respectively) breaks were performed in Reference 15. Included in these analyses are cases with one and two RB fan coolers in operation. From Reference 16, when considering instrument uncertainty and repeatability errors, the RB spray actuation setpoint J

may ba as low as 21.6 psig. Figures 9 through 12 provide the RB pressure 2

responses for the 0.01 and 0.04 ft oreaks with one and two RB fan coolers. A summary of the results is presented in Table 5.

Table 5. Error-Adjusted RB Spray Actuation Time for 0.01 and 0.04 ft: CLPD Breaks RB Spray CLPD Number of Actuation Break Fan coolers Time (sec) 0.01 ft 1

3600 2

8000 2

0.04 ft 1

675 2

800 l

Framatome Technologies, Inc.

24

- ~ -.. -

i 51-1266138-01

~

l The break sizes requiring EFW flow for accident mitigation are a function of the

+

number.of operating HPl pumps, which depends on the single failure that is being considered. For the LOBB solution set, only one HPl pump and one RB I

fan cooler will be available, and the break sizes requiring EFW are for areas of 2

approximately 0.04 ft and less. The RB spray actuation time for the LOBB case will be 675 seconds (conservatively 10 minutes. This value was generated based on. conservative-Appendix K--core decay heat values.

Generally, if l

realistic decay heat were used, the energy addition rate to the RB would be less and the resulting pressure increase would be slower, resulting in a longer time to actuate the RB sprays. Therefore, the values in Table 5 are conservative for spray actuation times.

l-i For the LOBB single failure, the minimum BWST drain time for the limiting break i

i size of 0.04 ft'is:

i i

(520 gpm)*(10 min) + (520 + 1600 gpm)*(X) = 263,536 gal j

X = 121.9 minutes i

Drain Time = 10 + 121.9 = 131.9 minutes (2.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months ) i This time is longer than that required to initiate CC cooling, i.e., one hour.

Therefore, the draining of the BWST will not set the operability limit for EFP-1.

EFP-2 Sinale Failute For the EFP-2 single failure case, an early drain time is conservative because of the operability limitations on EFP-1.

These limitations are initiation of recirculation mode of injection or operator initiated cooldown to the 500 psig EFW/LPI interlock. In this case, it must be demonstrated that break /HPl cooling, with two operating HPI pumps, will be adequate at one of these operability limits l

to remove core decay heat.

Using realistic decay heat values and assuming that an operator may initiate an RCS cooldown, the RCS pressure may be reduced to near 600 psig, Figure 21 (Reference 11). A conservative total pump flow rate at 600 psig will be used.

I Reference 5 provides a nominal 2-pump HPI flow at 600 psig of 981.5 gpm, rounded to 985 gpm.

The RCS pressure may not always decrease to the low RCS pressure setpoint i

for the ESAS. Therefore, an automatic actuation is not expected for all of the break sizes. However, on loss of SCM, the operators will manually initiate HPl.

l Loss of SCM will occur near the time that tl.e reactor trips on low RCS pressure.

l This is early in the transient, within 2 to 3 minutes after the break is opened.

i Framatome Technologies, Inc.

25

_.m._

'51-1266138-01 Therefore, it will conservatively be assumed that HPl starts when the break opens.

The net minimum inventory that can be drained from the BWST is the same as for the LOBB single failure, 263,536 gallons. The break sizes requiring EFW flow for accident mitigation are a function of the number of operating HPI pumps, which depends on the single failure that is being considered. With EFP-2 being the single failure, two HPl pumps and two RB fan coolers are operating, and the break areas of interest will be less than roughly 0.01 ft. For the EFP-2 single i

2 failure case, the RB spray actuation time will be 8000 seconds (conservatively 130 minutes). This value was generated based on conservative-Appendix K-l core decay heat values. Generally, if realistic decay heat were used, the energy addition rate to the RB would be less and the resulting rmssure increase would be slower, resulting in a longer time to actuate the RB sprays. Therefore, the 1

values in Table 5 are conservative for spray actuation times.

Using the RB spray actuation setpoint for the 0.01 ft case described above, the 2

minimum BWST drain time for a single failure of EFP-2 is, l

(985 gpm)*(130 min) + (985 + 3200 gpm)*(X) = 263,536 gal X = 32.4 minutes Drain Time = 130 + 32.4 = 162.4 minutes (2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months ) i If one of the operating RB spray pumps is stopped, the drain time can be extended to 182.4 minutes, or roughly three hours. A summary of the drain times for the single failures are presented in Table 6.

Table 6. BWST Drain Times Single Break No. of BWST Comment Failure Size RB Spray Empty Time (ft)

Pumps (hours) l LOBA

<0.005 0

18.84 Long drain time is conservative LOBB 0.04 1

2.2 Short drain time is conservative j

EFP-2 0.01 1

3.0 Short drain tirne is conservative EFP-2 0.01 2

2.7 Short drain time is conservative i

I i

Framatome Technologies, lnc.

26

~ - -

4 51-1266138-01 4.4 Matchup Time Calculation The matchup time will correspond to the time at which the energy absorption capacity of the HPI fluid that reaches the core through the intact HPI lines i

matches core decay heat. For these small breaks, if the RCS repressurizes, the hot leg piping will drain before the core will uncover and saturated steam will pass through the PSVs. Core cooling will be through HPl/ break /PSV cooling 1

with the core acting as if it were a single pass heat exchanger.

The fluid temperature in the BWST is 120*F (conservatively greater than the j

100*F maximum design basis temperature).

At atmospheric pressure, the l

enthalpy is 88 Btu /lbm. The fluid temperature at the exit of the decay heat coolers is conservatively taken as 140*F (Reference 5) and at a pressure of 135 psig, the enthalpy will be 108.3 Btu /lbm. The saturated steam enthalpy at 2575 1

psig (2590 psia) is 1083.2 Btu /lbm.

l From Table 2, the break configuration resulting in the least amount of HPI flow at I

the PSV lift pressure,2575 psig, is for the HPl line pinch break. The matchup times for all cases are determined assuming recirculation from the RBES. With j

only one HPI pump available, the BWST will be drained and recirculation from the RBES will be established before the HPI flow can matchup with the core decay _ heat.' With two HPI pumps available, with suction from the BWST, the BWST will also be drained before the HPl flow will absorb core decay heat. The

- greater HPl flow rate when in recirculation (as opposed to taking suction from the BWST) is necessary to match decay heat in either scenario. The rated CR-3 core power level is 2544 MWt. A conservative power level of 1.02 times 2568 i

l MWt will be t sed to determine the HPl/ core decay heat matchup time.

One HPl Case from RBES-Pinch Break The fluid density at 140*F and 135 psig is 61.4 lbm/ft*. The HPI flow into the RCS at the PSV lift setpoint for the pinch breaks is 119.4 gpm from Table 2. The resulting mass flow rate is:

l W,

=(119.4 gpm)*(61.4 lbm/ft )/(7,4805 gal /ft')/(60 sec/ min)

= 16.33 lbm/sec The energy absorption of the HPI fluid is:

Q,

= W,

Ah Q,

= (16.33 lbm/sec)*(1083.2-108.3 Btu /lbm) 15,924 Btu /sec

[

/

i Framatome Technologies, Inc.

27

51-1266138-01 1

The equivalent decay heat fraction, based on an initial core power level of 2568 l

MWt, is:

l OngQ.,

= (15,924 Btu /sec)/(1.02*2568 MWt

948 Btu /sec/MWt)

= 0.00641 From Reference 6, this corresponds to 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months using 120 percent of the 1971 l

ANS 5.1 decay heat standard. Using realistic decay heat, 90 percent of the 1971 standard, this time is reduced to about 14.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months .

Two HPl Case from RBES-Pinch Break The fluid density at 140*F and 135 psig is 61.4 lbm/ft'. The HPl flow at the PSV lift setpoint is 283.8 gpm from Table 4. The resulting mass flow rate is:

Wng

= (282.5 gpm)*(61.4 lbm/ft')/(7.4805 gal /ft )/(60 sec/ min)

= 38.65 lbm/sec

)

l The energy absorption of the HPl fluid is:

i On,

= Wng

Ah Ong

= (38.65 lbm/sec)*(1083.2-108.3 Btu /lbm) l

= 37,676 Btu /sec The equivalent decay heat fraction, based on an initial core p.ower level of 2568 MWt, is:

l OndQ,

= (37,676 Btu /sec)/(1.02*2568 MWt

948 Btu /sec/MWt)

)

= 0.015173 I

From Reference 6, this corresponds to 1.62 hours7.175926e-4 days 0.0172 hours 1.025132e-4 weeks 2.3591e-5 months using 120 percent of the 1971 ANS 5.1 decay heat standard. Using realistic decay heat, 90 percent of the 1971 standard, this time is reduced to less than one hour.

A summary of the HPl/ core decay heat matchup times are provided in Table 7.

1 i

Framatome Technologies, Inc.

28 l

1 51-1266138-01 Table 7. HPl/ Core Decay Heat Matchup Time (EFW Mission Time)

No. of HPl Decay Heat Matchup Time' EFW Mission Timer Pumps Realistic Appent'ix K Realistic Appendix K 1

.14.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months 35.0 hours0 days 0 hours 0 weeks 0 months 18.84 hours9.722222e-4 days 0.0233 hours 1.388889e-4 weeks 3.1962e-5 months 35.0 hours0 days 0 hours 0 weeks 0 months 2

< 1.0 hours0 days 0 hours 0 weeks 0 months 1.62 hours7.175926e-4 days 0.0172 hours 1.025132e-4 weeks 2.3591e-5 months 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months

1) Decay heat matchup time is based on HPl flow in the recirculation mode.

2) The EFW m_ission time is whichever is the longer between the decay heat matchup time or BWST drain time.

4.5 Conclusion The minimum time that the BWST can be drained for these specific scenarios and break sizes has been determined. With one train of ECCS equipment,(one i

HPl and one RB spray pump), the BWST can be drained in 2.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months for the 2

limiting break size of 0.04 ft.~ With two complete trains of ECCS' equipment (two HPl and two RB spray pumps), the minimum drain time is 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months for the 2

limiting 0.01 ft break. If one RB spray pump is secured, leaving two HPI pumps and one RB spray pump operating, this time can be extended to roughly three hours.

In the LOBB solution set, the limiting operational time limit to crosstie the EFW trains was believed to be when CC cooling must be established. The earliest that this can be accomplished is estimated to be one hour into the transient. The BWST can be drained in 2.2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months . Based on this calculation, the loading of the l

CC. chiller is still limiting relative to the time dependency to crosstie the EFW trains as compared to emptying the BWST. In this single failure, only one OTSG ADV is available. With a restricted relief capacity, the RCS can not be cooled to the 500 psig RCS pressure EFW/LPI interlock setpoint before CC cooling needs to be established.

For the EFW mission time, with one HPl pump, the BWST will be drained before the HPl flow into the RCS is sufficient to remove the core decay heat via break /HPl/PSV cooling. Accounting for a higher enthalpy of the injection fluid during the recirculation mode, or piggy-back operation, the time for the HPl to i

matchup with the core decay heat is 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months using conservative decay heat j

levels. With realistic core decay heat, this time is significantly reduced to 14.2 i

hours, but the BWST will not be emptied. Therefore, under realistic decay heat S

Framatome Technologies, Inc.

29 m

n

, ~ ~ -

- ~ - - ~

i 51-1266138-01 i

[

assumptions, the EFW mission time is when the BWST is drained, or 18.84 1

hours post-LOCA.

I s

For the EFW mission time during the EFP-2 single failure with two operating HPl pumps available, the energy absorption of the HPI fluid can match the core decay heat earlier in the transient than with one HPl pump. However, due to the i

lower HPl flow rate when supplied from the BWST, the energy absorption of HPl 1

l fluid is still not sufficient to match the core decay energy before the BWST is

[

emptied using conservative, Appendix K assumptions. Once recirculation from i

l the RBES is established, the increased HPl flow rate is sufficient to match core j

decay heat. Therefore, for the break sizes that require EFW for mitigation with a

[

j.

- single failure of EFP-2, the EFW mission time will be when the BWST is emptied, i

or 2.7 hours8.101852e-5 days 0.00194 hours 1.157407e-5 weeks 2.6635e-6 months , i

i i

5.0 EFP-2 Operation In developing the solution sets (Figures 2 through 4) and during the reviews of the CR-3 design bases, potential operational limitations were identified for the

]

turbine-driven EFW pump, EFP-2. Specifically, the CR-3 Improved Technical Specifications and bases for EFP-2 define the operability of the pump in terms of a low OTSG pressure (< 200 psig) and the capability to provide accident analysis l

flow rates. The required accident analysis flow rate is defined early in the transient, i.e., within the first couple of minutes when the OTSG pressure is near the main steam safety valves (MSSVs) lift pressure. By the time OTSG pressure has been reduced below 200 psig for SBLOCAs that require EFW for mitigation, i

core decay heat has decreased substantially, and the high EFW flow rates are no longer necessary.

L The LOBA and LOBB single failures require long term steam generator cooling using EFP-2. The potential exists for the steam pressure to decrease below 200 psig. This can be a result of operator actions to cooldown the RCS using the OTSG ADVs, or due to the operation of EFP-2 (which draws steam from the OTSGs),with the RCS and OTSGs coupled or decoupled.

Without good coupling-primary to secondary heat transfer--supplying steam to provide motive power to EFP-2 may cause the OTSG pressure to decrease without a commensurate reduction in RCS pressure. In either case, if the operability of EFP-2 is challenged before it can be demonstrated that HPl/ break /PSV cooling is sufficient to remove the core decay heat, inadequate core cooling may result.

In order to address these possibilities, two tasks were undertaken: (1) information from the EFP-2 pump and turbine vendors was solicited and (2)

" proof of principle" analyses were performed using RELAP5 to ascertain the long-term operating conditions for EFP-2, given how the RCF and OTSGs interact for these very small SBLOCAs. The following sections describe, for the Framatome Technologies, Inc.

30 l

51-1266138-01 single failures, the information that was received from the turbine and pump vendors and the results of the RELAPS calculations.

LOBA Sinale Failure For the LOBA scenario, EFP-1 is unavailable and EFP-2 alone must provide the required EFW. The issue for this single failure is the operability of EFP-2 at low' steam pressures. The low OTSG pressure can be a result of an operator-initiated cooldown, decoupling of the RCS and OTSGs while still providing motive steam to drive EFP-2, or simply providing motive steam to drive EFP-2.

LOBB Sinole Failure For the LOBB scenario,. both EFP-1 and EFP-2 will initially be operating.

However, EFP-2 must be crosstied to the 'A'-powered EFW control valves to provide the EFW before EFP-1 is secured. EFP-1 can not operate concurrent with either the CC cooling system operating, or the LPI pump when the RB.

sprays are operating, since this would overload EDG-1A. EFP-1 could also be tripped due to actuation of the 500 psig EFW/LPI interlock during an operator-initiated cooldown.

The issues related to the LOBB single failure are determining to how long it takes to crosstie EFP-2 such that flow can be coiitrolled through the 'A'-powered EFW control valves, and whether flow from EFP-1 can be preserved until that time.

The specific issues are: (1) loss of EFP-1 to initiate CC cooling or draining the BWST and initiation of recirculation from the RBES, (2) operator initiated cooldown of the RCS below the 500 psig EFW/LPI interlock setpoint, and (3)

EFP-2 operability at low steam pressures.

Section 4.0 addressed the loss of EFP-1 due to emptying the BWST, and concluded tnat having to start the control complex cooling system at one hour will occur before the BWST is drained. EFP-2 operability at low steam pressures and the RCS response to these SBLOCAs are discussed below.

When the EFW/LPI interlock is actuated, EFP-1 is tripped. The interlock is automatically actuated when the RCS pressure has decreased to 500 psig, concurrent with a LOOP. The interlock may be actuated because the break is large enough to pass all of the steam generated in the core or by an operator initiated cooldown of the RCS using the ADVs. For break sizes.that are large enough to depressurize the RCS to the interlock setpoint, EFW will not be required for mitigation because the break area is large enough to pass all of the core steam production. For the smaller break sizes, the main issue is operation and operability of EFP-2 over the mission time.

Framatome Technologies, Inc.

31

51-1266138 !

i in this single failure only one OTSG ADV is available. Wdh a restricted relief capacity, the RCS can not be cooled to the_500 psig RCS pressure EFW/LPI l

interlock setpoint before CC cooling needs to be established.

L EFP-2 Operability at Low Steam Pressures l

FPC and FTl contacted the EFP-2 pump vendor (Ingersoll-Rand now Ingersoll-Dresser) and turbine vendor (Terry Turbine-now Dresser Rand) to ensure a thorough review.

i FPC and FTl were advised by the Terry Turbine division of Dresser-Rand that l

there have been no changes to the -operation of the Terry Turbine since the l

original purchase specification (Reference 17). The purchase specification for l

the EFP-2 pump and driver called for a capability to provide EFW to the steam generators for heat removal and cooldown over the NSS steam pressure operating range from the steam generator design pressure of 1050 psig down to l

about 20 psig. This steam pressure is low enough to cool the RCS to the decay heat removal system cut-in temperature of 280'F.

Recent testing performed by Ingersoll-Dresser on a pump similar to the turbine-driven pump at CR-3 confirms operation of EFP-2 down to a turbine inlet pressure (assuming that the OTSG pressure and turbine inlet pressure are equal) of 20 psig will support a turbine speed of 1080 rpm.

Under these l

conditions, EFP-2 can provide total pump flow rates (injection plus recirculation) between 50 and 320 gpm, Reference 18.

FTl also performed an evaluation of the emergency feed water cavitating venturis that have been installed in the EFW lines to assess the effect on EFP-2 at low pressure (speeds). The characteristic response of the cavitating venturis is that the flow will be reduced with reduced inlet pressure, Figure 13. The flow I

limiting characteristics of the cavitating venturis, combined with the EFP-2 head capacity curve at 1080 rpm (Figure 14, Reference 19), less recirculation flow, yields the approximate maximum flow rate when operating at 1080 rpm. With i

the EFW valves to the steam generators full-open, the total EFP-2 pump flow is about 280 gpm l

l When the required OTSG level setpoint has been reached and EFIC has reduced the flow to both OTSGs to zero, the pump recirculation flow will be about 50 gpm at a pump speed of 1080 rpm. This satisfies the Ingersoll-Dresser l

recommendation and permits operation of EFP-2 at 1080 rpm with a steam generator supply pressure of about 20 psig. This also allows the RCS to be j

cooled to 280*F, the lower end of Mode 3.

Framatome Technologies, Inc.

32 u

I 51-1266133-01 Figure 15 provides the EFW flow necessary to match core decay heat. This figure includes realistic and 10CFR50 Appendix K decay heat contributions. As can be seen in this figure,230 gpm (280 gpm pump flow less 50 gpm for pump recirculation) is sufficient to remove core decay heat after about 70 minutes after reactor trip based on realistic decay heat. Using Appendix K decay heat,230 gpm will absorb core decay heat at 3 hours3.472222e-5 days 8.333333e-4 hours 4.960317e-6 weeks 1.1415e-6 months . It is also assumed that the OTSGs have been' filled to the loss of subcooling margin setpoint (LSCM) setpoint.

With the LOBA and LOBB single failures, only one HPl pump is operating. The mission tirre for EFW is about 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months , and by the time the OTSG pressure can be reduced sufficiently for EFP-2 to be challenged by low steam pressures,200 gpm of EFW flow will be more than sufficient to remove core decay heat. Based

. on these discussions, EFP-2 can perform the intended safety function at OTSG pressures as low as 20 psig.

SBLOCA RELAP5 Analyses Core cooling during a LOCA is assured by various safety-related systems and power supplies that supply ECCS inventory makeup to offset the break discharge and provide secondary side heat removal as needed to depressurize the RCS and augment the ECCS inflow rates. The key safety systems required to ensure successful LOCA mitigation are the emergency diesel generators, HPl pumps and the valves used to align the required flow paths, core flood tanks, LPI pumps and the valves used for flow path alignment, service water flow for the control complex coolers, building spray pumps and building coolers for containment pressure control, and emergency feedwater pumps and flow control necessary for steam generator heat removal.

These systems are used to mitigate any postulated break size from holes that exceed the normal makeup system capacity up a double-ended guillotine rupture of the hot leg pipe.

The successful mitigation of these events involves continuous removal of the core heat by liquid natural circulation, boiling, or forced convection cooling. The decay heat energy that is reraoved from the core must be either rejected to the steam generators or discharged out of the break.

The steam generators are the heat sink for normal operation and are the dominant heat sink for the smaller break sizes.

'If the break is sufficiently small, the HPl system replenishes the liquid lost out of the break and the RCS remains in single phase natural circulation. The EFW l

flow provides a continuous heat removal path when the primary system remains in natural circulation. As the break size increases or if the HPl RCS inflows are severely degraded, the HPI system will not be able to match the leak discharge and liquid natural circulation will be interrupted. The loss of circulation disrupts the continuous steam generator heat removal mechanism. Without circulation, the core will begin to boil or boil more violently.

Framatome Technologies, Inc.

33

51-1266138-01 i

j

'If the break is unable to accommodate all of the core energy and the steam i

j generator heat removal is not reestablished by either hot leg spillover or high-elevation boiler-condenser mode of cooling, the RCS will repressurize. If the break is very small, the break energy discharge will not match the core decay i

heat until the RCS reaches the pressurizer, safety valve lift pressure. Lifting of the pressurizer code safety valves increases the energy discharge from the i

system. The energy removal capability exceeds the core decay heat rate, and the safety valves resent after the RCS depressurizes. Cyclic pressurizer safety valve flow will continue until the core decay heat is removed without the safety valve energy discharge. The cyclic behavior will stop when the core decay heat declines with time or the reestablishment of steam generator heat removal. The i

reestablishment of steam generator heat removal may be in the form of a high-i elevation or pool boiler-condenser process With EFW available, the OTSG heat

}

removal ensures that the RCS can be depressurized to improve the HPI flow delivery to the core, and ensure that adequate core cooling is provided.

l Mitigation of the SBLOCA transient requires at least one HPI pump, one LPI pump, and one EFW pump with a sufficient supply of EFW liquid. There is a real

{

limit to the available EFW liquid source at the site and there may also be a 1

realistic limit to the conditions and specific time interval in which the turbine-driven EFW pump can be shown to be operable. These inventory requirements (Section 8.0) and mission times (Section 4.0) are functions of break size, break location, break type, and the HPl flow profile. The longest operability interval is determined by an HPl line pinch break (pinched on both the RCS and HPI pump side). The most restrictive HPI pump-side pinch is one that is " unrecognizable" at a pressurizer safety valve lift pressure of 2575 psig. The pinched line is unrecognizable when the pinched line is the highest flow line and it is less than 50 gpm (indicated) above the next highest flowing HPl line (Section 7.0).

After the BWST empties, the LPI pump is aligned to the sump and the LPl pump i

provides the suction for the HPl pump. At 2575 psig with only one HPI pump operable, the lowest imbw from the sum of the three lower flow lines is 119.4 gpm, Table 2.

From Reference 5, the actual individual HPI line flows at this pressure are: the pinched line is 120.2 gpm, second highest line is 41.7 gpm, and 41.7 and 36.0 gpm for the lower two lines. The instrument uncertainty is i

subtracted from the pinched line and added to the next highest flowing line to give a net difference of 50 gpm. This value corresponds to the revised HPI isolation criterion discussed in Section 7.0. If 119.4 gpm reaches the core and

- the RCS has repressurized to 2575 psig, this flow (when totally boiled-off from an inlet temperature of 140*F) will match the core decay heat at 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months using 1.2 times 1971 ANS 5.11971 fission product decay heat with B&W heavy actinides.

Framatome Technologies, Inc.

M i

51-1266138-01 i

i 1

\\

i The RCS evolution includes loss of EFW, with a loss of OTSG heat removal after

-i the level boils down to below the natural circulation setpoint or below the reactor coolant pump (RCP) spillover elevation. The loss of circulation and reduced HPl I

flow will result in core boiling that passes steam through the reactor vessel i

intemal vent valves and to the break.

At 2575 psig, an 0.00247 ft' RCS-side HPl line break will discharge a saturated steam mass flow rate that is equivalent to the HPI flow from the sump (The HPi

)

flow matches the core decay heat at 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months ). Breaks smaller than these sizes i

will cycle the pressurizer safety valves until the decay heat matches the break j

energy discharge. After 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months , adequate core cooling is assured because the amount of HPl flow that is'available for core cooling can not be totally boiled-off by the core decay heat. The long term cooling configuration without EFW is HPl through the pressurizer safety valves until the decay heat decreases and can be j

l matched by the break energy discharge, or offsite power is returned and EFW i

restored.

i l

To demonstrate the RCS and OTSG response to these break areas, a set of j

representative RELAP5 calculations were performed (Reference 11). The first case to be described corresponds to a 0.002 ft break in the CLPD region. In 2

this case, single-phase natural circulation was maintained and the HPl flow was throttled to maintain a 50*F subcooling margin.

EFP-2 was allowed to continually operate and no operator induced cooldown was modeled.

l The steam flow required to drive EFP-2 as a function of the steam pressure was modeled in the RELAP5 calculations. As can be seen in the pressure response, l

Figure 16, the steam drawn from the OTSGs to supply motive force to EFP-2, was more than sufficient to supply enough EFW to remove the core energy plus depressurize the OTSGs. The typical steam flow to drive the turbine-driven j

pump is approximately 10 lbm/sec at steam pressures sbove 280 psig. Since the RCS and OTSG remained coupled, this steam flow can be translated into an equivalent break area. At 1000 psia, a steam flow rate of 10 lbm/sec is roughly 2

equivalent to a 0.005 ft break. This area coupled with a small break in the RCS piping is more than adequate to depressurize both the OTSGs and the RCS several hours into the transient.

L At the end of the calculation, at 40 hours4.62963e-4 days 0.0111 hours 6.613757e-5 weeks 1.522e-5 months , the RCS is in stable single-phase L

natural circulation, Figure 17, and is 50*F subcooled, Figures 18 and 19. The i

OTSG pressure is about 235 psig and is decreasing at approximately 4.0 psi per hour. These results are based on an adiabatic steam piping boundary, SG isolation, and Appendix K core decay heat. From these results, as long as the RCS and OTSGs remain coupled, there will be sufficient steam production to i

j drive EFP-2 and the operators will not have to " manage" EFP-2 by cycling the steam admission valves.

I l

Framatome Technologies, Inc.

35

,v r-s., - - -

u

l l

51-1266138-01 In the second case, with a break size of 0.00h4 ft, the steam admission valves to EFP-2 were closed and EFW was stopped at 185 psig OTSG pressure. Since the RCS was in single-phase natural circulation, both the RCS and OTSG pressure began to increase. Once the SG pressure reached the MSSV lift setpoint, the steam admission valves to EFP-2 were opened, and EFW was started. The RCS and OTSG pressure response is included on Figure 20.

The mission time for EFW is 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months for these single failures.and HPl flow rates. In both cases described above, the RCS was adequately cooled and the OTSG pressure did not decrease below approximately 185 psig. Smaller break sizes will remain in natural circulation, and the expected results would be similar.

For larger break sizes, natural circulation may be interrupted. Therefore, a third case was investigated. This case, was for a slightly larger break size. 0.0035 ft.

2 Significantly larger break sizes will evolve to EFW spray or pool boiler-condenser cooling, and OTSG pressure will remain near the lift pressure for the MSSVs.

Smaller break sizes can remain in natural circulation as described above. In this l

case, the OTSG pressure, Figure 21, does decrease to approximately 135 psig in approximately 13 hours1.50463e-4 days 0.00361 hours 2.149471e-5 weeks 4.9465e-6 months because the steam flow required to drive EFP-2 l

exceeds the boil-off rate. The RCS 'and OTSG are loosely coupled, and the RCS remains saturated. The break /HPI cooling, with some OTSG heat removal, is l

sufficient to cool the RCS, i.e., the HPl and break flow rates are equal, if EFW is

{

l lost due to insufficient steam pressure to drive EFP-2, the RCS will repressurize.

However, in this case, the break is large enough at this point in the transient to l

limit the amount that the RCS can repressurize. Based on the HPI flow rates in i

recirculation mode (Table 2), a break area of 0.0035 ft, and at this time into the 2

l event, the maximum pressure that the RCS can reach is approximately 2400 l

psig. The mission time for EFW at this pressure is significantly less than if the RCS pressure is at the PSV lift setpoint due to the higher HPl flow rate at 2400 psig. In fact, the HPI flow is high enough in this case to refill the RCS and re-establish two-phase natural circulation.

The results presented above should be considered representative of a range of i

break siges and the results considered to be general representations of the system interactions. It can be concluded from the above discussions, that as long as the RCS and OTSGs remain coupled, little or no operator action to

" manage" EFP-2 is necessary.

However, if the RCS and OTSGs become decoupled or if the OTSG pressure decreases below 200 psig due to steam i

leakage, operator action may be necessary to preserve operability of EFP-2.

I l

4 1

i l

l Framatome Technologies, Inc.

36

)

51-1266138-01 i

l l'

6.0 EFP-2 Failure: Cooldown Restrictions The EFP-2 single failure solution set relies on EFW via EFP-1 for a relatively short period'of time during a SBLOCA to remove energy from the RCS. The mission time is relatively short (less than 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months or when the BWST is emptied, whichever is more limiting) compared with the other single failures because both

)

EDGs are available, and two HPl pumps are operating. Potential challenges to maintaining EFP-1 available are having to stop EFP-1 to load a decay heat i

pump on EDG-1A to maintain HPl injection capability during the recirculation cooling from the RBES, or an operator-initiated cooldown of the RCS below the 500 psig EFW/LPI interlock setpoint. This section will determine if, using the

{

available' steaming capacity, the RCS can be cooled below the 500 psig i

EFW/LPIinterlock setpoint.

In design basis analyses only safety-grade equipment is typically credited. For OTSG pressure control on the B&W-designed plants, the only safety-grade equipment available are the MSSVs.

The MSSVs are for overpressure protection of the OTSGs and not for cooling down the RCS. Therefore, in the design base accident analyses, no operator-initiated cooldown of the RCS below the MSSVs lift pressure is assumed.

Plant emergency operating procedures (EOPs), however, direct the operator to cool the plant down within the specific limits identified in the pressure temperature limits report (Reference 20). With a LOOP, the ADVs are used. For the EFP-2 single failure, it needs to be determined if the RCS can be cooled below the 500 psig EFW/LPI interlock setpoint before EFW is no longer needed.

6.1 RCS Cooldown To assess the effect of cooldown of the RCS toward the 500 psig EFW/LPI interlock setpoint, a RELAP5 analysis was performed. The RELAP5 model of the CR-3 plant was the same model used to assess the reduced HPi flow l

delivery rates prior to startup of Cycle 11 (Reference 21). This model was L

modified to use best estimate core decay heat (90 percent of the 1971 ANS 5.1 decay heat standard) and both ADVs. The' ADVs were limited to approximately 80 percent of valve wide open capacity (301,246 lbm/hr at 540'F and 948 psig),

consistent with the actual physical limitations placed on the valves, which is reflected in the main steam enhanced design bases documR' (Reference 22).

Using a more realistic decay heat is conservative for this analysis since, for a given break size, the RCS pressure will be lower-because of the lower decay

{

heat-and the effectiveness of EFW in cooling the RCS and reducing the system j

temperature willincrease.

Framatome Technologies, Inc.

37

I 51-1266138-01 A case simulating a failure of EFP-2 and LOOP, Reference 11, was run for a 2

0.0025 ft break in the CLPD region. In this case, two HPl pumps are available and operating. At 20 minutes (1200 seconds) into the transient, both ADVs were opened, and a forced cooldown of the RCS was imposed by decreasing the i

OTSG pressure at a rate consistent with a 100 F/hr cooldown. The 20-minute i

action time was judged to be a representative time based on typical operator actions required to mitigate a SBLOCA.

In the RELAP5 analysis, the required operator actions to make the transition i

from forced to natural circulation and to maximize HPl injection to the RCS was completed by 20 minutes into the transient, and the RCS cooldown was initiated.

Single-phase. natural circulation was maintained, and at 66.7 minutes (4000 seconds), the RCS was approximately 50*F subcooled, the minimum subcooling l

margin required by the EOPs. Based on the OTSG pressure response and i

throttling the HPI flow, it was projected that the 100 F/hr cooldown rate was fast enough that the RCS pressure would decrease below the 500 psig EFW/LPI l

Interlock setpoint. Therefore, the HPl flow was thr6ttled to maintain a 50*F subcooling margin, consistent with the EOPs, and the cooldown rate was l

reduced to 50 F/hr. These cooldown rates are conservatively in excess of those i

expected during a post-LOCA cooldown with adequate subcooling. At the time l

HPl was throttled, the RCS pressure was at approximately 1085 psig and the hot leg temperature was about 508'F (50 F subcooled).

At 2.5 hours5.787037e-5 days 0.00139 hours 8.267196e-6 weeks 1.9025e-6 months , EFW was stopped, simulating a switch to recirculation from the RBES. The OTSG levels were high enough to sustain natural circulation, and the cooldown rate was not affected. At the end of the analysis,2.8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months , the RCS was approximately 585 psig and 437 F (50*F subcooled). The OTSG pressure was roughly 285 psig.

Figure 22 provides the RCS and OTSG pressure and response. Figure 23 contains the RCS temperature response for this transient.

For larger break sizes, natural circulation may be interrupted and the RCS and OTSGs will decouple.

The OTSG pressure will continue to decrease with essentially no heat removal from the RCS. These larger break sizes, which require EFW, will evolve to an equilibrium pressure in the RCS in which the core l'

steam production and break flow are equal, and the 500 psig RCS pressure EFW/LPI interlock will not be actuated.

For the break sizes that are large enough to depressurize the RCS to the interlock setpoint, EFW will not be i

required for mitigation because the break area is large enough to pass all of the core steam production.

4 i

i

+

Framatome Technologies, Inc.

38

r 51-1266138-01 l

6.2 Conclusions in the cases analyzed, the RCS did not reach the 500 psig EFW/LPI interlock setpoint before HPI alone could remove core decay heat. However, had a uniform 100 F/hr cooldown been assumed with throttling HPl flow to maintain subcooling margin, the setpoint would have been reached. The RCS pressure is a function of HPI flow, breale size, core decay heat, and OTSG steam venting capability. Therefore, specifying a fixed cooldown limit is difficult. A more logical guidance, when both HPl pumps are available, would be to start the RCS cooldown at a reasonable time, maintain the cooldown rate within technical specification limits, and limit the OTSG pressure to a value above the 500 psig.

RCS pressure EFW/LPl interlock setpoint until an assured long-term source of EFWis available.

i initiating the RCS cooldown should follow the emergency operating procedures for mitigating a SBLOCA with a LOOP and loss of SCM. The limit on, OTSG pressure will depend on the kt leg / core exit subcooling margin and the core temperature difference. The case analyzed remained in single-phase natural circulation with approximately a 15'F temperature change across the cure.

However, a slightly larger break size or higher core decay levels could evolve into two-phase natural circulation and a different core AT. RCS subcooling and the temperature difference across the core are dependent on many parameters, i

l and it is difficult to specify a bounding cooldown limit. Therefore, to assure in all cases that the interlock is not actuated by an operator-initiated cooldown, it would be conservative to limit the OTSG pressure for a period of time until l

break /HPI cooling is adequate to remove core decay heat or an assured long-l' term source of EFW is available. For break sizes that are large enough to depressurize the RCS to the interlock setpoint, EFW will not be required for l

7 l

mitigation because the break area is large enough to pass a!! of the core steam l

production.

l i

7.0 Isolation of HPI Line Breaks l-The current HPI line isolation criterion contained in the CR-3 EOPs specif'es that using the low-range flow insNments, if only one injection line indicates flow greater than 75 gpm above the lowest line, the highest line should be isolated.

l' l

This is a one-time check that is performed at approximately 20 minutes after loss of SCM or actuation of the ESAS. The decision to isolate a broken line should

'also be made after all of the HPl lines ar? fully-opened, with normal makeup l

isolated. If normal makeup cannot be isolaud, or if an HPl injection valve in the l

broken line is failed open, additional actions may be required. In any case, the l

isolation criterion will identify the line that needs to be isolated, and as long as 39 Framatome Technologies, Inc.

i t

i

i 51-1266138-01 e

OTSG cooling is preserved, sufficient HPI flow will reach the RCS assuring adequate core cooling.

i This current criterion was based on one operating HPI pump and EFW always i

available. The criterion accounted for the limiting HPI pump in terms of head and flow, the CR-3 makeup system hydraulics model and flow uneartainties, and the limiting break location.

With two HPI pumps operating and EFW always j

available, the criterion was still valid.

4 With two HPI pumps operating and EFP-1 available for a period of time, the RCS

{

is in a safe and coolable state, but the existing isolation criterion may not be met when the operators reach the point in the EOPs that require isolation of a broken i

line if EFW is lost later in the transient, the break may not be large enough to l_

pass all of the steam produced by boiling in the core. The RCS pressure can j

increase, and the available HPl flow for core cooling will be reduced. If the broken line is not isolated, there will not be adequate HPl to cool the core. Since the existing HPI line isolation criterion is 091 periodically fosssessed throughout the duration of the transient. a new criterion must be established when two HPI pumps are operating and EFW is not available. This new criterion should be applicable for one and two operating HPl pumps and with or without EFW. The criterion should also be functional for other applicable single failures, e.g. an injection line failing to open or a makeup valve failing to close, and should not j

invalidate previous SBLOCA analyses.

i In the process of developing a revised HPl isolat!on criterion for CR-3, a concern was identified that was associated with the existing HPl isolation criterion. With the worst-case stack up of instrumentation uncertainty, particularly at low flow rates in the individual HPl lines, the potential exists for the operator to take a non-conservative action (i.e., failure to isolate) for a full HPl line break, given a

. single failure of a vital DC channel or EDG, With new instrument uncertainties, the current 75 gpm isolation criterion was shown to be acceptable, provided that EFW was available.

An altemative HPl line isolation criterion was devebped to more effectively discriminate a bro' ten HPl line, with or without EFW.

New instrumentation uncertainties were airo developed (References 7 and 8).

These new uncertainties were used to establish the revised HPl isclation criterion.

7.1 Breaks Considered for the isolation Criterion As a part of this effort to develop attemative HPI line isolation criterion, a broad range of breaks and single failures were considered. The major break classifications are: the classical SBLOCA, i.e. breaks in the CLPD piping, a core Framatome Technologies, Inc.

l i

51-1266138-01 flood line break, a full HPl line break, and the HPl line pinch break. Each of these breaks are discussed below.

Cold Lea Pumo Discharae Break - For the' worst-case break in the bottom of e

the RCS cold leg piping, the smallest amount of HPI flow will be delivered to

)

the core when the broken cold leg is supplied from the HPl line with the lowest hydraulic resistance (in the first 20 minutes of the accident). At CR-3, this corresponds to the A1 line which is the HPl nczzle fed with normal makeup. The HPl flow delivered to the core is conservatively calculated as the summation of flow through the three low-flow HPl lines. The minimum HPI flows for a CLPD break are given in Tables 2 and 4 for one-and two-HPl pumps, respectively. The flew raies v.'ere estab!!shed using a PIPF-PC

{

model of the CR-3 HPl system.

J Core Flood Line Break - A worst-case break of a core flood line at CR-3 will i

result in loss of ECCS injection capability from one core flood tank and one train of LPl. If a single failure disables the other train of LPI (e.g., loss of an EDG), then the core can only be cooled in three ways: (1) one core flood tank will passively inject through the intact core flood line, (2) the operating HPI

)

pump will inject through two HPl nozzles, initially, and then through four lines based on the timing of operator actions, with suction from the BWST, and (3) the operating HPl pump (s) will inject through all four HPI nozzles with suction 4

from the RB sump via the operating LPI pump (s). The HPI flow delivered to the core for this accident is calculated as the summation of flow through two HPl lines initially, and through all four lines thereafter consistent with the timing of operator actions.

Full HPl Line Break - For full HPl line breaks that meet the HPl line isolation criterion early during the transient (i.e., within the first 20 minutes), a break in the A1 line is assumed, which is the lowest hydraulic resistance HPl line prior to isolating normal makeup.

For breaks that are isolated later duiing accidents (e.g., single failure of EFP-2 with RCS repressurization and when HPl suction is switched to the RB sump), a break in the B1 line is assumed, which is the lowest hydraulic resistance HPl line after normal makeup is isolated. The HPI flow delivered to the core is conservatively calculated as the summation of flow through the three intact HPl lines.

l

. HPl Line Pinch Break - An HPl line pinch break is a break for which the cross-sectional area of the broken HPl line becomes constricted due to jet impingement and/or pipe whip. loads. For breaks that meet the HPl line

]

isolation criterion early during the transient (i.e., within the first 20 minutes), a spectrum of HPl line pinch breaks in the A1 line are assessed since the A1 line represents the lowest hydraulic resistance HPl line prior to isolating normal makeup. These HPl line pinch breaks are assessed at an RCS 4

l pressure of 1100 psig when judging the adequacy of operator actions. At 20 Framatome Technologies, Inc.

41

L

(

51-1266138-01 l

\\

I i

minutes into the event, EFW will be available and for this break size, the RCS pressure will approach the OTSG pressure. At higher RCS pressures, the flow difference between the broken and intact lines increases, and it will be i

easier for the operators to iaentify the affected line.

For breaks that are isolated later during the transient (e.g., single failure of l

EFP-2 with RCS repressurization), a spectrum of HPl line pinch breaks in the B1 line are assessed since the B1'line represents the lowest hydraulic

. resistance HPI line after normal makeup is isolated. This class of HPl line pinch break is limiting in terms of EFW mission time and is evaluated at an L

RCS pressure of 2575 psig, which corresponds to the error-adjusted lift setpoint of the pressurizer code safety valves because the HPl flow to the i

RCS is minimized.

The HPI flow delivered to the core for all types of HPl line pinch breaks is conservatively calculated as the summation of flow through the three intact HPl lines. The calculated HPl flows for a small-effective area HPl line pinch break are given in Table 2 for an identifiable pinch at 2575 psig. HPl flow

'l rates for other pinch-pressures are given in Table 3. These flow rates are based on a revised isolation criterion of 50 gpm between the highest and next-highest flowing HPI lines and the new flow measurement uncertainties.

7.2 Single Failures A variety of single failures were considered to assure the adequacy of the HPl isolation criterion. Each of the pertinent single failures, the consequences of the single failures relative to HPl system operability, and the reievant breaks listed above that are considered for the single failures are discussed as follows:

Loss of One Vital DC Channel-A single failure of one vital DC channel, up to and including a full battery (LOBA or LOBB), is postulated. The failure of one vital DC channel will result in the loss of one EDG. The loss of one EDG will result,in the loss of one HPI pump and the failure of two HPl injection valves in the closed position until the operator can cross-connect power to the operating EDG and open the valves.

This single failure is undesirable-because (1) it results in only one HPl pump being available for core cooling, and (2) it results in reduced HPl flow capability until a manual operator action is taken to open two of the HPl valves. The single failure of one vital DC channel is assessed for all of the breaks.

Failure of One HPl Valve in the Ooen Position - A single failure of one HPl valve in the open position is postulated in the B1 line since this is the HPI line i

with the lowest hydraulic resistance after normal makeup is isolated. Failure of an HPl valve in the open position is undesirable for those breaks that Framatome Technologies, Inc.

42 l

g-ai a

w w

e4-C"

i 51-1266138-01 require isolation (i.e., full HPl line break and HPl line pinch break). With i

respect to HPl flow delivery, the full HPI line break is always worse than the HPl line pinch break. Failure of the valve in the open position is undesirable l

because (1) it reduces the HPl flow delivery to the core, and'(2) it prevents termination of the HPl flow diversion. As such, a single failure of MUV-25.

(i.e., HPl valve in the B1 line) in the open position is assessed for a full HPl line break in the B1 line (Section 7.4).

Failure of One HPl Valve in the Closed Position - A single failure of one HPl valve in the closed pcsition is postulated in both the A1 or in the B1 lines i

since these are the lowest hydraulic resistance HPl lines before and after i

normal makeup is isolated. Failure of an HPl valve in the closed position is i

undesirable because (1) it reduces the HPl flow delivery to the core, and (2) it has the potential to mislead the operators when the HPI line isolation criterion i

is applied. With this in mind, a single failure of MUV-24 or MUV-25 is

{

assessed for all of the breaks lisied above.

i Inability to isolate Normal Makeuo - A single failure that results in normal j

makeup not being isolated is postulated. Failure to isolate normal makeup is j.

undesirable because (1) it reduces the HPl flow delivery to the core for l

breaks associated with the A1 line, (2) it has the potential to mislead the operators when the HPl line isolation criterion is applied, and (3) closure of l

the HPI valve (MUV-24) in the A1 line will not terminate the HPl flow diversion. With this in mind, a single failure that results in normal makeup not i

being isolated was assessed and is discussed further in Section 7.4.

i i

L Inability to isolate Seal Iniection - A single failure that rs.,gs in RC pump seal injection ngt being isolated is postulated. Failure to isolate RC pump seal

(

infection is undesirable primarily because it reduces the HPI flow delivery to i

the core. HPl flow diversion through the RC pump seci injection pathway is, i

however, limited by the hydraulic line losses in the RC pump seal injection pathway. Previous hydraulic analyses have shown that'even without isolating i

{

the RC pump. seal injection pathway, the HPl flow delivery with two HPl L

pumps operating is adequate to cool the core. As such, a single failure that results in the RC pump seal injection not being isolated does not require j

further evaluation.

j

. Failure of EFP In the current CR-3 HPl design, an interdependency exists between the HPl and EFW systems through the EFW/LPI interlock.

Historically, safety analyses have assumed that at least one train of EFW is available during the mitigation of all SBLOCAs. For CR-3, g!] EFW may be j

lost if a single failure of the turbine-driven EFW pump is postulated, and i

either the RCS pressure decreases below the EFW/LPI interlock actuation setpoint, or the BWST becomes depleted such that the HPI pump must be aligned to the RB sump via the LPI pump. If the RCS pressure decreases Framatome Technologies, Inc.

43

51-1266138-01 below the EFW/LPI interlock actuation setpoint before the BWST is depleted i

without any operator action to depressurize the OTSGs, it can be shown that EFWis no longer required because break /HPl cooling is adequate to remove L

. core decay heat. If the HPI pumps must be aligned to the RB sump when the BWST is depleted, then 311 EFW can be lost. If all EFW is lost and the RCS break size is very small, then the RCS may repressurize up to the lift pressure of the pressurizer code safety valvec.

i i

A single failure of EFP-2 is undesirable because (1) it appreciably reduces the HPl flow delivery to the core if RCS repressurization occurs, and (2) the j

effects of the failure will not be apparent early in the transient. Based on these considerations,' the operators will need to periodically check the adequacy of the HPI flow distribution to assure that a broken HPl line, if present, is isolated.

l For a single failure of EFP-2, the HPl line isolation criterion is evaluated for a l-CLPD break, full HPl line break, and an HPI line pinch treak. The core flood line break need not be considered because the break area is large enough L

that EFW is not required for core cooling.

7.3 Plant Configuration When evaluating the breaks and single failures described above, four major plant configurations were considered. These plant configurations are based on an assumed sing'e failure of one vital DC channel and an assumed set of operator actions (e.g., open all four HPl valves at 10 minutes). Note that every plant configuration does not have to be assessed for every break / single failure i

combination.

Rather, only the relevant plant configurations need to be considered.

The four major plant configurations are summarized as follows:

Confiauration A - Configuration A represents the 0-10 minute time period e

following ESAS action or loss of SCM, and assumes that HPI suction is from the BWST, two HPl injection valves are open, normal makeup is open, RC pump seal injection is open, and normal letdown is open.

Confiauration B - Configuration B represents the 10-20 minute time period, and assumes that HPI suction is from the BWST, four HPl injection valves l

are open, normal makeup is open, RC pump seal injection is open, and

{

normalletdown is open.

l Confiauration C - Configuration C represents the post-20 minute time period, e

and assumes that HPl suction is from the BWST, four HPI injection valves Framatome Technologies, Inc.

44 l

.~

51-1266138-01 are open, normal makeup is closed, RC pump seal injection is closed, and normalletdown is isolated.

Confiouration D - Configuration D represents the !ong-term operation after an e

' accident, and assumes that HPI suction is in recirculation mode, piggy-back, from the RB sump (via the LPI pumps), four HPl. injection valves are open, normal ~ makeup is closed, RC pump seal injection is closed, and normal letdown is isolated.

J 1

~

7.4 Single Failure to isolate an HPl Line Break or isolate Normal Makeup In developing an HPI isolation criterion, several single failures, in addition to those presented in Figures 2 through 4, are considered. Of the single failures i

described in Section 7.2, there are two cases where the HPl flow to a broken i

line can not be isolated:. failure to isolate an HPI injection valve arid failure to isolate the normal makeup isolation valve.

These single failures represent unique challenges to the plant operators in order to mitigate a SBLOCA.

In LOCA mitigation, there are two mechanisms to remove core decay energy, i.e., the OTSGs or break /HPI cooling. As long as EFW is avajable, the RCS pressure will remain low enough that sufficient HPl flow will be available to remove core decay heat. In these specific cases, two HPl pumps are operating and both EFW pumps are available. However, if one of these mechanisms are lost, like EFW for instance, operator action may be required to preserve (maximize) HPI flow to the core. Specific descriptions of these sing % failures are presented below:

Failure to isolate an HPl Line Break Failure to isolate HPl flow to the broken HPI line necessitates continuous OTSG cooling to maintain reduced RCS pressure and assure adequate safety injection flow.

Two trains of ECCS and EFW are immediately available to provide adequate RCS and core cooling. EFP-2 will effectively act as a steam dump, since it is powered by steam from the generators, while providing EFW flow to maintain OTSG level at the required setpoint. EFP-1 will also be available until the ECCS must be reconfigured to support recirculation, piggy-back operation.

At this coint, EDG-1A load management can be implemented (SW/RW pumps placed in " pull-to-lock"), and EFP-1 automatic trip on LPI actuation will be defeated. This essentially assures continued EFW availability to support OTSG l

cooling until RCS pressure and temperature reaches decay heat removal cut-in conditiens (Temperature < 300*F, Pressure < 284 psig). Once this condition is achieved, then one ECCS train can be configured for decay heat removal while the other train provides RCS makeup.

Framatome Technologies, Inc.

45

I 51-1266138-01 An additional strategy consists of splitting the MUP discharge header. This is discussed further in the text to follow.

if off-site power is restored, then the main condenser can be used in conjunction l

with the turbine bypass valves to facilitate a cooldown to decay heat cut-in conditions.

Once this condition is achieved, then one ECCS train can be configured for decay heat removal while the other train provides RCS makeup.

l Failure to isolate Normal Makeuo As with a single failure to isolate an HPl line break, both trains of ECCS and EFW are immediately available.

If normal makeup cannot be isolated, application of the isolation criterion will still identify the broken HPI line. Even if l

l the normal makeup isolation valve (MUV-27) is not closed, as long as EFW is available, sufficient HPl flow will be available to cool the core. In this scenario, i

the isolation criterion would identify the broken HPl line and would reduce the l

HP! flow to the broken line, but it would not terminate the HPI Sow diversion.

If EFW cannot be preserved, the operator can also split the MUP discharge header. This configuration results in one HPI pump feeding two intact injection lines while the other train HPl pump supplies flow to one intact line and the failed HPI line. Operator actions needed to accomplish this include power restoration to the motor operated cross-connect MU valve (MUV 3 or MUV-9) to enable a control room operator to close the valve. If power is not available and the l

radiation levels are tolerable, then an operator would have to close one of the manual cross-connect valves (MUV-4 or MUV-8) in the field.

7.5 New HPI Line isolation Criterion The above breaks, single failures, and plant configurations were evaluated to develop a revised isolation criterion. Based on the cases evaluated in Reference 23, a revised isolation criterion can be developed that provides sufficient HPl flow to the core for the various breaks and single failures listed above to demonstrate that core cooling will be assured. The new criterion is as follows:

At any time in an unthrottled condition, JE the highest-reading HPI line indicates tiow > 50 gpm higher than the next highest-reading HPIline,,[H@(isolate the high tiow HPIline.

This new criterion should be applied when all four injection lines are fully-open, with normal makeup isolated. If normal makeup cannot be isolated, or if an HPI

(

Framatome Technologies, Inc.

"6 1

I l

n

51-1266138-01 l

injection line valve in the broken line is failed open, additional actions may be required. In any case, the isolation criterion will identify the line that needs to be isolated. If subcooling margin has been maintained or restored before the l

criterion is met, and the HPl flow has been throttled, no isolation action should l

be taken for SBLOCA mitigation. If subcooling margin is subsequently lost, the i

operators should maximize the HPl flow before taking any action to isolate a broken line.

l l

The new criterion takes advantage of the latest CR-3 HPI hydraulics model and new flow measurement uncertainties, References 7 and 8. The new cnterion is logically similar to the existing criterion, but is much simpler.

The current l

criterion is biased to the lowest flowing lines which introduces significant uncertainties. The new criterion is biased to only the two high flowing lines, and there should be less flow measurement uncertainty. The new criterion would not result in the isolation of an intact line or fail to identify a line that is required to be isolated.

The revised criterion does require that the operators periodicaliy monitor the HPl flow splits to ensure that, for specific HPl line pinch areas, a broken line will be isolated if warranted.

1 8.0 EFW Inventory Requirements Based on the solution sets. EFW is required for a period of time until break /HPl cooling is sufficient to remove the core energy. Since EFW is required, an assessment was made of the inventory of emergency feedwater required until break /HPl cooling is adequate to cool the core.

This assessment was based upon the following assumptions and boundary conditions:

l initici Core Power of 1.02 2568 MWt with 10CFR50 Appendix K Decay Heat e

The 1.02 factor is required by Appendix K. The initial core power level of 2568 MWt is higher than the current licensed power level of 2544 MWt.

The one percent difference is conservative for a planned future power upgrade. The decay heat is required by 10CFR50 for t.OCA design bases accident analyses, i.e.120 percent of ANS 5.1 1971 using infinite l

irradiation.

1 i

l Framatome Technologies, Inc.

1 i

I 51-1266138-01 Over the period of interest, EFW fills the OTSGs to 95 percent on the operate range.

Consistent with EOP guidance, EFW will be used to fill the OTSGs to the loss of subcooling margin setpoint following a SBLOCA. The required inventory will include filling the OTSGs with subcooled liquid plus the balance to remove core decay heat. It is conservative to assume that the OTSGs are' initially empty and, in addition to removing core decay heat, EFW must also fill both OTSGs.

EFW must remove excess core decay heat until break /HPl/PSV cooling at 2575 psig is adequate to match core decay heat A pressure of 2575 psig assumes that EFW is lost and the RCS will instantaneously repressurize to the pressurizer safety valves. At this pressure, the HPI flow, heated to saturated steam, must match the core decay heat. This is valid because, before the core can uncover, the hot leg piping will be empty and steam will be available to pass to the pressurizer. The lift setpoint of the pressurizer safety valves is 2500 psig.

An additional three percent for valve accumulation is added for conservatism. The pressure at the top of the pressurizer will not be significantly different than that at the HPl injection nozzles, i.e. the RCS is saturated and partially void of liquid. Therefore, the same pressure can be used at both locations.

EFW temperature will be 120*F.

Primary metal and cooling of the fuel and cladding will not be accounted for in the calculation.

Over the relatively long time of interest, the potential heat addition from these sources will not significantly change the results.

Heat I,oss to the ambient will be ignored.

The required EFW inventory is calculated by integrating the core decay heat I

over time with an end time based on the mission time for EFW as calculated i

in Section 4.0.

During a SBLOCA, the core decay heat is removed through the break, via the J

L reactor vessel intemals vent valves, to the OTSGs, if natural circulation is maintained or by boiler-condenser cooling, and by heating the HPI fluid. Itis

{

difficult to determine the contribution of each potential source of energy removal

. because of the wide variations in RCS conditions that could exist. For instance, 5

Framatome Technologies, Inc.

48

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

51-1266138-01 if the RCS stays or becomes subcooled, the operators are given guidance to throttle the HPl flow to manage RCS subcooling margin. The amount of HPI flow i

is not known, and therefore it is conservative to not credit the potential for energy absorption. Similarly, all possible break sizes, types, and locations also need to be considered. - Since the fluid conditions at the break can change, a conservative assumption must be made to bound the break energy removal I

capability.

Since not all break sizes require emergency feedwater for the same period of l

time, a series of calculations for different break areas were performed. The l

results are presented in Figure 24.

The data presented in this ' figure are developed in Reference 11. These calculations provide an approximate value of the required EFW flow for various break sizes. A conservative assumption that the break (at critical flow conditions) will pass saturated liquid at 600 psia. For the break sizes that require EFW for mitigation, the expected RCS pressure will be greater than G00 psia. Below this pressure, other equipment is available to provide injection flow, i.e., the core flood tanks or LPI if the RCS pressure is low enough. The calculation integrates the core decay heat and then subtracts the integrated break energy. The balance must be removed by the OTSGs. Once the energy removed through the break exceeds the core decay heat, the required EFW inventory to remove core decay heat for a given break size is known. The amount of EFW needed to fill both OTSGs to the loss of SCM setpoint must be added to the integrated value to determine the total inventory requirements.

4 The liquid volume to fill both OTSGs to the loss of SCM setpoint, 95 percent on the operate range, is calculated below using information from Reference 27:

1 Loss of SCM setpoint level

= {(0.95)*(292 inches)+(102 inches)}/12 in/ft

= 31.62 ft The OTSG cross sectional area, from References 24 and 25, is OTSG Area = (43.4955 + 23.3308)*1.0152 2

= 67.84 *t The totalinventory in both OTSGs is:

2 OTSG inventory

= (67.84 ft )*(31.62 ft)*(2)*(7.4805 gal /ft')

l

= 32092.9 gallons As seen in Figure 24, break sizes above approximately 0.04 ft effectively do not 2

l need EFW to mitigate the transient and the inventory required is what is necessary to fill both OTSGs. At the opposite end of the curve, a "zero" break l

area, requires approximately 339,000 gallons of condensate over 35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months after Framatome Technologies, Inc.

49

51-1266138-01 5

reactor trip. This is based on the mission time calculated when one HPl pump is available, in the recirculation mode, with the limiting pinch break HPI flow rates in Table 2, Section 4.0.

l The 339,000 gallon inventory required over the 35 hour4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months EFW mission time is available on site. The total capacity of the EFW tank is 184,000 gallons, with a 150,000 gallon minimum capacity, as required by the Technical Specifications.

j in addition, water in the condensate storage tank (200,000 gallon capacity) and i

in the condenser hotwells (200,000 gallon combined capacity) is also available.

i i

in developing the isolation criterion contained in Section 7.0, additional single failures are considered. Specifically, if the single failure is one in which an HPl i

line fails to close or the makeup isolation valve fails to close, the mission time for i

EFW can be quite long. For these cases, Figure 25 was developed. This figure provides the total EFW inventory requirements as a formtion of time for a "zero" 1

break area using a conservative core decay heat value. In effect, the integrated i

core power based on the initial core power level (1.02 times 2568 MWt) is mu!tiplied by the integrated full power seconds (FPS) taken from Reference 6 at several times. No credit is taken for the energy removed by the break flow or to j

heat the HPI fluid. It is assumed that EFW must absorb all of the decay heat, so at each of the times, the integrated core decay energy is divided by the enthalpy -

l rise of the EFW fluid, approximately 1100 Btu /lbm, and the result is converted to j

a volume. Assuming a realistic core decay heat,90 percent of the 1971 ANS 5.1 standard, the required inventory can be reduced by approximately 30 percent.

i If credit is taken to split the HPI pump discharge header, then EFW mission time and EFW inventory requirements will be of the same magnitude as for the other

]

worse-case single failures evaluated, i.e.,35 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months and 339,000 gallons, i

i i

9.0 References

1.

Florida Power Corporation interoffice Communication, NOE 97-0584, Startup Team Solution Set - Safety Assessment (Revision 1), 5/23/97.

2.

Small Break Loss-of-Coolant Accident Analysis for B&W 177-FA Lowered-Loop Plants in Response to NUREG-0737, item II.K.3.31, Babcock &

Wilcox, BAW-1976A, Lynchburg, Virginia, May 1989.

' 3.

FTl Document 51-1245866-00, Reevaluation of HPl Requirements During Small Break LOCAs, April 1996.

4.

Florida Power Corporation Nuclear Operations Engineering, Crystal River Unit 3, Enhanced Design Basis Document for the Decay Heat System, System Code: DDH, Tab 6/3, Revision 6,9/27/96.

Framatome Technologies, Inc.

50

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

i 51-1266138-01 r

5.

FPC Calculation M-97-0026, Revision 0, HPl Pump Flows for SBLOCA.

j 6.'

FTl Document 32-1258134-00, Decay Heat for LOCA Analysis,9/26/96.

I J

7.

FPC Calculation 189-0037, Revision 5, Make-Up/HPi Flow LOOP Accuracy (Low Range),6/10/97.

8.

Florida Power Corporation Interoffice Communication, NOE 97-0230, CR-3 HPI isolation Criteria, 4/12/97. (See Also Suspected Design Bases Issue Review, PC 97-1505.)

9.

FPC Calculation I-91-0012, BWST Level Accuracy, Revision 3,11/20/96.

}

10.

FPC Calculation M-93-0017, Revision 2, Borated Water Storage Tank (DHT-1) Volurne,2/14/97.

11.

FTl Document 32-1266136-00, CR-3 Startup Team SBLOCA Analyses.

j

12.

FPC Calculation H-97-0001, Revision 0, Control Complex Transient l

Temperature Model.

13.

FPC Calculation M-95-0016, Revision 1, BWST Swapover and Minimum Allowable Level Evaluation,3/4/96.

14.

Florida Power Corporation Nuclear Ope ations Engineering, Crystal River Unit 3, Enhanced Design Basis Documerit for the Reactor Building Spray System, System Code: BS, Tab 6/4, Revision 5, Dated 3/3/97.

15.

FTl Document 32-1266137-00, CR-3 RB Analysis with 2 Fan Coolers.

' 16.

FPC Interoffice %rrespondence, PROG 96-0003, BS and LPI Actuation Points, S.K. Balliet to K.R. Campbell, June 28,1996.

17.

EFP-2 Purchase Specification: FPC RO 2931, Addendum B, FPC-322-013, PR-1374, November 18,1969.

18.

Ingersoll-Dresser Letter Paul J. Kasztejna (Supervising Design Engineer Ingersoll-Dresser Pump Co.) to Mark Liebmann (Florida Power Corporation), dated May 7,1997.

19.

Ingersoll-Rand, Pump Curve 082A, 9.25-inch Impeller, Part Number 4X9N3A/3B, May 7,1997.

Framatome Technologies, Inc.

51

l

+

51-1266138-01 j

20.

Methods of Compliance With Fracture Toughness and Operational

)

Requirements of 10CFR50, Appendix G, Babcock and Wilcox, Lynchburg, Virginia, BAW-10046A. Rev. 2. June 1986.

21.

FTl Document 32-1244465-00, RELAP5 CR-3 SBLOCA Spectrum.

22.

Florida Power Corporation Nuclear Operations Engineering, Crystal River Unit 3, Enhanced Design Basis Document for the Main Steam System, System Code: MS, Tab 6/10, Revision 7, 3/21/97.

23.

FTl Document 51-1266161-00, CR-3 Revised HPl Line Isolation Criterion.

l 24.

B&W Document 32-1159004-00, _ Task AS-4 Operator Actions to Reestablish Natural Circulation,2/20/87, 25.

BWNT Document 32-1229132-00, Oconee R5 LBLOCA EM,3/8/94.

1 1

26.

B&W Document 32-1158464-01, ANS 5.1 (1979) Decay Heat Curve Fit, 9/30/86.

27.

BWNT Document 51-1212232-01, Key Elevations for All Plants, 3/24/94.

1 i

The documents marked.with an asterisk are maintained and controlled by Florida Power Corporation. Per FTl procedures, use of these references are-i allowed in safety-grade calculations with the approval of the cognizant unit manager or contract manager. The signature below authorizes the use of l

these documents for input to this evaluation.

t i

r// A 4 mre m

%/n (Unit Manager / Contract Manager)

(Date)

L l

Framatome Technologies, Inc.

52 i

FIGURE 1.

w%

B SOLUTION SET FLOW PATH r

a O'

Si i

I e

f LIMITING SINGLE j

q l

e FAILURES i

i O>

I o

LOBA l

0 Loss of "A" Battery

~.

m b

I INITIAL l

N CONDITIONS 3

O r

r

,r r

RX TRIP MODE SINGLE

?

+

SBLOCA +

+

LOBB

?

I FAILURES LOOP Loss of "B" Battery t

2 e

E I

S e,~

=

i e

S i

e

w D

b h

w a

t; i

?'

i e;

i

N FIGURE 2. LOBA Single Failure Solution Set E

f s

t 1 HPl MUP-1B or MUP-1C s

PUMP a

a e

9 S

i o

'i i

3 i

U o

r No MITIGATION 09 STRATEGY g

,4 LOBA+

C 1 HPI PUMP FRAMATOME g

g EFP-2 I

EFW a: ADMIN -> AND +

O EFW FLOW

{

ANALYSIS C

l l

u EFW' i

FLOW E

S FWP-7 '

r DEFENSE IN DEPTH FWP-7 ->

DIESEL MOD S

SUCCESS PATH CHALLENGES V

j DEFENSE IN DEPTH t

M 7

U1 C

Coo

FIGURE 3.

LOBB Single Failure Solution Set

{

h M

?

Qg r

Q 1 "PI

(

MUP-1A or MUP-1B

)

PUM o

w a

3 d '

m m

INSUFFICIENT r

OPERATIONAL

%q

-> EFP-1 --*

EDG LIMITATION

" ~ ~ ~ ~ ~ ~ '

CAPACITY g

I i

S 3

=

-CONTROL COMPLEX I

0 o

COOLING (~1.0 hr) l

[ o j

O MITIGATION k

P 1

ea STRATEGY

' TIME DEPE,NDENCY

-BWST DEPLETION j

U LOBB->

1 HPI PUMP l

-500 PSIG N

=

(Operator Cooldown) e l

3 A

C i

.O EFW FLOW l

............_a l T s

8 g

o J

l N i

Y I

C

' r

+

r EFW '

FRAMATOME r

EFP-2 ->

CROSSTIE EFW TRAINS +

EFW

--> ADMIN + AND

FLOW W

ANALYSIS E

S S

r t

FWP-7 DEFENSE IN DEPTH i

r DlESEL SUCCESS PATH OD CHALLENGES l

DEFENSE IN DEPTH f

I m

i3 w

a S

t M

6 1

FIGURE 4 EFP-2 Single Failure Solution Set M

g 2

3 Og rm j

3 (b

2HPI HPI FLOW

[

r q

PUMPS ANALYSIS tb i

O

~

s L

s S

O

%O i

Og

~.m MITIGATION U

N I

STRATEGY N

FRAMATOME TIME g

EFP-2+

2 HPI PUMPS ANALYSIS C

~

(

EFW FLOW C

OPERATIONAL LOAD EFP-1 TRIP

~

e f

LIMITATION MANAGEMENT

+

DEFEAT 7*

AND -> ADMIN +

j t

SWITCH EFW LuoD.

E FLOW

= ES REACWAMN

- BWST

{ MOD (PTL)

DEPLETION m

P l

- 500 Psig S

(Operator Cooldown) i S

" FWP-7 '

r

-> FWP-7 -->

DIESEL DEFENSE IN DEPTH SUCCESS PATH MOD 2

CHALLENGES DEFENSE IN DEPTH V y

{

g n

i E

c:

2 I

l

=

FIGURE 5.

CR-3 0.0025 ft Break - Pressure Versus Time D,N 3

D 2400 q

N

---LEGEND g

PRZ, VOL 400-10

[

RV-UP, VOL36041 SEC SG-1, VOL 670-01 oy 2000-C n

^

SEC SG-2, VOL77041

~~

CFF, VOL 897 01 O

I

~

O 1

I.

~.m

'w t

1600-i r-

+

W t-t

+

3

_~_._i_.

--m, d

'[

l A

o

=

y y

g-i m, _.= = = n; -

l

; -

[

m

+

1 800-i

-e-

-4 I

4Q0 t

-e

-e-

-e th r

m is i

y o

0 8

16 24 32 40 48 56 64 g

x10 '

6 Time (sec)

( x10 3) l l

2 FIGURE'6.

CR-3 0.003 ft Break - Pre sure Versus Time i

L w

2400 g

?

N LEGEND Q

F

%o 0

PRZ, VOL 400-10

[

g RV-UP, VOL36041 g

^

SEC SG-1, VOL 670-01 t

N' l

i C

SEC SG-2, VOL 770-01 q

i

CFT, VOL 897-01 m

i n

I e

6 l

3 O

t I

O i

l 5

09 1600-4-

N l

D b

\\

M l

l 3

A i

t

,o I

~

i 1200-D i

4-m

-e 3 0 -.".

=. -

i y

t g

I t

800-s-

2 t

+

i

^

..,g

_.g:

i e

.p-..

I t

\\

t.n j

t.n L

i 0

7 i

0 8

16 24 32 40 48 56 64 i

i Time (sec)

( x10 3)

P m

o

~

i f

g FIGURE 7.

CR-3 0.0035 ft Dreak - Pressure Versus' Time MQ3 Q

w 2400 O

LEGEND m

63

PRZ, VOL 400-10 Q

RV-UP, VOL 360-01 y

SEC SG-1, VOL 670-01 r

3 2000-p _,_

j O

SEC SG-2, VOL 77041

+

4---

CFI',

VOL 897-01 I

g Q

N h

, 1600-i t

4 m

1200-L a

m i

800-I L--

4 I:

400-1

-1 i.-

_~_

tn n

0

?

16 24 32 40 48 56 64 M

x10 b

3 Time (sec)

( x10 3)

~

FIGURE 8 2

RCS and Secondary Pressures, 0.04 ft Break at RC Pump Discharge N4 3

Q O

E M

0 2r 3000 i

i i

a i

I 3O O

Legend:

P2 00 RCS Pressure

~

i O

- - - - - Secondary Pressure J

2000 E.

s y"

. - [-

F i

1500 I

i 1000 _%

i f

N

/

v 2

o o

500 t

i i

i I

cw S

O 500 1000 1500 2000 2500 3000 3500 c,o u

o Time, sec

~

s

51-1266138-01

.m k

~

o ed x

O 0

N

\\

c l

\\

l

.X a

\\

m j

.G i

u l

m i

l

\\

l O

\\

A

\\

a i

o

(

N

\\

i

-a

\\

l W

l I

ao

~m o

u i

o i

d 0

bf.

am I

m i

O4A h

i E

s s

l e

.s W

x m

1M-e U

m e,

ta M

D0 H

l A

_'s e

e e

e a

a e

e a

a a

a s

d i

d 6

i 4

4 i

n n

n i

i e

361SS3Bd 200 IN3NIW1N03 Framatome Technologies, Inc.

61

51 1266138 01 mu*

00 A

O

\\

N O

g

~'Ns I

\\

.u

\\

h, b

m

's o

1 a

O

\\

n

\\

4 N s N

a N,

e

~

u Ou

'r O

\\

u

)

d m

t ut m

0 w

e g

b_.!!

mm It x

t3x W

M m'

O

.m O

ed w

D0 H

k4 B

B B

B 5

3 B

B B

B E

4 e

x e

4 e

e e

e a

a 331SS38d 39n89 IN3 NIB 1N03 Framatome Technologies, Inc.

s2

i t

51-1266138-01 u*

H E_

O

" ~.

OU CO

\\

N i

N l

3 N

Q ma

's I

U

\\

no

\\

w

\\

~o 1

l 5

i

\\.

b u

O M

e

\\

u x

a

\\

v t

l t/J

\\

M e

l k

W f

a

-)

e we

=

W 1

^ \\t L

x y

i 7

l C

l e

O

~

e-4

.a 1

0 l

r3 Hk m._

B B

3 3

8 n

s s

e n

e v

4 x

e n

e e

e t

l 361SS3Bd 3DfBD IN3 NIB 1NO3 i

Framatome Technologies, Inc.

s2

1 51-1266138-01 l

m u

G (El i

~

O OO l

O f

E

\\

l

\\

l i

\\

x

\\

l

5 t

o e

u

~

w m

1 O

m a

x O

\\

~.o

\\

W O

J

\\

~m i

u O

~

s

\\

B

\\

d t

l 8

\\

W l

m

\\

0u 1

m 8

\\

~h c

e

=

s W

l G

k HX to i

W x

w-I C40

.m 1

N 1

ed i

M o

l C

l H

l m

's t

i e

a e

e a

e e

a a

n a

s i

n i

e n

n n

i e

e WISS38d 3DfBD IN3NIW1N0')

Framatome Technologies, Inc.

64

N NQ3D FIGURE 13.

EFW Cavitation Flow Versus Inlet Pressure

'D 3m em 750 O

[*-

5 700

/

a g.-

k 650 4

/*#

Q-i 550 s

/~~A' s D

.D g.

,i*

..a o

3 350 E

/

o 300 250

/

00 f

EF-62-FO l

/

150

- - - - - -EF-63-FO

/

100 l

g 50 0

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 Inlet Pressure (psia)

U1 C

Os C

?'

i

,N Q

FIGURE 14.

l 3

Florida Power Corp. Feed Pump Characteristic Curve for 1080 RPM l

D M

i.".. C,,'S,.;, ".O,""e"e*s*.* E*s*a.'.* *.*.*1 Curva tto. - CURVEOS2A Date. 07 h

t ea B kB#9 E' E

O

, JINUE L

e.

.w.r

. - e.e

. a, e

g i

e-

_ site.

g l

l

[

(9

~

I

._{_

Imp

Ear D'= meter-9.25 b

I

[

i linposer part no. - 4XSN3A/3B l

O

_l._ _i_.

.s_.

._ _f D $

i BW -NlN TF

3 3

20

[

{

I

._ l k

g g

q_

Notes : stECOMMENDED wa O

g l

O.q i

l MINIMUM SPEED O

g.

2 l

-I_

l A L_--

l 10ao steu w a

_4 i.

i.

4 i___.

b w

W I

i.

8 4

W g

a

~

o i

I

.j.

I

_.i_

I I

l 1

_l _

.._i___.

i._

z t

8 t.,

l I

l- - -

go

___1_

EF7 CCNpY l'

l

-i

_. i._ _ _; _

l l

z

_ l-

_; y

)

80 I

i

{

o i

i I

i TQTA t. HEA D l

- r---

g---- -L

's

. -t --

t M..r l

l l

1

280 3/

,I g, -

/

_l l

O

_-i 7

L

_,i _.

70 s

' I, k

_ i_ _

7 e-A I

i i

24o t

I

=

I l

7~-- _

e I

60 h

_l-l..- f' _

L Custom curve for:

1

~

l

/

n 2,

I 3'

FLORIDA POWER CORP E

200

]

/

l l

-t-3 Service - FEED PUMP 50 m.

l b

160 i

l--

7 j

._ !.. l

__. _._ j-g

__j_.

..__ q_ _

y_.

3 i

4OO f

l 30 I

I_._

l

-l'--

ownr

_ i.__

i

___ p J_ _

i l

i i

i i i i lI

/

i Characteristic Curve 20

~

_t_ _. _..

_. L M

os j

-.}

I Piznp she & type - 4X9ff-7 m

speed-1080 RPM 1O 1

l

__ l _ -.

j _.._

seeses no. l 6

l i

g order no. - -oo

~

O SO 100 150 200 250 300 350 O

OALLONS PER MINUTE

51-1266138-01 i

l 1

i i

. 8 7

E I

i I

. 8 I

I i

is I

8 m

I

=

N

)e g

Q

-l -

g lx l

O a

l U

.c

'. b

]

_ _ - -l

- - G e

T E

I, f

.E W

O l

4J

~l i~ 8 y

a

~.

Y" 3

I I

w w

8 7

I e;

I I

g am*

j~ g g

l c --

H f

I

1. 'R

/

h '/

. # l o

o o

o (md6) moLi Irl:1 Framatome Technologies, Inc.

N N

2 Q

FIGURE 16.

CR-3 0.002 f t Break - Pressuro Versus Time 3

Q

%D 20 3

i t

LEGEND

[

PRZ, VOL 400-10 o

RV-UP, VOL 360-01 D*

SEC SG-1. VOL 67041 D

2000-O SEC SG-2, VOL 77041 g

i

CFT, VOL 897 01

~

09

~m 4

i 3

,60o.

,i

.n 6

w

~

i T

Y

~ ~ ' * * "

p. = = c -

i n

gon -

y t-L i

= = 2: =so g

4 i

+

e,

~==== = =i=.

7; i

m c

03 0

Y 0

2 4

6 8

10 12 14 16

~

Time (sec)

( x10 9

2 FIGURE 17.

CR-3 0.002 ft Break - Mass Flow Versus Time Q3 Q

3000 4

LEGEND

~Q 0

Clel A, JUN 16541 q

m C1,1B JUN I85 01 O

Cle2A, JUN 26541

~D 750-4 i

0 Cle2B, JUN 285-01

~

+--

4 O

I

~.

%b 500-1 D

\\

9 2

mA

\\

.g 1

y"'

b f',M,TW O k h,bir 6... -.

4 CO to 1

42

~

1 e

l

-250-

- - - +

+

y --.

b ta 500 oo

.b 0

2 4

6 8

to 12 14 1;

Time (sec)

(x104)

h FIGURE 18.

CR-3 0.002 ft Break - Temperature (CVs 175/275) Versus Time Q

.Og 650 m

LEGEND M

- SG-1 Inlet, VOL 110-1 h

SG-2 inlet. VOL 210-1 2.

l A, RV INLET, VOL 175 C

2A, RV INLET, VOL 275 O

600- h-VOL352 Tsat NO i

OQ

~.

"U-E-R l550-N --

j I

"--- -~- -

cr _-

s j

I I

m 500-

. ------j

+s -

p i

t G

N

.-m, 4

...s..

L

.s--~~-.

A s

e i

g, 4

y

..._w 7

rer i

u

~

i 350 o

0 2

4 6

8 10 12 14 16*' '

Time (sec)

( x10 4)

N Q

IGURE 19 CR-3 0.002 ft Break - Temperature (CVs 195/295) Versus Time r

il Q

650 O

N LEGEND 0

SG-1 Inlet, VOL 1101 q%

SG-2 inlet, VOL 210-1 D

t.

i I A, RV INIEr, VOL 195

[

600- 1

,i O

"tA, RV INLET, VOL 295 i

+ ~ -

o i

VOL 352 Tsat O

OQ t

~.

i Cc 4

u 550-N N

+-

4

+

r

+-

5 le i

bm 500-

+

+ - - -

+

- + - - - - -

e

a

$ 450-I-

a 4 --

l--

i 400-

+-

M w

H 350 i

a

=

0 2

4 6

8 10 12 14 16 6

Time (sec)

( x10 4)

N NQ FIGURE 20.

CR-3 0.0014 ft Break - Pressure Versus Time 5

Q 2400 T-O N

LEGEND M

9 O

PRZ, VOL 400-10 RV-UP, VOL 36041 h

^

SEC SG-1, VOL 670-01 2000-

] --

j C

SEC SG-2, VOL 77041 3

-t

--+

4---

O

CFT, VOL 897-01 O

~

4 1600-r i

+ - - -

-4 g

b-i W 1200-D

+

E i

= =

q I

300 I

. I_

i m

i i

400-g. :

. 5

+ - - -

+

--+

t.#t i

I U

m m

4" C

0 0

2 4

6 8

10 12 14 16 2

t Time (sec)

( x10 4)

N N

2 Q

FIGURE 21.

CR-3 0.0035 ft 3

Break - Pressure Versus Time Q

2400

s M

LEGEND M

O 0

n

- PRZ, VOL 400-10 RV-UP, VOL36041 f.

SEC SG-1, VOL 670-01 3

2000-j-

a --

0 O

SEC SG-2, VOL 77041

-+

g 09

CFT, VOL 897-01

~

M

.59 N

1600-4-

-4

-- b-3O i

i

-i 4-d I

~

i MA 1200-a 4--

H--

-4 e

i m

i A

b

~

g g

4

)

i i

s i

e : :

- - t 5

i 400

-3 7

p-I I

7-

._v.

T l

I C

I m

a 0

[

?

I i

w 0

8 16 24 32 40 48 56 64 6

Time (sec)

( x10 3)

FIGURE 22 CR-3 0.0025 ft2 Break (2 HPI Pumps) - Pressuro Versus Time Q4

-D 2400 Q"

i 3

__ LEGEND m

0

PRZ, VOL 400-10 9m X

RV-UP, VOL36041 0

SEC SG-1, VOL 67041 2000 l

O O

I SEC SG-2, VOL 77041

+

CET, VOL 89741 3

o t

09

~

m 3

t.

i 1600-4 u

t i

- + - - - - - - - - -

9 eA i.!

i M

i gE 1200-E-

o m

4--

i-t

~

6 s

300 _ _ _.. l.. -

c-

_4 _

-j

-__.-..-j.

i I

e i

g.

9,._~.

.y 9._.,.._...;.

l ta 1

-m m

i G

i 1

e -.

a O

e m

i

=.

0 2

4 6

8 10 12 14 16 L

-e r

Time (sec)

( x10 3) x 10,

t l

l'

=-. -

FIGURE 23.

CR-3 0.0025 ft Break (2 HPI Pumps)

. 43

- Temperaturo Versus Time 4

650

%O i

LEGEND

~

SG-1 Inlet, VOL 110-1 h

SG-2 inlet, VOL 210-1 O

IA, RV INLET, VOL 175 y

600-

+ - - - - -

5-- -

C 2A, RV INLET, VOL 275 3

4 O

VOL352 Tsat

%O 09

~3m 6

350- s-N-

n 1.----

6 sm g

300 4 _..

N

._p 9

[

i p

i 9' <so-r J

400-

--i-j m--

+

1 i

(A.

i i

u 0

2 4

6 8

10 12 14 16 Time (sec)

( xio 3)

%l$gw?3 1

0 90 0

no ig e

R 80 D

0 PLC n

i 70 0

A CO LBS 60 t

0 so P

)

2R d

(

e e

5 z

r 0 i S i

0 k

u ae q

r e

B R

w 4

o 0

0 lF W

FE 30 l

0 a

toT 20 4

0 2

ERUG 1

0 I

0 F

0 0

=

5 0

0 0

3 3

2 5

0 5

1 1

iUElrE kw N3gp%ON% 9%Dh3O O9%%b 3o-0 y

at

.k~

51-1266138-01 8

a 8

I a

4J O

H 8

R

%ex 8

~

g2 e

1 0w R

3 b

z b

\\

R M

i B

i 8

8 t$

\\

~

W

.oo N

l n

n (le6) pennbeg WWE Framatome Technologies, Inc.

77

i 1

FLORIDA POWER CORPORATION i

CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302/ LICENSE NUMBER DPR-72 ATTACHMENT E i

l ASSESSMENT OF LIMITED USE OF RELAP5 a

1 i

l l

\\

su F R AM ATO M E TF4HHOL0GIa5 Integrated Nuclear Services June 11,1997 l

INS-97 2323 1

j Mr. F. X. Sullivan (NAlE) be:

E. E. Organ /OF54 l

Flosida Power Corporation J. C. Seals'0153 Crystal River Energy Co aplex J. D. Carlton/OF49 l

15760 West Power Line Street J. J. Cualin/OF53 Crystal River, FL 34428 6708 T. G. Stack /OFS4 B. L. Brooks /OF54 Attention:

Mr. D. F. Kunsemiller (SA2D)

Record Center NSS7/TI.2 Mr. D. Ri:e

Subject:

Assenment of J.imited Use of RELAPS

Reference:

FPC Contract NPM010AD, WA73, Task 4100929 - Safety Analysis Support

Dear Mr. Sullivan:

At the request of Mr. D. F. Kunsemiller and Mr. D. R!ce, FTI is prodding an assessment of limited use of RELAPS. Tlic anessment is attached.

T!te assessment finds that FPC has satisfied NRC critena for reliance on BAW-10192P and finds RELAPS as the licensing basis code to confirm EFW requirements for long term cooling.

This work was perfortned under the referenced FPC Work Authorir.ation. Should you have.any questions, please give me a call at 804/832-2574 Ve.ry truly yours, L. M. Lesniak Customer Service Manager LML/bec i

Attachment c:

R. J. Finnin/OF57 l

R. W. Knoll /FPC (N102) 3315 Old For est Rved, P.O. Scx 1o935, Lynchburg. VA 24506-0935 Telephone: 804 632-3000 Fax: 604 832-3663 Internet hnp//WWW.tramatech.com

ASSESSMENT OF LIMITED USE OF RELAP5 I. Background FTI performed engineering evaluations based upon RELAP5/ MOD 2 analyses to support the mission times for and operability of the turbine-driven emergency feedwater pump (EFP-2) during certain small break LOCA transients. These evaluations were used to demonstrate the RCS pressure and temperature responses and their relationship to EFP-2 operability. The RELAP5 analyses were performed with FTI's NRC-approved version of RELAP5/ MOD 2, and, with two exceptions, were done with the ECCS evaluation model (EM) as approved in BAW-10192-P. The EM adequately determines the transient evolution; however, its primary purpose is to calculate a conservative peak cladding temperature (PCT) following a postulated LOCA. The EM models and methods are developed to predict core uncovering and resulting PCTs for transients with typical end times of less than 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months . The EFP operability analyses examine very small breaks for durations of several days with abundant core cooling. Given that these analyses were not specifically performed to address core cooling, two adjustments were made to the EM input deck to facilitate code execution.

These changes were (1) increases in the uur requested maxiruum time step and (2) a switch from equilibrium to non-equilibrium control volumes in the heated core region.

The approved evaluation model requires a small time step and equilibrium core volumes for SBLOCA core uncovering and PCT analyses.

These changes will not affect the RCS or secondary-side pressure or temperature evolution as demonstrated in previous EM sensitivity studies.

II. NRC Topical Report BAW-10192-P The NRC documented its review of the subject topical report (dated February 1994) along with Framatome's May 6,1996; October 11, 1996; and January 7,1997 responses to NRC requests fc additional information in a February 16,1997 letter to Framatome.

On the basis of its review, the NRC concluded that BAW-10192 (hereinafter, "the Topical") is acceptable for referencing in licensing applications in the analysis of LOCA accidents for once-through steam generator plants. The NRC also noted that when the report is referenced in a license application (as exists in the subject case), it will not repeat its review of matters described in the report and found acceptable.

I See letter from James E. Lyons, Acting Chief, Reactor Systems Branch, NRC to J.H. Taylor, Manager, Licensing Services, Framatome Technologies Inc.; " Acceptance for Referencing of Topical Report BAW-10192-P, ' Loss-of-Coolant Accident Evaluation Model For Once-Through Steam Generator Plants' (TAC No.

M89400)".

of the NRC's February 18, 1997, letter states, in part, that use of the Topical methodology for reference in licensing applications involving large and small break LOCA analysis for B&W plants is acceptable, subject to eleven conditions. Accordingly, justifying use of the Topical in licensee actions by comparing Code use to the eleven conditions is necessary for plant specific utilization of the Topical. In that regard, this text addresses those eleven conditions and demonstrates that use of the Topical in support of this license amendment is acceptable.

III. NRC Conditions for I icensee Use of BAW-10192-P FTI provides below, its evaluation of the eleven conditions that must be satisfied for a licensee to justify use of the Topical:

1. The LOCA methodology should include any NRC restrictions placed on the individual codes used in the evaluation model.

Response

FPC (with Framatome being the 'implementer of the Code) has satisfied all NRC restrictions placed on the use of RELAP5 as defined in the evaluation model presented by Framatome to the NRC in its letters dated February 1994 and as supplemented in correspondence dated May 6,1996 except for the two input options listed in Item I and described in detail in Item III.2 below.

2. 1he guidelines, code options, and prescribed input specified in Tables 9-1 and 9-2 in both Volume I and Volume II of BAW-10192P should be used in LBLOCA and SBLOCA evaluation model applications, respectively.

Response

Given that these analyses were demonstration cases used for proof of principle, and that no core uncovering would be predicted for the break sizes analyzed, an increase in the maximum time step is acceptable so long as the code convergence is maintained. The j

increased time step size is required to reduce the clock time required for analyses that have end times in excess of 30 reactor hours. The code convergence was maintained by keeping the time step increase within the confines of sensitivity studies and benchmark analyses used to validate the RELAP5/ MOD 2 code for EM application. The maximum time step used in these analyses was increased from the typical EM analysis value of 20 milliseconds per time step to 50 or 100 milliseconds per time step in these analyses. EM sensitivity studies performed in Reference 1 evaluated the variations on results with time step sizes ranging between 10 and 50 milliseconds. The RCS pressure behavior was nearly identical for all of the cases. Typical MIST SBLOCA benchmark analyses (Ref. 2) successfully used 200 to 250 millisecond time steps. Therefore, the time steps used for these SBLOCA cases are well within justified ranges for acceptable r.nalytict.1 convergence of the RCS evolution and its relationship to EFP-2 operability.

I i

i l

The switch from equilibrium to non-equilibrium in the core was made to facilitate code execution during long-term quasi-steady RCS pressure periods. Previous EM analyses for l

HPI line break cases have encountered execution difficulties with code water property l

searches during these periods. The dif6culty has been traced to the generation of vapor in the core by the subcooled boiling wall heat transfer model and subsequent condensation by the high interphasic heat transfer coefficients that are forced by the application of the equiliSrium control volume option. The code failures can be overcome by reductions in the maximum time step; however, this simply increases the run times without altering the calculational results. Abundant fuel pin cooling is predicted during subcooled or saturated nucleate boiling, and the non-equilibrium switch will not alter this conclusion or affect the RCS pressure behavior. The non-equilibrium volumes can alter the state conditions used for the EM heat transfer package at very high void fractions in the post-CHF regime. The transition and film boiling heat transfer correlations included in this package were formulated based on thermo-dynamic equilibrium conditions.

Use of non-equilibrium control volumes has been shown to be reasonable in sensitivity studies, but its use is not totally consistent with the heat transfer package formulation. The applicability of core heat trassfer package using the non-equilibrium formulation for the EFP-2 operability cases is supported by these previous studies and even more so by the fact that the post-CHF heat transfer regimes will not be predicted during any of these analyses since there is no core uncovering.

If core uncovering and heatup were to be predicted, the equilibrium formulation must be used in all control volumes that predict a transition or film boiling regime.

3. The limiting linear heat rate for LOCA limits is determined by the power level and the product of the axial and radial peaking factors. An appropriate axial peaking factor for use in determining LOCA limits is one that is representative of thefuel and core design and l

that may occur over the core hfetime. The radial peaking factor is then set to obtain the limiting linear heat rate. For this demonstration, calculations were performed with axial peak of 1.7.

The general approach is acceptable for demonstrating the LOCA limits methodology. However, as future fuel or core designs evolve, the basic approaches that were used to establish these conclusions may change. Fil must revalidate the acceptability i

of the evaluation modelpeaking methods if: (1) sigmficant changes arefound in the core i

elevation at which the minimum core LOCA margin is predicted or (2) the core maneuvering analyses radial and axial peaks that approach the LOCA LHR limits difer i

appreciablyfrom those used to demonstrate Appendit K compliance.

i

Response

Tae analyses performed did not relate to core kw/ft limits, so this restriction does not apply.

4. The mechanistic ECCS bypass model is amptable for cold leg transition (0.75 ft* to i

2.0ft ) and hot leg break calculations. 7he nonmechanistic ECCS hpass model must be i

used in the large cold leg break (2 2.0 ft*) methodology since the demonstration l

calculations and sensitivities were run with this model.

)

i 1

,,w w

.~

4

Response

For SBLOCA analyses, the ECCS bypass model is typically not executed. ECCS bypass is a phenomenon applicable to blowdown during a large break of the reactor coolant piping in -

the reactor coolant pump leg. Therefore, this restriction does not apply to the analyses referenced in this report.

5. Time-in-hje LOCA limits must be determined with, or shown to be bounded by a specipc application of the NRC-approval evaluation model.

Response

No LOCA limits were calculated; therefore, the restriction does not apply.

6. LOCA limits for three pump operation must be established for each class of plants by application of the methodology described in this report. An acceptable approach is to demonstrate that three pump operation is bounded byfourpump LHR limits.

Response

L No LOCA limits were calculated, therefore; the restriction does not apply.

7. The limiting ECCS conpguration, including minimum versus maximum ECCS must be determinedfor each plant or class ofplants using this methodology.

Response

This restriction applies to analyses of large break LOCAs. The use of minimum flow rates is per se the limiting case for SBLOCA analyses. In the analyses done to support the l

evaluations in this document limiting ECCS tlow rates were assumed, i

8. For the small break model, the hot channel radial peaking factor to be used should i

correspond to that of the hottest rod in the core, and not to the radialpeakingfactor of the 12 hottest bundles.

Response

.The hot channel radial peaking factor used corresponded to the hottest rod in the core, i

Since'no core heatup calculations were performed, this restriction is not relevant.

.9. The constant discharge coeficient model (discharge coefcient = 1.0) referred to as the "High or Low Break Voiding Normalized Value," should be used for all small break analyses.

The model which changes the discharge coefcient as a function of void fraction, i.e.,' the " Intermediate Break Voiding Normalized Value", should not be used unless the transient is analyzed with both discharge models and the intermediate void methodproduces the more conservative results.

~

l

Response

The constant discharge coefficient model (discharge coefficient = 1.0) referred in the BAW-10192P analysis as the "High or Low Break Voiding Normalized Value," was used for the SBLOCA analyses performed to support the evaluations in this document.

l 10.For a specific application of the FTI small break LOCA methodology, the break size which yields the local maximum PCT must be identiped.

In light of the diferent possible behaviors of the local maximum, FTL should justify its choice of break sizes in each application to assure that either there is no local maximum or the size yielding the maximum local PCT has beenfound. Break sizes down to 0.01ft* should be considered.

Response

The analyses done to support the evaluations reported in this document were not done to predict limiting PCT consequences for SBLOCA, therefore the requirement does not apply. The break sizes analyzed were smaller than those specified in the requirement.

11. B&W-designed plants have internal reactor vessel vent valves (RVVVs) that provide a path for core steam venting directly to the cold legs. The BWNTLOCA evaluation model credits the RVVV steam flow with the loop steam venting for LBLOCA analyses. The possibility existsfor a cold leg pump suction seal to clear during blowdown and then reform during reflood before the evaluation model analyses predict average core quench. Since the REFLOOD3B code cannot predict this reformation of the loop seal, FTIis required to run the RELAP5/ MOD 2-B&W system model until the whole core quench, to conprm that the loop seal does not reform. This documentation should be performed at least oncefor each plant type (raised loop and lowered loop) and bejudged applicablefor all LBLOCA break sizes.

Response

This restriction applies to post-blowdown RCS behavior following a large break LOCA.

None of the cases analyzed to support the evaluations in this document were for break sizes sufficient to produce formation of a cold leg pump suction seal. Therefore, the restriction does not apply.

IV. Interface of RFI APS Use and CRAFT / THETA GSAR Chanter 14)

FPC has evaluated whether the use of RELAP5 for SBLOCA analyses is compatible with the us of CRAFT / THETA in the CR-3 FSAR Chapter 14 accident analyses. In sum, FPC concludes that no interface problems exist as a result of the use of the two codes because (1) the RELAP5 code has been reviewed and accepted by the NRC for application to SBLOCA transients, and (2) the analyses performed for this application do not supplant or alter any of the SBLOCA analyses reported in Chapter 14 of the CR3 FSAR.

1

{

I

1 l

V. Licensing Basis Based on the above discussions, FPC has satisfied the NRC's criteria for reliance on BAW-i 10192-P in this license amendment.

Accordingly, for purposes of SBLOCA analyses to confirm EFW requirements for long-term cooling, FPC considers the' licensing basis code to be RELAP5 (BAW-10192-P) as approved by the NRC in its letter dated February 18,1997.

l References

1. BWNT Document 32-1224876-00, "EM SBLOCA Sensitivity Studies I," March 1,1994.

2. Klingenfus, J.A. & Parece, M.V., "Multiloop Integral System Test (MIST): Final Report," NUREG/CR-5395, EPRI/NP-6480, BAW-2078 (Vol.10), December 1989.

i

FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302/ LICENSE NUMBER DPR-72 ATTACIBIENT F SUPPORTING INFORMATION i

i l

l t

ATTACHMENT F l

SUPPORTING INFORMATION 1

As discussed in the cover letter, Florida Power Corporation (FPC) is addressing the issues l

related to small break loss of coolant accident (SBLOCA) mitigation for Crystal River Unit 3 (CR-3) through a combination of Technical Specification revisions, Emergency Operating l

Procedure (EOP) revisions, and plant modifications. This integrated approach ensures that

(

necessary accident mitigation systems are available and maximizes defense in depth. To l

support NRC's review of this submittal, the following listed information is provided in this l

attachment.

]

l Table Title

i 1

NRC Identified Unreviewed Safety Questions (USQ) 2 Modifications (MOD) l 3

Operator Actions (OA) l 4

Summary of Planned FSAR Changes L

5 Related LERs (LER)

Letters in parentheses indicate cross-reference designations i

used for item references to other tables in this attachment.

i

1. NRC Identified Unreviewed Safety Questions This table lists those issues that have been previously identified by the NRC as containing Unreviewed Safety Questions (USQ). The resolution of these NRC Identified USQs along with references to other tables in this attachment are provided.

2. Modifications This table provides a listing of the modifications to be performed during this outage related to SBLOCA mitigation issues. These modifications are currently in vario2s stages of implementation and, therefore, the information provided in this table is draft.

Implementation of these modifications will support the Technical Specification / Bases changes in Attachment C, resolve issues identified as USQs, and resolve outstanding issues related to SBLOCA mitigation.

3. Operator Actions To support the NRC's review ai the TSCRN 210, two tables are provided to identify certain operator actions required for the mitigation of SBLOCAs as presented in

}

Attachment B.

However, the EOPs affected by this TSCRN are in various stages of revision and, therefore, the information provided in these tables are draft. Table 3A i

i provides a complete list of operator actions relied upon in the first 20 minutes for these SBLOCA scenarios. Table 3B identifies the additional operator actions, beyond those i

4 1

l 2

+

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 2 currently in the operating procedures, required after the first 20 minutes of these SBLOCA scenarios.

Net all of the operator actions listed in this table would be required for all SBLOCAs. As indicated in the table, some of these actions would only be required to be implemented under certain low probability scenarios. Additionally, some of these actions are considered to be " defense in depth" and are not considered in design basis mitigation analyses.

FPC has been able to reduce the number of operator actions required in the first 20 minutes of these SBLOCA scenarios relative to the previous requirements. Specifically, one operator (OA2) will address more than one previously required operator action. These operator actions in conjunction with the proposed Technical Specification changes and plant modifications have been taken in an effort to minimize the operator burden during the response to a SBLOCA.

FPC requests specific NRC review of the operator actions presented in Tables 3A and 3B as an integral part of the amendment review.

4. Summary of Planned FSAR Changes This table is a summary of planned changes to the CR-3 FSAR based on the modifications and changes to the SBLOCA analysis as provided in this submittal. The information in this table is draft and based on the most recent information available for the modifications scheduled to be completed this outage to support the SBLOCA analysis. The finalized changes to the FSAR will be completed via the FSAR change program. As stated in the cover letter, these changes will be addressed prior to restart.

5. Related LERs This table lists those events related to this submittal that have been previously reported to the NRC as voluntary reports and unanalyzed conditions. The descriptions provide a reference to previously reported issues and refer to resolutions contained within this TSCRN.

U.S.' Nuclear Regulatory Commission I

3F0697-10

~ Attachment F Page 3 p

Table 1 j

NRC Identified Unreviewed Safety Q--l u j

USQ USQ Description References Resolution f

1 EDG Inadino (ASV-204 Mod)

A. This modification increased the EEI 96-12-02 Modifications The implementation of MAR % potential loading of the A EDG such MODI, MOD 5,' MOD 6, j

12-01 (removal of EFIC automatic that the design load limit of 3500 kW LER1 MOD 7, MOD 8, MOD 9,,

open signal for ASV-204) prevented (TS Basis 3.8.1; FSAR 8.2.3) would MOD 10 l

the automatic start of EFP-2 on a be exceeded for short periods of one

[

I failure of the B train vital DC power, to three seconds during cenain EDG TS/Raus Chances thus preventing a NPSH block loadings.

)

concern for EFP-1. However, EFP-1 B. This modification increased the -

Parts 2, 3 would need to pump more water and automatically connected accident thus, if there were a concurrent load at the one-minute interval to j

LOOP, would represent a larger load 3159 kW in excess of the minimum j

on the A EDG.

test load specified by TS SR

[EEI %-12 first example]

3.8.1.11. The TS Basis stated that l

the minimum load of 3100 kW

}

provides margin above the predicted j

worst-case automatically connected accident load at one minute.

C. This modification increased the motor-driven EFW pump load to 666 kW which exceeded the TS SR Basis j

3.8.1.8 statement that the largest single post-accident load (that the A

'i EDG would have to reject) was 616 kW.

3 I

l

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 4 Table 1 NRC Identified Unreviewed Safety Questions USQ USQ Description References Resolution 2

EDG Loading (EOP-13 Revisiold A. Design load limit of 3500 kW (TS EEI 9612-02 Modifications The procedure change for EOP-13 Basis 3.8.1; FSAR 8.2.3).

MOD 5, MOD 6, MOD 7, increased the motor-driven EFW B. Automatically connected accident LERI MOD 8, MOD 9 pump post-accident load from 666 kW load at one-minuie exceed 3100 kW to 713 kW by directing operators to (TS SR 3.8.1.11)

TS/ Bases Changes take manual control and increase EFW flow. The C. Largest post-accident load for reject Parts 2,3 resulting EDG load was greater than (TS SR Basis 3.8.1.8) that calculated to support MAR % 12-01 and, therefore, the same three USQs as discussed above were introduced.

[EEI 96-12-02, second example]

3 EDG Loading (OP-402 Revision)

EEI 96-12-02 Modifications The procedure change for OP-402 increased the post-accident IIPI pump load on MOD 7, MOD 8, MOD 9 the A EDG by 75 kW and on the B EDG by 86 kW by allowing operators to ES LERI select the swing B IIPI pump to either EDG. The post-accident load of the B IIPI TS/ Bases Changes pump (691 kW) exceeded the TS Basis value of the largest single post-accident Part 3 load (616 kW).

[EEI 96-12-02, third example]

U.S. Nuclear Regul: tory Commission 3F0697-10 Attachment F Page 5 Table 1 NRC Identified Unreviewed Safety Questions USQ USQ Description References Resolution 4

EFW NPSII EEI %-19-03 Modifications TMAR T87-10-09-01 and MAk %-04-124)1 (removal of EFIC automatic open MODI, MOD 2, MOD?

signal for ASV-204) increased the probability of a malfunction of equipment LER2,LER3 MOD 4 important to safety (damage to EFP-2 due to insufficient NPSH). Additionally, the plant design basis relied upon EFP-2 to share the EFW flow with EFP-1 TS/ Bases Changes (EFV-12 crosstie) in order to maintain the A EDG within its loading limits. The Part 1 safety evaluations failed to identify this potential USQ.

[EEI 96-19-03]

l 5

ASV-204 Automatic Ooen Signal Removal EEI 96-19-06 Modifications Removal of the automatic open signal from ASV-204 disabled one of the two MODI automatic steam supplies to EFP-2. This reduced the reliability of EFP-2, which LER 1, LER2 in turn, increased the probability of a failure of EFP-2. This was an increase in TS/ Bases Chances the probability of occurrence or malfunction of equipment previously evaluated in Part I the FSAR.

[EEI 96-19-06]

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 6 Table 1 NRC Identified Unreviewed Safety Questions l

1 USQ USQ Description References Resolution I

6 Ooerator Actions NRC Inspection Report FPC's response to the Added and revised certain operator actions associated with the mitigation of small 50-302/97-06 Violation has not been break LOCA prepared at this time.

[97-06-01]

Those operator actions addressed by the Violation are included in this submittal and FPC requests NRC review of these operator actions as an integral part of the amendment review.

i f

h

., ~,.. -.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 7 Table 2 Modifications MOD MAR Subject Description Reference 1

96-11-01-01 ASV-204 EFIC Auto Restores the automatic opening of ASV-204, the steam admission valve to EFP-2, USQl,USQ4, opening reinstallation on an "A" EFIC actuation. This will restore the load sharing capability of the USQ5,LER1, Emergency Feedwater System for the LOCA concurrent with LOOP and loss of LER2,LER3, EDG-1B in order to reduce the load on EDG-1 A LER4 2

96-10-02-01 Emergency Installs passive flow restricting devices on the discharge side of both EFP-1 and USQ4,LER2, Feedwater Cavitating EFP-2. This will prevent excessive pump flow resulting in possible failure LER3,LER4 Venturis mechanisms of mnout or inadequate NPSH available.

3 96-10-10-01 EFV-12 Valve Replace valve EFV-12 on the cross-tie piping between EFW train A and train B, USQ4,LER2, 96-10-10-02 Mods, MOV a manual operated gate valve, with a motor operated gate valve. This will LER3,LER4 96-10-10-03 Installation, Conduit facilitate operator action to open this valve remotely and route discharge of EFP-Supports 2 through the cross-tie piping to the OTSGs.

4 97-01-04-01 EFP-2 Flow Installs flow indication from the cavitating venturis installed downstream of EFP-USQ4,LER2, Indications

2. This control room indication of EFP-2 flow rate will be powered from the LER3,LER4 opposite train ('A' side) to provide flow indication should a 'B' side failure disable its flow indication. This will provide feedback to the operator of flow from EFP-2 when EFP-1 needs to be secured for EDG load management.

5 97-04-01-01 EFP-1500 psig Trip Installs a control switch to allow operator action to defeat the automatic trip of USQl,USQ2, Defeat Switch EFP-1 (500 psig RCS pressure). Defeating this trip will allow EFP-1 operat;on LER1,LER4 during a SBLOCA. This switch will allow continued EFP-1 operation when DHP-1 A starts on a 500 psig actuation, after EDG-1A load managemmt by operator action.

6 97-04-02-01 RW/SW Pumps Pull-Replaces existing control switches with a Pull-To-Lock switch on Nuclear Service USQl,USQ2, To-Lock Switches and Decay Heat Seawater pumps RWP-2A and -2B and Nuclear Services Closed LER1,LER4 Cycle Cooling pumps SWP-1A and -1B. This will prevent automatic restart of these pumps on subsequent Engineered Safeguards actuation signal facilitating EDG-1 A load management.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 8 Table 2 Modifications MOD MAR Subject Description Reference 7

96-12-17-01 EDG Small Load This modification will remove the auto-start function from both nonsafety control USQl,USQ2, Reduction circuits of the Flush Water Pumps. This will prevent them from auto-loading USQ3,LER1, Modifications, DOP onto the EDGs.

LER4 2A/2B 8

96-10-05-01 Diesel Power Uprate Implements modifications to increase the service ratings of the EDGs. (1) The USQl,USQ2, Project combustion air flow rate will be increased by replacing nozzle rings in USQ3,LER1, turbochargers with larger ones, and (2) combustion air intercoolers will be LER4 replaced with a dual pass intercooler.

9 96-03-12-01 Emergency Diesel Installs more iccurate power meters (kW indication) for EDGs-1 A and -1B.

USQ1,USQ2, Generator Indication Accuracy was further improved by changes to CT/FT's. EDGs can be leaded USQ3,LER1, associated Upgrade higher because of improved instmment accuracy.

LER4 FCN's 10 96-06-02-01 EFIC Integral Installs windup reset on integral controller on the EFIC system. This will USQl,LER1, Windup Reset provide for faster response of EFW for control of flow to the OTSGs. This LER4 reduces EFW flow and consequential EDG-1 A loading upon initiation.

11 97-03-01-01 Standby Generator Installs a new diesel generator (not safety-related) to provide an alternste backup for FWP-7 power supply for FWP-7.

12 97-02-17-01 MUV-27 HPI Changes the Engineered Safeguards automatic actuation logic for the normal Autoclosure Makeup supply valve MUV-27 to add automatic closure upon receipt of a diverse containment isolation signal (which also initiates HPI). The purpose of the modification is to aid in HPI flow balancing actions in the event of a broken HPI line. MUV-27 must be closed to help ensure accurate HPI flow indication.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 9 Table 3A Operator Actions Less Than 20 Minutes O

Operator Action Assumed Time Basis Reference A

in Design Analysis 1

Trip all running RCPs Yes

<2 Required for NRC leiter to FPC dated 5/29/86 (Generic Ixtter 86-05) minutes loss of refers to B&W Owners Group (BWOG) studies which subcooling concluded that compliance with 10 CFR 50.46 is achieved if margin operator action to trip RCPs is taken within 2 minutes.

based on voiding condition of reactor coolant.

i f

1 1

i P

l b

ame

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 10 Table 3A Onerator Actions Less Than 20 Minutes i

O Operator Action Assumed Time Basis Reference A

in Design Analysis 2

If Subcooling Margin Yes

< 10 Required for NRC letter to FPC dated 7/6/79 (SER for Order dated (SCM) is lost and ES has minutes loss of 5/16/79 based on TMI-2 Accident) recognizes CR-3 revision not actuated, initiate (isolation subcooling to Emergency Procedure EP-106, which defines operator l

manualIIPI and Reactor ofletdown margin action in response to a spectrum of break sizes. States EP-l Building Isolation and and (precedes 106 was " judged to provide adequate guidance to the Cooling (RBIC) makeup automatic operators to cope with small break LOCA." EP-106

- isolates letdown were initiation)

(currently EOP-03, "Los.s of Subcooling Margin") contained l

(USQ6) previously guidance to initiate HPI and ensure adequate IIPI flow.

i

- initisms IIPI flow identified f

- isolates normal as makeup (USQ6) separate (contingency actions actions) are provided in OA 4 I

within 20 minutes if power is not available) l

- isolates RCP seal control bleed off j

valves

- actuates EFIC

- initiates Emergency l

RB coolmg i

l

=

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 11 Table 3A Onerator Actions Less Than 20 Minutes O

Operator Action Assumed Time Basis Reference A

in Design Analysis 3

Ensure all four IIPI Yes

< 10 Required NRC letter to FPC dated 5/29/79 " Permanent Solution to injection valves are open minutes only for loss SBLOCA Issue" recogmzes operator action to turn

- switch power supply of 1 train of associated transfer switcec, open affected IIPI valves by 10 for affected injection Class IE minutes.

valves by manipulating power i

switches in control room e

k f

f k

b l

m

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 12 Table 3A Oncrator Actions Less Than 20 Minutes O'

Operator Action Assumed Time Basis Reference A

in Design Analysis 4

Isolate RCP seal Yes

< 20 Required to FPC letter to NRC dated 2/28/79, answers a previous injection (USQ6) minutes maximize question of whether or not it was necessary to isolate any HPI flow to flow paths in the makeup system after a LOCA. FPC refers reactor to RCP seal injection and normal makeup and refers to a (As a contingency Gilbert Associates report that concludes adequate HPI flow action, if power is lost to is achieved without these lines isolated. NRC letter to MUV-27 (normal licensees with B&W designed systems (Generic Letter 86-makeup) and MUV-18

05) dated 5/29/86 states the cooling water sources (RCP seal injection),

supperting the RCP with the potential of being isolated are transfer to an energized seal injection, seal bleedoff, component cooling water to seal bus and close valves) line coolers, and component cooling water to RCP motors and oil coolers. The need to isolate RCP Seal Injection was discovered in 1995 to be necessary due to discovery that operators relied on non-Reg Guide 1.97 instrumentation to measure this flow when determining HPI pump runout flow limits (see LER 95-026). Seal injection isolation was also determined necessary during Refuel 10 in 1996 upon discovery that 'vorst case instrument error may result in inadequate HPI flow (see LER 96-006).

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 13 Table 3A Operator Actions Less Than 20 Minutes O

Operator Action Assumed Time Basis Reference A

in Design Analysis 5

Ensure adequate HPI Yes

< 20 Required FPC letter to NRC dated 10/27/89 states HPI must be successfully flow (USQ6) (isolate a minutes only for balanced to support SBLOCA mitigation as described in various broken injection line break in B&W topical reports accepted by NRC. Subsequent FPC letter using new isolation IIPI line dated 10/31/89 states that mitigation strategy employed from the criteria) late 1970's through reviews done in response to NUREG 0737 relied on balancing HPI flow for breaks in HPI injection lines.

'Ihese letters relate to LER 89-037, issued in November 1989 reponing a design basis condition in which instrumentation used for balancing HPI flow was inadequate. NRC letter dated 12/20/89 confirmed verbal concurrence to resume power operation with the HPI instrumentation problems. One condition was operator action for HPI flow balancing. NRC letter dated 2/17/95 from Gary Holahan to Ed Jacks (BWOG Operator Support Committee) states staff has completed its review of BWOG response to NUREG 0737 Item I.C.1 regarding EOP Guideliaes and is finalizing an SER on the topic. Balancing HPI flows was a part of the ATOG/TBD guidelines incorporated into FPC procedures. FPC issued LER 96-007 on 3/15/96 to report another design basis condition involving HPI flow instmmentation. The flow deficiencies described therein were addressed by revised SBLOCA analyses provided by Framatome Technologies in April 1996 which required isolation of the affected HPI line versus balancing. Most recent FTl analyses have provided new isolation criteria.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 14 Table 3A Onerator Actions Less Than 20 Minutes i

O Operator Action Assund Time Basis Reference i

A in Design Analysis 6

Ensure adequate EFW Yes

< 20 Raise OTSG B&W (Taylor) letter to NRC (Baer) dated 5/1/78 provides flow (USQ6) minutes levels to topical report 10104, "B&W's ECCS Evaluation Model,"

ISCM which notes operator action is necessary during early stages (EFIC was initiated in setpoint of the accident te mitigate consequences and meet 10 CFR OA2; therefore, ensuring (95%)

50.46. Auxiliary feedwater is assumed to be available.

EFW flow is a NRC letter to FPC dated 7/6/79 provides a SER for actions confirmation step only) taken in response to Commission Order dated 5/16/79. The SER states that a generic review of B&W analyses entitled i

This step manually raises

" Evaluation of Transient Behavior and Small RCS Breaks in OTSG levels to the the 177 Fuel Assembly Plant" resulted in a principle finding Inadequate Subcooling that reconfirms SBLOCA analyses demonstrate a Margin, ISCM level combination of heat removal by the steam generator and the HPI system combined with opera;ar action to ensure adequate core cooling. These results are applicable to CR-3 considering the ability to manually start the redundant EFW pumps and HPI pumps from the control room, assuming failure of automatic EFW actuation. NRC letter to FPC dated 8/30/85 provides a SER for NUREG 0737 Item II.K.3.30, "S3LOCA Methods." Section IILS.a of the SER states "the timing of operator action to raise the secondary system water level to 95% was found not to be critical."

P i

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 15 Table 3B New Operator Actions After 20 Minutes I

OA Operator Action Failure Cycle 11 Only Basis Scenario 7

If 'B' DC power is lost, crosstie EFP-2 to A train (EFV-12)

LOBB Yes EFP-1 can only provide flow for a specific time period, then EFP-AND 2 must be aligned.

Secure EFP-1 8

If only EFP-2 is supplying feedwater to the OTSG, then LOBA No Defense in depth for possible there will be no operator initiated RCS cooldown whether or LOBB No decoupling of the OTSGs. Use not offsite power is available. If other sources of feedwater of FWP-7 provides additional are available, cooldown may be initiated.

resources available to operators during a LOOP (Mitigation strategy includes operation of diesel backed FWP-7 as a Defense in Depth action) 9 If EFP-2 is not operating when in a LOOP condition with EFP-2 Yes If EFP-2 is not available, steps inadequate subcooling, limit cooldown prior to the EFP-must be taken to ensure EFP-1 1/LPI Interlock operates as long as needed.

10 Periodically re-evaluate llPI line break criteria on RCS LOBA No Required for specific IIPI line repressurization LOBB No pinch areas to ensure a broken EFP-2 No line will be isolated if warranted

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 16 Table 3B New Operator Actions After 20 Minutes OA Operator Action Failure Cycle 11 Only Basis Scenario 11 Manage operation of EFP-2 by closing ASV-5 and ASV-204 LOBA No For a LOBA, or a LOBB (to on low OTSG pressure (Cycle EFW).

LOBB Yes manage EDG Icad), EFP-1 would be secured and EFW flow would rely on EFW-2. EFP-2 would be cycled due to operational limitations on low OTSG pressures.

12 Put EFIC in manual permissive LOBB Yes Required to prevent cycl;ng of the limited duty motors on the EFW AND block valves. This action may be included in the EOPs for both Close EFW block valves trains of EFW.

13 Manage EDG load in order to extend EFP-1 operation by -

EFP-2 Yes Defense in Depth action for Shutdown SWP-1A & RWP-2A after verifying postulated single failure of the redundant pumps are operating and placing switches loss of EFP-2. These actions in Pull-to-Lock to prevent reactuation of pumps (EDG extend the time EFP-1 is loading) available for OTSG cooling.

Place EFP-1 Trip Defeat Switch in defeat position to prevent automatic trip of EFP-1 on RCS pressure of 500 psig

}

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 17 Table 4 Summary of Planned FSAR Changes Section Section Title Summary of Planned Changes 6.0 Engineered (1) Add a discussion that certain size small break LOCAs require EFW to maintain primary to Safeguards secondary cooling via the OTSG until reactor core decay heat can be removed solely by IIPI and flow out the break.

6.1 Emergency Core (1) Add discussion that EFW is needed for some sizes of SBLOCA and clarify current discussion Cooling System about break sizes. (2) Update discussions of RCS and HPI line breaks based on FPC's Safety Assessment. (3) In 6.1.2.1.1 add item d. to state that the normal makeup supply valve (MUV-27) is closed to facilitate accurate HPI flow indication for HPI flow. Also, add reference to the normal makeup supply valve (MUV-27) as being supplied by either of two channels of ES electrical buses.

(4) Table 5-9, item no. 9, add "(A/B)" to MUV-27 and add function that diverse containment isolation signal isolates normal makeup from HPI line.

7.1 Protection In 7.1.3.2.3, delete reference to " Flush Water Pumps."

Systems

_ 7.2 Control Systems (1) In 7.2.4.2, reflect that trip module in "A" cabinet starts turbine-driven EFP-2. Add discussion that starting of both EFW pumps on "A" EFIC actuation is necessary to assure that EFP-2 will operate with failure of "B" DC system and loss of offsite power. EFP-2 is relied upon to share the EFW load with EFP-1 to decrease load on EDG-1 A. Also, EFP-2 operation is necessary in SBLOCA with LOOP and loss of Battery "B". (2) Revise 7.2.4.1, EFIC " Design Bases" for flow rate control of ERV from 600 gpm to 550 gpm, and delete discussion of flow rate control of EFW when OTSG pressure is less than 600 psig. (3) Add discussion to 7.2.4.2 that cavitating venturis are to choke flow at approx. 750 gpm to avoid high flow problems that could occur if control valves fail open. (4) In Table 7-8A, add EFP-2, with I channel. (5) In 7.2.4.1, in number 3. add: " Manual control may be required under certain circumstances such as low decay heat rates and the transition between low and high range level instruments."

7.3 Instrumentation (1) From section 7.3.5 ?, in Table 7-12, add EFP-2 Flow for Cross-tie, Type and Category: D, 2.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 18 Table 4 Summary of Planned FSAR Changes Section Section Title Summary of Planned Changes 8.2 Electrical System (1) In 8.2.3.1.3 revise 2000 hour0.0231 days 0.556 hours 0.00331 weeks 7.61e-4 months rating from 2851-3000 kW to 2851-3200 kW and revise 200 hour0.00231 days 0.0556 hours 3.306878e-4 weeks 7.61e-5 months Design rating from 3001-3250 kW to 3201-3400 kW. Change the stated seven-day rQM volume of storage tanks from " maximum continuous rating of 2850 kW" to "200 hour0.00231 days 0.0556 hours 3.306878e-4 weeks 7.61e-5 months rating of 3400 kW".

Change the stated required volume of day tanks. Add discussion that EFP-2 will be automatically started by opening of ASV-204 and share the EFW load thereby reducing EDG loads, and that this is needed to keep EDG load within 2000 hour0.0231 days 0.556 hours 0.00331 weeks 7.61e-4 months rating. (2) Update discussion of loads based on EDG load testing; (3) In Tables 8-1 and 8-2 delete Flush Water Pump DOP-2A as an auto connected load on EDG-1 A. (4) Section 8.2.2.4, revise last sentence of first paragraph as follows: The other bus section, designated Reactor Auxiliary Bus 3, provides power for the Auxiliary Feedwater Pump (FWP-7) through transformer MTTR-5 or a Standby Diesel Generator (MTDG-1).

9.5 Cooling Water (1) PRELIMINARY: In 9.5.2.1.1 add discussion of the new " Pull-To-Lock" switches which replace Systems the existing control switches for Nuclear Service and Decay Heat Seawater System (RW) pumps RWP-2A and RWP-2B, and Nuclear Services Closed Cycle Cooling System (SW) pumps SWP-1 A and SWP-1B, and that this will allow the operator to block restart of SWP-1A and RWP-2A as part of EDG load management required as part of SBLOCA mitigation.

i U.S. Nuclear Regulatory Commission 3F0697-10 Attaciunent F Page 19 Table 4 Summary of Planned FSAR Changes Section Section Title Summary of Planned Changes 10.5 Emergency (1) In 10.5.1 add discussion of cavitating venturis EF-62-FO and EF-63-FO as being designed to Feedwater System provide 750 gpm flow. (2) In 10.5.2, discuss how recirculation lines provide EFW pump protection against low flow damage and cavitating venturis provide protection against high flow damage. Add discussion to 10.5.2 that the turbine driven EFP-2 is independent of AC power and starts by opening of either ASV-5 or ASV-204 when activated by EFIC. Also, add discussion of dependencies: (a)

Starting of both EFW pumps on "A" train EFIC actuation is necessary to assure that EFP-2 will operate on failure of ES "B" DC system or EDG-1B coincident with loss of power, EFP-2 will be relied upon to share the EFW load with EFP-1 to decrease the load on EDG-1A. (b) Operation of EFP-2 is necessary in a small break LOCA with a LOOP and loss of "B" battery as the RCS cooling source to reduce RCS pressure from 500 psig to 200 psig where LPI cooling is effective. (3) Add section 10.5.2.7, "EFW Cavitating Venturis' to discuss details about the cavitating venturis. The venturis, installed in the discharge from each pump, are designed to choke EFW flow into the OTSGs at 750 gpm and protect against high flow damage due to pump runout and low NPSII. The venturis are a passive protection feature in the event any control valves fail open. The venturis protect the OTSG tubes from flow induced vibration problems. Add discussion that EF-62-FT across cavitating venturi EF-62-FO provides EFP-2 flow measurement for EFV-12 cross-tie operation. (4) In 10.5.2.3, clarify that both ASV-5 and ASV-204 open upon actuation from EFIC.

(5) In section 10.5.3, add ASV-204 as receiving power from Battery 'A'. Add discussion that EFP-2 will supply flow through cross-tie and utilize 'A' train instruments an control valves to monitor and control flow to the OTSGs. (6) Add new section, "EFW System Cross-Tie", discussing the cross connect piping between discharge of EFP-1 and EFP-2, the valves in that piping, and conditions when cross-tie operation is required. (7) Table 10-1, add EFV-12, as a 6-inch Atwood & Morril gate valve.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 20 Table 4 Summary of Planned FSAR Changes Section Section Title Summary of Planned Changes 10.6 Auxiliary (1) In section 10.6.2, add to discussien of power sources for AFW pump motor that on a LOOP, the Feedwater motor is fed from MTDG-1, a non-safety related diesel generator, not automatically started but under manual control from the MCB. (2) In section 10.6.5 add that a non-safety Standby Diesel Generator (MTDG-1) is capable to be lined up to the 4160V Reactor Auxiliary Bus 3 which will enable FWP-7 to operate in a LOOP.

14.2 Standby Section 14.2.2.5, " Loss of Coolant Accident", revise and a6d additional details to the discussion of Safeguards SBLOCA based on revised analyses performed by Framatome for different sizes and locations of Analysis breaks.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 21 Table 5 Related LERs LER LER #

Title Description Reference Resolution (Event Date) 1

% -020 Unreviewed Safety An USQ was identified _which was associated

USQ1, Modifications (9/10/%)

Questions Concerning with MAR %-04-12-01 (ASV-204 EFIC Auto

USQ2, MODI, MOD 5, MOD 6, Diesel Generator Loading Open Removal). This modification

USQ3, MOD 7, MOD 8, MOD 9, Caused by Interpretation contributed to increased in loading of the A USQ5 MOD 10 of Regulatory EDG. It was discovered that, contrary to Requirements other than information contained in the TS, EDG '

TS/ Ram Chances.

Prescribed.

Ioading analyses indicated the worst case Parts 1, 2, 3 maximum automatically connected accident load at one-minute (3100 kW) would be.

exceeded, maximum EDG design rating of 3500 kW loading would be exceeded for up to 3 seconds during three of the six EDG block I

loading sequences, and the single largest rejected load was greater than previously identified.

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F Page 22 Table 5 Related LERs LER LER #

Title Description Reference Resolution (Event Date) 2 96-024 Plant Modification During a review of a EDG loading

USQ4, Modifications (10/11/96)

Creates Unanalyzed calculation, it was determined that the USQ5 MODI, MOD 2, MOD 3, Condition Regarding calculation assumed EFP-2 was running when MOD 4 Emergency Feedwater EFP-1 received an ES automatic trip signal at Availability a RCS pressure of 500 psig for LPI actuation.

TS/ Bases Changes MAR 96-12-04-01 (ASV-204 EFIC Auto Part 1 Open Removal) had removed the automatic start from EFIC A system to prevent runout and NPSH concerns with EFP-2 during certain accident conditions when its flow control valves would fail open. EFP-2 would remain available but require operator action to crosstie EFP-2 flow to the EFP-1 flow path control valves which would have control power available during a loss of DC power train B scenario.

U.S. Nuclear Regulatory Commission i

3F0697-10 i

Attachment F Page 23 Table 5

{

Related LERs LER LER #

Title Description Reference Resolution 1

(Event Date) 3 97-001 Ineffective Change FPC had not explicitly reported a condition USQ4 Modifications l

(1/28/97)

Management Results in that existed prior to May 1996 involving MODI, MOD 2, MOD 3, Unrecognized NPSH inadequate NPSH affecting EFP-2. In MOD 4 i

Issue Affecting addition to an EDG load management Emergency Feedwater concern, the postulated loss of B DC power TS/ Bases Chanoes Availability single failure coincident with a SBLOCA and Part 1 LOOP, could have resulted in two situations in which EFW may not have been available to perform its intended functions. These include a design feature which trips EFP-1 at a RCS pressure of 500 psig, and a point in thne at which EFP-1 would need to be secured in order to load the LPI pump onto the EDG in order to provide adequate NPSli to the HPI.

i t

i i

L i

f t

U.S. Nuclear Regulatory Commission 3F0697-10 Attachment F t

Page 24 Table 5 Related LERs LER LER #

Title Description Reference Resolution (Event Date) 4 97-005 Unanalyzed Condition A team studying EDG capacity and EFW Modifications (2/14/97)

Regarding Small Break system dependency issues discovered an MODI, MOD 2, MOD 3, LOCA Mitigation Caused unanalyzed condition involving a SBLOCA in MOD 4, MOD 5, MOD 6, by Misunderstanding the reactor coolant cold leg discharge piping MOD 7, MOD 8, MOD 9, Reliance on Emergency line coincident with a LOOP and single MOD 10 i

Feedwater failure of EFP-2. For a certain range of l

small breaks, the RCS may repressurize. For TS/ Bases Changes an unisolable cold leg pump discharge line Parts 1,2

[

break or an isolable broken HPI line that may not be identified, if EFW is lost early in the transient, there may be inadequate HPI flow to the core, even with two HPI pumps t

operating. With a combination ofinsufficient HPI injection and no EFW,10 CFR 50.46 requirements may not be met.

[

f i

i i

i

-a

4 FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT 3 i

DOCKET NUMBER 50-302/ LICENSE NUMBER DPR-72 ATTACHMENT G LIST OF ACRONYMS AND ABBREVIATIONS USED

ATTACHMENT G LIST OF ACRONYMS AND ABBREVIATIONS USED ADV

- Atmospheric Dump Valves AFW ---

Auxiliary Feedwater ATOG Abnormal Transient Operating Guidelines B&W-Babcock and Wilcox BWOG

-Babcock and Wilcox Owners Group BWST-

- Borated Water Storage Tank CLPD Cold Leg Pump Discharge CFR -----

Code of Federal Regulations CR Crystal River Unit 3 CREVS -

Control Room Emergency Ventilation System DC

-Decay Heat Closed Cycle Cooling Water System DC Electrical Direct Current Electrical DHP---

Decay Heat Pump DHR Decay Heat Removal ECCS Emergency Core Cooling System EDG

-Emergency Diesel Generator EEI -----

Escalation Enforcement Item EFIC Emergency Feedwater Initiation and Control System EFP

-Emergency Feedwater Pump EFV --

Emergency Feedwater Valve EFW -

-Emergency Feedwater Emergency Operating Procedure EOP ESAS Engineered Safeguards Actuation System F---

Fahrenheit FPC

-Florida Power Corporation FSAR Final Safety Analysis Report FTI Framatome Technologies Incorporated (formerly B&W)

FWP Auxiliary Feedwater Pump gal gallon gallons per minute gpm HPI

High Pressure Injection kW kilowatts LCO ---

Limiting Condition for Operation LER-

Licensee Event Report LOBA Loss of Battery A LOBB Loss of Battery B LOCA Loss of Coolant Accident LOOP-Loss of Offsite Power LPI -

Low Pressure Injection MCAP Management Corrective Action Plan MSSV Main Steam Safety Valve 1

I U.S. Nuclear Regulatory Commission i

3F%97-10 l

Attachment G Page 2 MUP Makeup Pump.

MWt

- Megawatt (Thermal)

NPSH -

Net Positive Suction Head i

NRC Nuclear Regulatory Commission OTSG ---

-Once Through Steam Generator i

PCT Peak Clad Temperature psi -

--pour.ds per square inch psig pounds per square inch gauge l

PSV

Pressurizer Safety Valve i

RBCU

-Reactor Building Cooling Units j

- RCP -

Reactor Coolant Pump RCS Reactor Coolant System l

RW

---Nuclear Services Seawater System i

RWP

---Nuclear Services Seawater Pump j

l SBLOCA Small Break Loss of Coolant Accident SER Safety Evaluation Report SR-Surveillance Requirement i

SW Nuclear Services Closed Cycle Cooling System SWP Nuclear Services Closed Cycle Cooling Pump l

TBD-Technical Bases Document -

TMI Three Mile Island TS --

Technical Specification TSCRN ------

Technical Specification Change Request Notice l.

USQ-

-Unreviewed Safety Question p

I l

i i

i l

l

,

ML20148K796

Text

1.0 Background

1.0 BACKGROUND

; -

Subject:

Reference:

Dear Mr. Sullivan:

Response

Response

Response

Response

Response

Response

Response

Response

Response

Response

Response

Navigation menu

Search