ML20213D965

Small Break LOCA Analysis for B&W 177FA Raised Loop Plant in Response to NUREG-0737,Item II.K.3.31

ML20213D965
Person / Time
Site:	Davis Besse
Issue date:	10/31/1986
From:	BABCOCK & WILCOX CO.
To:
Shared Package
ML20213D890	List: ML20213D886 ML20213D965
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.K.3.31, TASK-TM BAW-1981, NUDOCS 8611120356
Download: ML20213D965 (56)

Text

..

BAW-1981 l

77-1165793 00 October 1986 I

I I

g Small Break Loss of Coolant Accident I

Analysis for The B&W 177FA Raised I

Loop Plant in Response to I

NUREG-0737, item II.K.3.31 I

I i I

)l Babcock &Wilcox l

jeA"!888 ggaagg Pon

i BAW-1981 October 1986 8

I SMALL BREAK LOSS-OF-COOLANT ACCIDENT I

ANALYSIS FOR THE B&W 177-FA RAISED-LOOP PLANT IN-RESPONSE TO NUREG-0737, ITEM II.K.3.31 I

I Prepared For The B&W Owners Group

.I Toledo Edison Company I

.l THE BABC0CK & WILCOX COMPANY Nuclear Power Division P. O. Box 10935 Lynchburg, Virginia 24506-0935 Reviewed By:

h[

SupefvisoryfErffineer Approved By:

8Nt I

Projet Manager i

Babcock &WHcom a McDermott company

I Babcock & Wilcox Nuclear Power Division Lynchburg, Virginia Report BAW-1981 October 1986 I

Small Break loss-of-Coolant Accident Analysis For The Babcock & Wilcox 177-Fuel Assembly (FA) Raised-Loop Plant in Response to NUREG-0737. Item II.K.3.31 M. A. Rinckel, G. E. Anderson, J. R. Paljug Key Words: NUREG-0737. II K.3.31. SBLOCA g

ABSTRACT Small break loss of-coolant accident (SBLOCA) analyses have been performed g

utilizing the revised evaluation model (EM) created in response to NUREG-5 0737, Itum II.K.3.30.

The evaluation presented herein is in specific response to Item II.K.3.31 of NUREG-0737.

This report describes S3LOCA transient behavior, compares revised EM results with previous results, and determines the applicable conservatism of previous SBLOCA spectrum analyses.

The evaluation concludes that the SBLOCA spectrum, inclusive of previous and revised EM results, satisfies the requirements of NUREG-0737, Item II.K.3.31, and therefore confirms that the Babcock & Wilcox 177-fuel assembly (FA) raised-loop plant can be maintained within the limits of 10 CFR 50.46 should that plant experience an SBLOCA.

I I

I I

I

- 'i -

a ucoermore company

I I E

'I CONTENTS I

Pa;s 1.

INTRODUCTION.........................

1-1 2.

BACKGROUNO..........................

2-1 3.

SPECIFIC REQUIREMENTS OF NUREG-0737, SECTION II.K.3.31 3-1 4.

SMALL BREAK LOSS-OF-COOLANT ACCIDENTS.............

4-1 4.1.

Category 1: SBLOCAs Too Small to Interrupt Natural Circulation...................

4-2 I

4.2.

Category 2: SBLOCAs That May Allow The RCS to Repressurize in a Saturated Condition..........

4-3 4.3.

Category 3: SBLOCAs That Allow RCS Pressure to Stabilize at Approximately SG Secondary Pressure.........

4-3 I

4.4.

Category 4: SBLOCAs Large Enough to Depressurize The RCS Sufficiently to Permit Low-Pressure Injection..

4-4 5.

SBLOCA EVALUATION MODELS....................

5-1 5.1.

Summary of The SBLOCA Evaluation Models.........

5-1 5.2.

Model and Input Changes.................

5-1 6.

SBLOCA SPECTRUM ANALYSIS WITH THE REVISED EM.........

6-1 6.1.

Introduction......................

6-1 I

6.2.

Justification For The SBLOCA Cases Selected For Reanalysis 2.....................

6-2 6.2.1.

0.04-ft Break.................

6-2 E

6.2.2.

0.01-ft Break.................

6-2

5 6.3.

Justification For Excluding Categories 1, 3 and 4 l

Breaks From The Reanalysis Spectrum...........

6-3 6.3.1.

Category 1 6-3 I

6.3.2.

Category 3 6-4 6.3.3.

Category 4...................

6-5 7.

RESULTS OF THE SBLOCA REANALYSIS 7-1 7.1.

Introduction......................

7-1 l

7.2. 0.04-ft2 Break.....................

7-1 7.2.1.

Results Using The Early EM...........

7-1 i

7.2.2.

Results Using The Revised EM..........

7-2 7.2.3.

Comparison of The Results From The Analyses Using The Early and Revised ems 7-4

- iii -

I 4 McDermott company I

l CONTENTS (Cont'd)

Page 8.

CONCLUSIONS...........................

8-1 9.

REFERENCES......'.....................

9-1 I

list of Tables Table 5-1.

Comparison of Key Input Parameters For The Early, Intermediate and Revi sed ems........................

5-3 5-2.

Comparison of AFW Flow Rates Vs Secondary SG Pressure For The Early, Intermediate, and Revised ems.........

5-4 5-3.

Comparison of High Pressure Injection (HPI) Flow Rates Vs RCS Pressure For the Early, Intermediate and Revised ems 5-5 7-1.

Comparison of The Resu ts of The Early and Revised Evaluation Models For The 0.04 ft Break.................

7-7 I

List of Fiaures Figure Page 4-1.

Characteristic RCS Pressure Response for SBLOCA Categories 1 through 4..........................

4-6 4-2.

Transient Core Mixture Height Vs SBLOCA Break Size 4-7 4-3.

Cladding Temperature Response - SBLOCAs That Produce Partial Core Uncovering...............

4-8 4-4.

Comparison of HPI Flow Rates Vs RCS Pressure.........

4-9 5-1.

CRAFT 2' Noding Diagram For Small Breaks - Early Model.....

5-6 l

5-2.

CRAFT 2 Noding Diagram For Small Breaks - Intermediate Model 5-7 5-3.

CRAFT 2 Noding Diagram For Sma}1 Breaks - Revised Model....

5-8 7-1.

RCS and SG Pressures, 0.04-Ft Brea, Early Model 7-8 7-2.

Inner Vessel Liquid Volyme, 0.04-ft Break, Early Model 7-9 7-3.

Leak Flow Rate, 0.04-ft' Break, Early Model 7-10 RCS and SG Pressure, 0.04-ftz SG Secondary Pressure, 0.04-ftgreak, Revised Model......

7-11 7-4.

Break, Revised Model.....

7-12 7-5.

l l

7-6.

Inner Vessel Liquid Volyme, 0.04-ft' Break, Revised Model 7-13 7-7.

Leak Flow Rate, 0.04-ft' Break, Revised Model 7-14 I

I

- iv -

a uconmoit como.ny i

I CONTENTS (Cont'd)

Figure Page 7-8.

RCS Pressures, 0.04-ft2 Break, Comparison of Results of Early and Revised Models...... $...............

7-15 7-9.

SG Secondary Pressures, 0.04-ft Break, Comparison of Results of Early 'and Revised Models

2.............

7-16 7-10.

Inner Vessel Liquid Volumes, 0.04-ft Break, Comparison of Results of Early and Revised Models.............

7-17

'I 7-11.

Leak Flows, 0.04-ftZ Break, Com and Revised Models...... parison of Results of Early

2...........

7-18 7-12.

Integrated Flows to Containment, 0.04-ft Break, Comparison of Results of Early and Revised Models............

7-19

I I

,.I i

I

'I iI E

l

-v-a woermore company

I I

I 1.

INTRODUCTION E

In response to the requirements of NUREG-0737,Section II.K.3.31, The Babcock

& Wilcox Owners Group (B&WOG) has performed design-basis small break loss-of-g E

coolant accident (SBLOCA) analyses for the 177-Fuel Assembly (FA) raised-loop plant design.

These analyses repeat certain studies that have been pre-l viously submitted; these latest studies, however, employ updated modeling techniques, inputs and assumptions that are discussed specifically or by reference in this document.

The results of these analyses, presented and described herein, confirm the findings of previous studies:

The Babcock &

Wilcox (B&W) designed raised-loop 177-FA plants can be maintained within the limits of 10 CFR 50.46 should an SBLOCA occur.

I I

~

I I

I I

t I

I I

1 1

1-1 mggg a uconmott company

I I

I 2.

BACKGROUND As a result of the March 28, 1979 accident at Three Mile Island Unit 2 (TMI-2), the Bulletins and Orders Task Force was formed within the Nuclear Regula-tory Commission (NRC) office of Nuclear Reactor Regulation.

The Task Force I

was charged, in part, with reviewing the analytical predictions of feedwater transients and SBLOCAs to ensure the continued safety of all operating reactors, and with determining the acceptability of operator emergency guidelines.

As a result of their reviews, the Task Force concluded that, l

while there were no apparent safety concerns, additional system verification of the SBLOCA model (as required by II.4 of Appendix K to 10 CFR 50) was needed in certain areas. These improvements and concerns, as they applied to each light water reactor (LWR) vendor's model, were documented in the various Task Force reports for each vendor.

The review of the B&W SBLOCA model was I

documented in NUREG-0565, " Generic Evaluation of Small Break Loss-of-Coolant Accident Behavior in Babcock.& Wilcox Designed 177-FA Operating Plants,"

January'1980. The review of the reactor coolant pump model was documented in NUREG-0623, " Generic Assessment of Delayed Reactor Coolant Pump Trip During Small Break Loss-of-Coolant Accidents in Pressurized Water Reactors,"

November 1979.

On October 31, 1980, the NRC issued NUREG-0737, "Clarifica-tion of TMI Action Plan Requirements."

Included in NUREG-0737 is the I

requirement for an industry review of NUREG-0565 and -0623 and the develop-ment of a program that addresses the NRC concerns therein.

The Sr:all Break

,gl3 LOCA Methods Program was developed by the B&WOG to address the requirements of NUREG-0737,Section II.K.3.30.

The results of the Small Break LOCA Methods Program have been documented in References 1 and 2.

These references address the revision to SBLOCA codes and models in response to the issues identified in NUREG-0565 and -0737.

The NRC has reviewed and approved the results of the SBLOCA Methods Program with the issuance of the May 5,1985 Safety Evaluation Report (SER) for the B&WOG t

i 2-1 Bakock &WWilcom a ucoermore company

lIl 1!

SBLOCA Evaluation Model.

This SER is presented on pages v - 1111 of Ref-erence 1.

The preceding documentation completed the requirements of NUREG-0737,Section II.K.3.30.

However, it then became necessary to address the requirements of NUREG-0737,Section II.K.3.31.

As is discussed in the following sections, a program for compliance with II.K.3.31 was formulated and carried out by the B&WOG.

I I

I I

I' I

I I

I I

I' I

I I

2-2 Babcock &Wilcom a ucoermore comoany

I I

I l

I 3.

SPECIFIC REQUIREMENTS OF NUREG-0737, SECTION II.K.3.31 E

Section II.K.3.31, eatitled:" Plant-Specific Calculations to Show Compliance with 10 CFR Part 50.46," requires that plant-specific SBLOCA analyses using models that meet the requirements of NUREG-0737 be performed to show g

compliance with 10 CFR 50.46 and be submitted for NRC approval.

The 5

requirements are applicable to all operating reactors and applicants for an operating license.

Additional information concerning the requirements of Section II.K.3.31 was provided by the NRC letter entitled " Clarification of TMI Action Plan Item II.K.3.31 (Generic Letter No. 83-85)," D. G.

Eisenhut, November 2, 1983.

This letter states "...the requirements of II.K.3.31 can be satisfied by each l

licensee by submittal of a plant-specific analysis that demonstrates that current SBLOCA analyses using previously approved evaluation models are more limiting than analyses using the revised (II.K.3.30) models.

This bounding demonstration can be done on a generic basis through the owners groups or vendors and submitted individually by each licensee."

Furthermore, the aforementioned SER (Reference 1) states, "It is the inten-I tion of the B&W Owners Group (B&WOG) to provide generic analyses, by plant configuration, in response to NUREG-0737 II.K.3.31 which will demonstrate that the current FSAR SBLOCA results are conservative.

This will be accom-plished by selecting a limited break spectrum for the evaluation.

The break spectrum will be selected to exercise the emergency core cooling system (ECCS) and span the previously identified limiting break size."

In view of the requirements and clarifying documentation, the intent of the B&W0G through the analyses presented in this report is to demonstrate compli-ance with II.K.3.31 by showing conformance with 10 CFR 50.46.

Conformance I

will be demonstrated through a qualitative assessment of the SBLOCA spectrum to determine the critical break sizes and then a quantitive analytical evaluation of these critical break sizes.

For the purposes of this 3-1 gggg

I discussion, a critical break size is one that produces a fuel cladding temperature in excess of primary coolant saturation temperatures or provides evaluation of phenomena indicat've of the transition from one break size i

category to another.

Discussions of break size categories and of the selection of break sizes to be analyzed with the revised evaluation model (EM) are contained in the following sections of this report.

1 I

I I

I I

I t

i I

I I

I I

I 3-2 Bakock&Micom a ucoermore company

{

?

~

, q.

w sN

,, ~.

2' s

4.

SMALL BREAK LOSS-0F-COOLANT ACCIDENTS Based on the results of design-basis analyses of B&W's 177-FA plants, a loss-of-coolant accident is defined to be "small" when its cross-sectional area is p.

0.5 ft2 or less.

Design-basis SBLOCA assumptions are discussed in section k

6.1 of this report.

An SBLOCA involves a relatively slow system depressurization.

Flow conditions within the reactor coolant system change F

L gradually and sacethly.

Temperature and pressure gradients between regions tend to be small.

The lack of agitation allows partial phase separation of

[.

steam and water and, in some situations, countercurrent. flow.

Rather than the distinct blowdown and reflood phases associated with large breaks, small

[

breaks have a smooth transition from a period of relatively high core flow to one~ of relatively quiescent conditions.

During the early phase, heat transfer in 'the core is flow-controlled and is adequate to cool the fuel cladding.< La'ter, during the quiescent period, a two-phase froth level can g

develop in the core region of the reactor vessel.

To ensure adequate core L

cooling, a two-phase mixture level must be maintained at or near the top of thecoralsaminimum.

In this manner, the generated core decay heat can be

[

removed from the fuel rods by pool boiling or, if the core is slightly uncovered, by convection to superheated steam. The ECCS has been designed to p

, provide. the necessary_ fluid makeup to the reactor coolant system (RCS) to ensure at least this adequate level of core decay heat removal.

p Design-basis SBLOCA analyses have shown that different sizes of SBLOCAs exhibit various characteristic RCS responses.

For very small breaks (less 2

than approximately C.003 ft ), natural circulation will be maintained and the j

k pH eary system can Es.kept in, or returned to, a subcooled state.

For larger sma1Ibr'eaks(approxi_mately0.02ft2 and larger), the circulation flow phase will en'd soon after the RC pumps are tripped.

After the end of the " forced flow" circulation portion of a small break, the

(

reactor coolant " settles out."

That is, the water falls by gravity and I

~

\\

E l

~

4-1 mweg

.m a McDermott comparty

.s

I collects in the lower regions of the RCS, and the steam separates from the l

liquid phase and collects in the high points of the RCS.

A " boiling pot" water level will exist that will vary depending on break size and location, g

primary-to-secondary heat transfer, ECCS performance (which depends largely on RCS pressure) and decay heat levels.

These variables can cause the response characteristics of the RCS to change in different ways after the reactor coolant system " settles" into the

" boiling pot."

However, the following four main categories of SBLOCAs have been designated:

1.

SBLOCAs too small to interrupt natural circulation.

2.

SBLOCAs that may allow the RCS to repressurize in a saturated condition.

3.

SBLOCAs that allow RCS pressure to stabilize at approximately secondary side pressure.

4.

SBLOCAs large enough to depressurize the RCS sufficiently to permit low-pressure injection (LPI).

4.1.

Category 1: SBLOCAs Too Small to 5

Interruot Natural Circulation 5

Typically for Category 1 breaks (sizes smaller than approximately 0.003 ft ),

2 the make-up and high-pressure injection (HPI) systems can compensate for the break flow and maintain the primary coolant loops essentially full of liquid so that circulation is not interrupted.

Note that the approximate size of this category is less than the same category for the 177-LL plant (.005 ft ),

2 This is due to the 177-RL plant having a separate make-up system of slightly smaller capacity than the 177-LL plant HPI capability in the make-up mode.

An example of the expected RCS pressure response for breaks in this category is shown as curve 1 of Figure 4-1.

While RCS pressures may tend to remain high, the operator has equipment available (reactor coolant pumps, pressuri-zer sprays, makeup system, and steam generators) to effect a near-normal cooldown and depressurization of the RCS EE I

4-2 Babcock & WHcox

it i

4.2.

Category 2: SBLOCAs That May Allow The RCS to Repressurize in a Saturated Condition g

Category 2 Nreaks, (sizes between approximately 0.003 ft2 and 0.02 ft )

2 initially cause the RCS to depressurize and become saturated.

Later, however, these breaks can cause the upper hot leg elbows to void sufficiently I

to interrupt circulation.

This loss of circulation leads to a loss of

~

primary-to-secondary steam generator (SG) heat transfer.

Without SG heat I

transfer, the energy additicn from the core and the mass addition from the ECCS may exceed the ability of breaks of these sizes to discharge mass and energy; as a result, the RCS may repressurize.

As the RCS continues to lose inventory, but while the RCS mixture level is l

still above the top of the core, a condensing surface will be exposed inside the SG tubes. The beginning of steam condensation will establish the boiler-condenser (b-c) mode of heat removal.

This heat removal mode will terminate any pressure increase and will facilitate the control of RCS pressure at a value sufficiently low to ensure adequate HPI flow for core cooling.

I The transient pressure behavior for a Category 2 break is illustrated as Curve 2 of Figure 4-1.

This transient is taken from Reference 3 and includes I

sattuted repressurization followed by depressurization as the RCS evolves into the b-c mode of cooling.

4.3.

Category 3: SBLOCAs That Allow RCS Pressure to Stabilize at Acoroximately SG Secondary Pressure 2

2 and 0.10 ft ) result Category 3 breaks (sizes between approximately 0.02 ft in RCS depressurization until the primary and secondary systems reach thermal equilibrium. At this pbint, SG heat removal is essentially lost.

A continu-ous-RCS depressurization is maintained, but the rate of this depressurization can be relatively slow and dependent on the rates of the following events:

1.

Decrease in core decay heat generation.

2.

Cooldown of fluid and metal components in the primary and secondary systems.

3.

Release of mass and energy from the break.

4.

ECCS injection.

I 4-3 Babcock &WHcox a McDermott comparty

I An example of the possible RCS pressure response for breaks in this category is shown as Curve 3 of Figure 4-1.

SBLOCA studies (Reference 4) have shown that partial core uncovering, accompanied by increases in fuel cladding temperature may occur for some Category 3 breaks.' Figures 4-2 and 4-3 show the results of the analyses of a series of Category 3 breaks. for B&W's 177-FA lowered-loop (177-LL) plants (Reference 4).

In addition to showing the effects of core uncovering and core recovery, these 177-LL studies also show that if core flooding tank (CFT) injection begins while the core is still covered, then core uncovering g

will be prevented.

These 177-LL results for Category 3 breaks apply to the a

177 FA raised loop (177-RL) plant because of the plant design similarities.

The lowered loop results are in fact conservative for the 177-RL design because 1.

A greater portion of RCS liquid inventory is above the top of the l

core in the 177-RL design than in the 177-LL design.

2.

For the type of RCS pressures that would exist during a Category 3 break transient, HPI flow would be greater for the 177-RL plant than for the 177-LL plants, assuming a single failure within the l

HPI system.

(See Figure 4-4 and Section 6.)

4.4.

Category 4: SBLOCAs Large Enough to Depressurize The RCS Sufficiently to Permit Low-Pressure In.iection 2

2 and 0.5 ft ) will cause the RCS to Category 4 breaks (sizes between 0.1 ft continually depressurize, first to saturation, then to an equilibrium condi-tion with the secondary system.

At that point, SG heat transfer ceases.

E However, the break is large enough to allow the RCS depressurization to 5

continue without SG heat transfer.

As RCS pressure falls below secondary I

pressure, heat energy flows in the reverse direction, from secondary to j

primary.

This relatively insignificant amount of added energy would tend to f

prevent further RCS depressurization; however, Category 4 breaks are large enough to allow continued RCS depressurization, despite the slight addition of secondary heat.

Of more significance than the amount of energy added to the RCS by reverse heat transfer is the overall impact of SG modeling on transient results. The I

4-4 Babcock &Wilcox a McDermott company

L

[

type of SG model used (previous versus revised) has essentially no impact on I

the remainder of an SBLOCA transient once sustained reverse heat transfer is h

achieved. The major dependency of this category of SBLOCAs is on break flow.

Typical RCS pressu'e behavior for a Category 4 break is illustrated as curve r

p H

4 of Figure 4-1..For this category of breaks, the RCS pressure continues to decrease, due to the relatively large break area, until it falls below the shutoff head of the LPI system, ensuring adequate core cooling.

s f

L I

L r

L IL E

L IL r~

9 E

L N

{

4-5 Babcock &WIIcom a McDermott company

Figure 4 Characteristic RCS Pressure Response for SBLOCA Categories l' through 4 3000 2

Category

~ Break Area Range (ft )

l 2500 2

1 Less than.003 ft 2

2 2

0.003 ft to 0.02 ft 4

2 2

3 0.02 ft to 0.1 ft 2

2 j

2000 4

0.1 ft to 0.5 ft 2

?

E cn g

1500

s i

{

2 E

L 1000 m

l 500 4

1 0

I I

I 500 1000 1500 2000 2500 3000 3500 4000 i

1 Time, sec a

W 3

M M

N M

M

'M N

M M

M M

M-M M

M M

M 'M 5

i l

Figure 4-2.

Transient Core Mixture Height Vs SBLOCA Break Size

}5' I

Leoend Break Area 0.04 FTZ 30

^

L (177-LL Analyses Data)

-r 0.055 4

0.07 FT 45 0.0es rT2 2s n

0.10 FT2

- 0.15 FT2 g

h (d

"Q ' ' ~ ~ ~ ~ _ _

5 15

\\.

1. % n-N~ D.

s n# W

{

Top of Core W ~~~

ar n

- - - - ^ *-

,..r 10 5

i i

i a

a a

i a

a a

a n

0 200 400 600 800 1000 1200 1400 1600 1800,2000 2200 2400 260' Time,sec

l Figure 4-3.

Cladding Temperature Response - SBLOCAs That Produce Partial Core Uncovering 1100 (177-LL Analyses Data) 4 1000 11-2 :

e j

9m o

a Legend

~e Break Area h

800

-x-0.055 ft2 4

F ai 8

0.07 ft j

g 2

R 700 0.085 ft 1

d O

1 500

-w-./

~

I I

I l

i I

I 0'

300 600 900 1200 1500 1800 2100 2400 i

Time, sec 4

J 1

1

\\

l 1

I E

0

,I 0

P 9

I E

E P

H d a d e a H e

E H h g

0 wi, 0

o H 8

L p

p o M

o o o L L

d d e e

e r 0

r s e 0

Me u

i w 7

s a o s

R L r

A A P

F F 7 7 MR S

7 7 C

1 1 d

0 p

n 0

m s

e 6

u V

g Ms p

e

/

L m

e t

pa a

R Ml w

e o

0 t

0 a

F 5

R I

w

\\'

o Mf PH l

\\

F I

o P

n 0

H Mi o

0 4

s rapm MC o

i 0

0 4

3 M4eru g

MF i

i 0

02 M

e 0

g

' 0 1

g 0

0 0

0 0

0 0

0 g

0 0

0 0

0 0

0 5

0 5

0 5

3 2

2 1

1 g

3E.E={k 0" 9*

g 1i i1

I I

5.

SBLOCA EVALVA. TION MODELS 5.1.

Summary of The SBLOCA Evaluation Models Beginning in the early 1970's, SBLOCA analyses on the 177-RL plant have been performed utilizing the following series of three SBLOCA evaluation models (ems).

1.

The Early EM: This model, which was used prior to about the year 1976, combined the entire RCS into a single loop (see Figure 5-1.).

A spectrum of SBLOCAs analyzed with this model is document-ed in Reference 5.

2.

The Intermediate EM:

This model was used from about 1976 t E

1985.

The two primary loops were modeled separately (see Figure al 5-2.).

References 3 and 4 document analyses performed utilizing this model.

3.

The Revised EM:

This model resulted from the SBLOCA Methods Program (see References 1 and 2).

The Revised EH contains two separate primary loops, with the primary and secondary SG regions modeled in greater detail than in the intermediate model (see g

Figure 5-3).

5.2 Model and Inout Chances l

The revised SBLOCA EM evolved from the intermediate SBLOCA EM as a result of l

the SBLOCA Methods Program.

This program was developed to address the requirements of NUREG-0737,Section II.K.3.30, and resulted in certain code modifications being made to the intermediate SBLOCA EM.

These modifications are described in detail in Reference 1, and include the following:

1.

A non-equilibrium pressurizer model.

2.

A two-phase RC pump model.

3.

A mechanistic steam generator model.

I 5-1 mggg%

a ucoumoir company

I 4.

A revised auxiliary feedwater (AFW) model.

Additionally, the leak discharge models differ between the early EM and the other ems.

In the early EM, the Moody Correlation was used to calculate both subcooled and satdrated leak discharge.

In the intermediate and revised ems, subcooled discharge rates are calculated with the orifice equation and saturateddischafgeratesarecalculatedbytheMoodyCorrelation.

In addition to these model differences, key input parameters differ among the three ems, as is noted in Tables 5-1 through 5-3.

These changes in the input parameters reflect an evolution that took place between 1970 and 1985.

This evolu' tion affected the understanding and interpretation of the interfaces between the requirements for conservatism contained in 10 CFR 50.46 and 10 I

CFR 50 Appendix K and the actual setpoints, functions, and conditions that exist in the operating plants.

In addition to the input parameter changes listed, pressurizer surge line form-loss coefficients were also changed in the revised modal.

The surge line form-loss coefficients being used for the analyses documented in this report are best-estimate values that are appropriate for those SBLOCA transients in which forced RC flow is of relatively short duration.

I

'I I

I t

I I

I I

I s-2 a McDermott comparty I

Table 5-1. Comparison of Key Input Parameters For The Earl y. Intermediate. and Revised ems Inout Value Early Intermediate Revised Parameter M

Model Model Low RCS pressure reactor 2065 1900 1900 trip setpoint, psia Low RCS pressure safety 1515 1465 1585 E

features actuation a

system (SFAS) setpoint, psia Delay time from SFAS 25 35 35 actuation until start of HPI flow, sec Main feedwater (MFW) 43 14 7

coastdown time after reactor trip, see Delay time from RC pump 36 40 40 trip to the start of auxiliary feedwater (AFW) flow, see Number of SGs supplied with AFW 2

2 1(a)

Secondary SG level for AFW 32 10 10 cutoff, ft (a)The analysis discussed in Reference 1, which was performed with the revised EM, assumed that both SGs were supplied with AFW.

The study documented in this report assumed that only one SG was supplied with 3

AFW.

5 I

I l

I 5-3 hock &WWilcox a ueonmone company

I I

!I l.

Table 5-2.

Comparison of AFW Flow Rates Vs

'm Secondary SG Pressure For The j

Early. Intermediate. and Revised ems AFW Flow Rates oer SG aom j

Secondary SG Early Intermediate Revis9d l5 Pressure, psia EM EM EM l

0 540 570 1140 815 540 570 1140 1136 378 400 772 1244 298 255 314 1 evised EM assumes one AFW pump available with all flow supplied to one I

R SG, not one-half of flow to each SG as in Intermediate EM.

I

'I lI I

I I

I l

I I

lI 5-4 m g ggg,,

a ucoermot company

Table 5-3 Comparison of High Pressure Injection (HPI) l Flow Rates Vs RCS Pressure For The Early, Intermediate and Revised ems I

HPI Flow Rates, com Early Intermediate Revised RCS Pressure. osia 1

EM EM 1640 0

0 0

1615 0

158 150 1515 170 279 265 1215 374 468 444 1015 471 554 526 615 624 684 649 215 750 788 748 0

800 855 812 I

l

~

I I

I I

I I

I I

5-5 y

l

E i

Figure 5-1. CRAFT 2 Noding Diagram For Small Breaks -- Early Model I

I I

r 7

g g

o_

e @

g)

I g

s--s I

ho a

-O i

I g

~

l O

- Ge oe9 I

--g;O l

e- -e 5

0:

m I

e I

I E

5-6

I Figure 5-2. CRAFT 2 Noding Diagram For Small Breaks -- Intermediate Model I.

t,. '

28

,g CPT 4

12 43 g

7 g

3 4

18 17 e

l-e s

GE~

I 30 k 'O' (

g e 6

g

~

. 23 10 1

t@

@~

l

'y b

I 110!!!: AN;f.lBN44 B&f 4 NOT $NOWI 3 Glagar gg lI006 13 11 CentAlma nt n00t i4 21 Q

g g

w v

v Patu 31 11 Ltas Patu Pete Betis T

• _I gv v

fl CSITAtletNT l'I PAIN l1 strygg ([3s Path peg Calfatnetur TO Ittat n00t PAIN REP 41tuf! CDfalmeint $Peaf

!?!!IE 400t 40 10tNflP IC8'l h P4 ft1 48 10tWf t PitAf t elt I1 00EIC00tt t.2 CORE I

L0tte Pttisus 3.4.20. 29 N0f Lil PIPlh6 g

C001 & uPPit Ptlleus S.30.41,43 N87 LIE uPPt3 3 18 met tts PIPtat 8.3 18 fugt!

4 17 18 L # Ptt ut&O f.32 3G LO'tE NEA0 3.10 Iftte Cluttaf00 fugt!

O Caet OfPat!

6 19

!! Con 64RT. 14 9.21.33 COLD Ltt PtPteis 7 20 IG LOM E ning 10.23.34 Puert I 14 21 COLO LIG PIPlot 18.11.14.21.35.38 COLD Lit PtPint

f. t l. 2 2 COLO tts PtPlus ll.Is 00suttutt I

10 UPPtt SteeCSER 20 LPl t!

Pell!vGilla 13.14 UPPit 00meCDIR 13 ChialustNT 17 P'tI3U'llt' g3 UPPts Pttrue il VEuf VALVI 24 21 18 UPPit ut40 10 !?

at 30 Cott f0 uPPit PLinus y

og 43 18 UPPts ut 0

30. !:

Ltas & affuna pain I

I l

l 5-7

I Figure 5-3. CRAFT 2 Nodin9 Diagram For Small Breaks -- Revised Model i

e 8

lE ll it

. e i.

,g

(=

~X-a 3

. )(e.

..)

G.,

6

(-

"X-I

,,,,)(,,,e n

4.

6.

e, A

6

("

"X'

,,)(,.

)

6 t

6.

e I

e a

G, g.

-sqM,,

m,a, eg_..

'i e,

0, e-V_c I

-e a a gI) a i,=>.-,i.=i

- p I

e O

nog nani==r===

t..

I h

I9tafIf f RAfIRA h

IStafIfIlaf tet I

90 tegstemte

0. 8 test 0

69998 Pitete 0.0.80.89 99f LIG 989504 8

6ett 8 tette fif tte 6.84.44.50 set kle. Opt 00 0.80 999 LIS 909000 9

test ef9400 F. It to 43558 M68 9.18.00 4060 hit 969:04 0.04.89 4060 684 Pleest t e. it. Bt feest I

9.48.83 4048 480 Pettet e t, s t. it.at.36.s4 sett LIG #69808 to 98900 Gesettelt 16.04 Sevetesee 79 98880081300 it 496 I

is (Hniceset i t. a ente 0 es eis 88 88780 P50000 90 94808501810. 30000 test 94.80 Se e#988 Stat 68 fitf f atti I

10 feet 96 96 79HS 80.47 aft et f ate 60 le feH8 Of testesentet staael 48.84,49.64.6.80 St=484Ht449 0890 08.09 tett it 588t9 # iese L

4 0. 9. 6 0. St. 4. 4 F 66 60testatt SISS et feeg 68 Se fetet 66.49 60 00stStett of fate 70 It feast S. P. PF. 7 0. l f. S e 86 0880s0409 litt I

et 04. fl. f t. 46, t e St.88testaaf toeg

49. Il 8089 e4f 89 I

I 5-8

I I

I 6.* SBLOCA SPECTRUM ANALYSIS WITH THE REVISED EM 6.1.

Introduction The requirements of NUREG-0737,Section II.K.3.31, have been previously discussed in sections 1 and 2 of this report.

Basically, a demonstration of

" plant-specific" compliance is required by plant configuration. This section presents the qualitative arguments and analyses necessary for the raised-loop g

plant configuration evaluation to comply with the requirements of II.K.3.31.

Note that in order to satisfy the requirements of 10 'CFR 50 Appendix K, SBLOCA reanalyses with the revised model will assume design-basis conditions.

These assumed conditions include:

1.

120% of the 1971 ANS 5.1 decay heat generation standard.

2.

Failure of one emergency diesel generator to start operating l

following a reactor and turbine trip and a coincident loss of offsite power (LOOP).

This failure results in the inoperability of one HPI pump, one LPI pump and one AFW pump.

3.

The coincident reactor and turbine trips combined with the coincident LOOP and the single failure of one emergency diesel generator result in no AFW injection in the intact loop SG.

Note also that only breaks at the RC pump discharge location were considered for reanalysis.

This location has previously been identified as the worst-g case location for B&W plants (see Reference 6).

W The break spectrum to be analyzed and the sufficiency of this spectrum will g

be justified in the following paragraphs.

In all cases, the demonstration of E

reasonably close similarities between results obtained with the previous and revised models will 1.

Verify the acceptability and conservatism of the previous EM and of the results obtained using th,at model.

I 6-1 hd &Mhz a ueoeemote company

I 2.

Demonstrate that the revised EM is an acceptable analytical tool for use in future SBLOCA studies.

3.

I Verify that a complete break spectrum that meets the requirements of 10 TFR 50.46 now exists for B&W's 177-RL plant.

6.2 Justification For The SBLOCA Cases Selected For Reanalysis 6.2.1.

0.04-ft2 Break The 0.04-ft2 break is considered to be a transition break size between Categories 2 and 3.

The evaluation of such a transition break is necessary to determine how the transient is influenced by changes incorporated in the revised EM.

Analyses (Reference 5) with the early EM indicate that this particular size limits the probability of saturated RCS repressurization, a characteristic of Category 2 breaks that indicates primary-to-secondary decoupling, while maimizing the potential for SG influe' ce on a Category 3 n

break.

Therefore, the possibility of significant changes in transient response resulting from the revisions to the EM and the historical interest in breaks in Categories 2 and 3 dictate the selection of the 0.04 ft2 break for reanalysis.

In addition, the 0.04-ft2 break at the RC pump discharge location has been previously analyzed for the 177-RL plant (Reference 5).

Following the reanalysis of the 0.04-ft2 break, a basis for direct comparison between the results obtained using present versus previous modeling techniques will then exist.

6.2.2.

0.01-ft2 Break l

The 0.01-ft2 break was chosen because it is representative of Category 2 break sizes. Historically, Category 2 breaks require SG heat removal for RCS depressurization.

If SG heat rc." val is interrupted, Category 2 breaks are small enough to allow saturated RCS repressurization. Therefore, this is the

>I category of most interest when considering the potential effects of the EM revision, particularly those pertaining to SG and AFW modeling.

The 0.01-ft2 break reanalysis for the 177-RL plant was performed as part of the SBLOCA Methods Program.

The details of this reanalysis are discussed in I

e-2 a McDermott company 1

I Appendix F of Reference 1 and will not be repeated here.

However, three 2 break.

points should be noted about the 0.01-ft 1.

A reanalysis of a Category 2 break was performed mainly to demonstrate that the revised model can predict the repressuriza-tion /depressurization phenomena.

Showing that the core did not uncover was not a concern since historically no core uncovering has been predicted for Category 2 breaks on B&W's 177-FA plants.

2.

As stated in Reference 1 Appendix F, the revised EM did predict l

2 break.

the expected phenomena for the 0.01-ft 3.

The 0.01-ft2 break reanalysis documented in Reference 1 assumed that AFW was supplied to both SGs and that the SG secondary level setpoint was at 10 feet.

The injection of AFW into only one SG is not expected to detract from the revised EM's ability to predict Category 2 break phenomena.

Asymetric SG behavior has been evaluated and reported in Reference 3.

Feeding only one SG during a Category 2 break is also not expected to result in core uncovering, which verifies established conclusions for this category of SBLOCA.

Minimum AFW flow and SG level setpoint requirements are determined by studies of SBLOCA demands for b-c cooling. Specific SBLOCA transient analyses verify these minimum 2 break reanalysis requirements.

Therefore, the existing 0.01-ft presented in Reference 1 is considered to be valid and an g

additional reanalysis of this break size is not necessary.

W 6.3.

Justification For Excluding Categories 1, 3 and 4 3

Breaks From The Reanalysis Soectrum g

l 6.3.1.

Cateaory 1 Analyses of Category I breaks with the intermediate EM (Reference 3) have demonstrated the capability of the make-up and HPI system under design-basis conditions to maintain sufficient RCS liquid inventory to sustain natural circulation and prevent core uncovering.

Revisions made to the EM are not expected to result in significant changes in transient behavior for breaks in l

this category.

At most, a slight shift may occur in the maximum break size defined for Category 1.

This shift would only involve movement between I

6-3 Babcock &WHcom a moermon company l

I Categories 1 and 2.

A representative Category 2 break (the 0.01-ft2 I

break) has been reanalyzed with the revised EM.

In comparison with other break categories, Category I breaks have been shown to be insignificant in terms of transient severity and the potential for core uncovering.

The reanalysis of a Category I break is therefore not required to ensure that the break spectrum is compl'ete.

6.3.2.

Catecory 3 A representative Category 3 break (the 0.07-ft2 break) has been reanalyzed for the 177-LL plants using a revised EM and assuming design-basis conditions l

(see Reference 8).

That 177-LL reanalysis, which assumed AFW injection into both SGs, demonstrated that 1.

For the 0.07-ft2 break, the core did not become uncovered.

2.

The previous 177-LL EM, which essentially ' corresponds to the intermediate EM for the 177-RL plant, produced conservative results (Reference 4) relative to those from the revised 177-LL EM.

The previous 177-LL results (Reference 4) did predict partial core uncovering for some Category 3

breaks, but also predicted that fuel cladding temperatures would be maintained well below the 10 CFR 50.46 limit of 2200F (see Figure 4-3).

The previous 177-LL results are therefore conservative I

relative to the results of the revised 177-LL studies.

I Those previous 177-LL Category 3 results are also expected to be conservative for the 177-RL plant, for the following reasons:

1.

For the range of RCS pressures that will exist during most of a Category 3 break transient (see Figure 4-1, curve 3), the HPI system for the 177-RL plant will provide higher flow rates than I

the HPI system for the 177-LL plants (see Figure 4-4).

2.

I By design, the 177-RL plant has more of its RCS inventory above the top of the core than does the 177-LL plant.

This means that more RCS liquid is available to drain into the reactor vessel and cover the core.

I I

e-4 A McDermott comparty

I 3.

Category 3 breaks result in fairly rapid RCS pressure decreases to values equal to or less than secondary pressure.

This means that, for much of the event, the SGs would not be a heat sink for the RCS.

Instead, the SGs would probably be a heat source. This point.is important because a.

When the SG, stops functioning as a heat sink, the type of SG model used (previous versus revised) becomes irrelevant and g

b.

The amount of heat that could potentially be transferred from the secondary system to the primary system in the present 177-RL EM is lower than in the 177-LL EM due to the lower level setpoint for AFW cutoff and to the asymetric AFW supply.

Therefore, since the conclusions of the previous 177-LL Category 3 SBLOCA analyses are conservative for the 177-LL plants, they are also considered to i

be conservative for and applicable to the 177-RL plant.

The existence of a conservative Category 3 break spectrum, which shows that Category 3 breaks are not limiting and do not result in violations of 10 CFR 50.46, eliminates the need for a Category 3 reanalysis for the 177-RL plant.

l 6.3.3.

Cateaory 4 Category 4 breaks exhibit a fairly rapid and constant depressurization to pressures below the LPI shutoff pressure, thus assuring long-term core protection.

Saturated RCS pressure rapidly decreases below secondary pressure.

When this occurs, reverse heat transfer begins; revisions to the SG model will therefore have only limited and short-term effects on the l

overall transient.

This category of SBLOCAs has been analyzed for the different plant types, l

177-LL and 177-RL, and with the various ems.

While early EM analyses pre-dicted some degree of core uncovering for this category, EM technical improvements demonstrated the large conservatism of the early EM, with revised results (intermediate EM) predicting no core uncovering. Analysis of this category SBLOCA with the intermediate EM has been performed for both the 177-LL and 177-RL plant with similar results; the RCS pressure is controlled by the leak discharge such that the CFTs an,d LPI are actuated before any core l

I 6-5 gg a uconmois company

I uncovering occurs thus traintaining satisfactory core cooling throughout the transient.

For these break sizes, the features that have the most significant influunce on the transient are the critical flow and leak discharge models; neither of these have been revised.

Therefore, the reanalysis of Category 4 breaks is I

not considered necessary to demonstrate compliance with NUREG-0737 II.K.3.31.

I I

I I

I I

I I

l3

.I

'I

!I

!I I

e.e J MCDermott COmpar1y I

I I

I s

7.

RESULTS OF THE SBLOCA REANALYSIS l

7.1.

Introduction The discussion that follows is divided into three sections:

1.

Discussion of the 0.04-ft2 SBLOCA early EM analysis.

2.

Discussion of the 0.04-ft2 SBLOCA revised EM analysis.

3.

Comparison of the two transients and a discussion of their g

significant differences.

7.2.

0.04-ft2 Break g

7.2.1.

Results Usino The Early EM The transient responses of selected parameters for the 0.04-ft2 break, which were obtained using the early EM, are presented in Figures 7-1 through 7-3.

The sequence of events is provided in Table 7-1.

The break caused the RCS to depressurize until, at the low RCS pressure setpoint of 2065 psia, the reactor tripped.

The turbine tripped and a coincident LOOP also occurred at approximately the time of the reactor trip.

Primary pressure rapidly decreased to approximately 1130 psia, when system l

saturation slowed the rate of depressurization.

The inventory depletion continued, causing natural circulation to be interrupted as the primary loop drained.

Boiler condenser (b-c) cooling began at 530 seconds.

Both primary and g

secondary depressurization were caused by AFW injection.

At approximately 820 seconds, AFW was terminated when the secondary SG level reached the 32-ft setpoint, and both the primary and the secondary depressurizations were ended.

Primary and secondary pressure remained relatively constant from -830 seconds until -1900 seconds, after which time the primary pressure fell below the secondary pressure.

The increase in the RCS depressurization rate after

-1900 seconds was caused by an increase in leak quality.

This increase in l

~

hd &MICOM a McDermott company

I I

quality coincided with the time at which the primary loop drained (i.e. the only liquid remaining within the RCS piping external to the reactor vessel (RV) was that which) was trapped in the pump suction piping).

Primary depres-surization continued, and at 2920 seconds both CFTs actuated.

The transient l

was terminated at; 3500 seconds, at which time the RCS liquid loss rate was l

essentially matched by the. ECCS injection rate.

For the 0.04-ft2 break analyzed with the early EM, the core was predicted to remain covered.

I Therefore, no increases in fuel cladding temperature above the saturated liquid temperature occurred.

7.2.2.

Results Usina The Revised EH The transient responses of several key parameters for the 0.04-ft2 break, I

analyzed with the revised EM, are presented in Figures 7-4 through 7-7.

A sequence of events for the predicted transient is provided in Table 7-1.

A rapid primary system depressurization followed the break opening and resulted in a reactor trip with a coincident LOOP, a turbine trip, and MFW and RC pump coastdown when pressure decreased to the 1900 psia setpoint. The pressure continued to decrease to saturation, after which steam voids formed and increased in volume within the upper hot leg elbows (the U-bends). This steam formation drastically reduced primary flow until the primary and secondary systems became decoupled.

Following decoupling, an RCS repressuri-zation occurred between -50 to 120 seconds.

The repressurization ended when primary flow was reestablished.

This flow restoration occurred as a result I

of the flashing of liquid within the upper RV.

Eis flashing forced more fluid into the hot legs, increasing the hot leg levels sufficiently to restore flow over the U-bends.

A rapid primary system depressurization followed this flow surge.

After the RCS pressure decreased relatively rapidly to -1300 psia (at -200 seconds), the rate of depressurization slowed due to a second interruption of l

hot leg flow by voids in the U-bend.

This flow interruption reduced SG heat removal sufficiently to allow a slight RCS repressurization between -200 and 250 seconds.

The condensation of primary steam (b-c) by the AFW spray began at -220 seconds.

However, AFW was stopped at -225 seconds interrupting b-c when the secondary level in the broken loop SG (SG-1) reached the 10-foot I

I 7-2 Bakoctr &Micom a uconmois company

I level setpoint.

(Note that the intact loop SG, SG-2, was not supplied with AFW.)

By the time the AFW was shut off, a primary-to-secondary differential temperature (61) of -10F existed. Pr imary-to-secondary heat transfer resumed at -240 seconds when the RCS liquid level became sufficiently low to allow l

the condensation of primary steam by the secondary liquid pool.

Once this form of b-c was established, the large primary-to-secondary AT resulted in g

increased SG (SG-1) heat transfer.

The high rate of SG heat transfer combined with the effects of the break and HPI cooling caused the RCS to depressurize at a fairly constant rate until -1000 seconds.

By that time, the primary and secondary systems had reached thermal equilibrium and SG heat removal essentially ended.

HPI cooling and energy losses out of the break allowed the RCS pressure decrease to continue, with primary pressure falling l

below secondary pressure after approximately 1000 seconds.

The resultant reverse heat transfer caused the depressurization of the broker. loop SG (SG-1) between 1000 and 1500 seconds (see Figure 7-5).

The reverse heat transfer was reduced substantially at 1500 seconds as the liquid inventory within the broken loop hot leg piping and SG primary side (SG-1) was depleted. Again, the intact loop SG (SG-2) did not exhibit the same pressure characteristics as did the broken loop SG (SG-1) because AFW was not delivered to the intact loop SG.

The break and HPI cooling caused the primary pressure decrease to continue between 1500 and 3200 seconds.

The l

CFTs began to inject at 2650 seconds and system refill began immediately thereafter.

The transient was continued until 3200 seconds, when the leak i

flow was matched by the HPI flow.

The inner RV liquid volume is shown in Figure 7-6.

The minimum RV liquid volume occurred at approximately the same time as the actuation of the CFTs.

Even though the minimum RV liauid volume is slightly below the top of the core, the mixture level stayed well above the top of the core throughout the transient.

Consequently, no fuel cladding temperature increase above RCS l

saturation temperature was predicted for this break.

I E

1 I

7-3 ggg g gg, a uconmore company

I 7.2.3.

Comparison of The Results From The Analyses Usino The Early and Revised ems A comparison of selected results from the early and revised studies is shown in Figures 7-8 through 7-12.

A comparison of the sequence of events is provided in Table 7-1.

'l The reactor trip occurred at 5 seconds in the early case and at 10 seconds in the revised case.

This variance is due mainly to the differences in the assumed low RCS pressure reactor trip setpoint:

2065 psia in the early model versus 1900 psia in the revised model. The variance in reactor trip time has I

little effect on the transient as RC pumps and MFW pumps are still opera-tional.

Therefore, SG heat removal at full power is maintained and a heat generation-heat removal balance occurs.

Following the reactor trip, RCS pressure in both cases decreased rapidly to saturation.

The early model predicted a lower saturation pressure than did the revised model. Saturation pressure is a function of energy removal, indicating that the early model predicted greater primary-to-secondary heat transfer during the first 100 to 200 seconds of the transient.

The higher energy release in the early model I

is attributed to the following events:

1.

A longer predicted pump coastdown time in the early case.

2.

The maintenance of a relatively high primary-to-secondary heat I

transfer coefficient by the early SG model throughout the entire transient.

I The assumption in the early model of a 43.5 second main feedwater 3.

coastdown time, as compared to 7 seconds for the revised model.

4.

The overprediction by the early model of the duration of two-phase natural circulation because of the noding arrangerrent within the top of the hot leg U-bend (see Figures 5-1 and 5-3).

5.

The availability of the equivalent of two SGs in the early model and only one SG in the revised EM.

The time at which the RCS pressures for the two cases equalized was -1750 I

seconds (from Figure 7-8).

This equalization occurred despite the fact that the early case predicted a much greater RCS depressurization rate between time zero and -250 seconds (see Figure 7-8).

By -250 seconds, the RCS l

7-4 Babcock & Wilcom a MCDermott CompJrty

{

I pressure in the early case was -250 psi lower than in the revised case.

The major difference between the two cases that allowed RCS pressure in the revised case to decrease more rapidly between -250 and 1750 seconds was the cooling effects that resulted from larger HPI flow rates in that case (Table 5-3).

Also contributing to the faster depressurization in the revised case was a relatively'high rate of b-c cooling that took place on the secondary liquid pool (in SG-1) after the AFW flow had stopped.

Figure 7-9 does show a sharp drop in SG pressure in the early case between -530 and 830 seconds.

This pressure decrease resulted from the interruption of F.CS flow (at -820 seconds) while AFW was still being injected. The rapid decrease in secondary l

pressure indicates that a large primary-to-secondary AT also developed in the early case during that period.

g This increasing AT did result in an escalation in the RCS depressurization rate between -530 and 820 seconds (Figure 7-8).

However, by -820 seconds (when the AFW flow stopped), the RCS pressure and temperature had decreased g

sufficiently to bring the primary and secondary systems to a near-thermal equilibrium condition. SG heat removal then became very minimal as indicated by an RCS pressure stabilization between -830 and 1900 seconds in the early case.

While RCS pressure was stabilized in the early case, two sets of conditions were working to continue the RCS depressurization in the revised case.

1.

Boiler-condenser cooling, HPI cooling and energy losses through l

the break between -250 and 1000 seconds.

2.

HPI cooling and energy losses through the break after -1000 seconds.

These combinations were able to overcome the energy addition to the RCS from core decay heat and secondary-to-primary heat transfer (a very small effect that was present after -1000 seconds).

The resultant continuous net energy loss from the RCS after -250 seconds prevented the RCS pressure from stabilizing in the revised case.

Consequently, the RCS pressure in the revised case was able to first equalize with and then decrease to slightly g

below the RCS pressure in the early case.

I 7-5 Babcock & WHcom i

^

lI Another significant finding of these analyses concerned the collapsed liquid levels in the RV.

As a result of the differences in ems and input assump-tions between the two esses, the revised model predicted a slightly lower

I minimum RV collansk liauld level (Figure 7-10). This is mainly attributable to differences in.SG heat transfer, modeling discretization and the subcooled

.l~

leak discharge model. Up to ~250 seconds of the transient, the revised model i

predicts approximately twice as much leak flow as the early model (Figures 7-11 and 7-12).

This is due primarily to the reduced SG heat transfer perfor-mance of the revised model resulting in an elevated RCS pressure response jg (Figure 7-8), and a more conservative subcooled leak discharge model (orifice W

equation).

After ~250 seconds of the transient, both early and revised ems use the Moody Correlation for saturated blowdown but the more finite modeling of the revised model allows for higher quality discharge resulting in lower mass flow rates.

The RV mixture heiaht in both cases remained above the top of the core, thus preventing fuel cladding temperature increases above the RCS fluid saturation temporature.

While the previously discussed differences in the ems (the input assumptions and the transient responses between these two cases) are noteworthy, they are not significant in-their overall impact on the 0.04-ft2 break.

Again, the core remained covered in both cases, so the fuel cladding temperature responses were fundamentally the same.

Furthermore, as is illustrated in Figures 7-8 through 7-12, the responses of key parameters differed at times during the transient, but the transient ended with the primary and secondary systems essentially in the same condition in both cases.

I ll

I

I

-lE 7-6 sabcockaWHcom a M(Dermott company

-+<w-w--_,. _ _ _

,9 y__

-..r M

e

'-4 e

rew e-+- ' - -W-- - - " " - -

I Table 7-1. Comparison of The Results of The [arly and Revised Evaluation Models for The 0.04 ft Break Transient Time. see Early Revised Event Hgdal Model Break occurs 0

0 Reactor trips on low RCS pressure 5

10 E

(2065 psia for early model; 1900 psia for revised 5

model), turbine trips, RC pumps trip on coincident LOOP, MFW coastdown begins, AFW system actuates Hot legs saturste 42 25 AFW injection cosumances 41 50 MFW flow coastdown ends 50 17 SFAS actuates on low RCS pressure, 30 20 (1515 psia for early model; 1585 psia for

[

revised model)

~

High-pressure injection begins 55 55

~

Break region saturates 530 210 Flow-controlled phase ends 530 230 SG natural circulation level reached (32 ft for 820(*)

225(b)

E early model; 10 ft for revised model), AFW shuts 5

off Primary and secondary systems reach thermal 1200 1000 equilibrium CFTs actuate 2920 2650 Analyses ended 3500 3200 I

(*)The primary-to-secondary differential temperature at this time l

was less than IF.

(b)The primary-to-secondary differential temperature at this time was -10F.

I 7-7 mggg a McDermott company

[

[

Figure 7-1.

RCS and SG Pressures, 0.04-ft2 Break, Early Model 3000 i

i i

i i

i Legend 2500 -

RCS Pressure

- - - - - - - Secondary Pressure

\\

2000 2

E g

1500 h

[

i M

1000

[

500 E

Oh i

i e

i i

0 50G 1000 1500 2000 2500 3000 3500 Time,sec E

7-8

Figure 7-2.

Inner Vessel Liquid Volume, O.04-ft2 Break, Early Model 3000 2500 2000 m

C 0

N-1500 y

Top of Core a

3 E

3 1000 Bottom of Core 500 a

0 0

500 1000 1500 2000 2500 3000 3500 Time,sec Note: Liquid volume is indicative of collapsed water level. To determine mixture height, void fraction must be considered. The mixture height is used to determine core cooling

14 2

Figure 7-3.

Leak Flow Rate, 0.04-ft Break, Early Model 1

1000 i

i i

8 8

800

~

8 R

600 5

~

i l

E.

1 2

400 -

l O

E l

is

.S 200 0

500 1000 1500 2000 2500 3000 3500 l

Time,sec i

i l

2 Figure 7-4.

RCS and SG Pressure, 0.04-ft Break, Revised Model 2500 i

i i

i 2000 -

Legend RCS Pressure

~

~~~~~~~

1500 l

2 i

E 7

C y I --4 + --

- m ~ _ -

P 1000 g

E a-500 l

0 I

i 8

500 1000 1500 2000 2500 3000 3500 4000 Time,sec W

W W

W W

W W

W W

W W

W W

W W

W W

m m

n l

l llll 1

)i{

0 00

~

~

4 W

m FA W

F 0

h A

0 i

5 t

i i

o 3

w n

1 2

G G

S S

wl 0

e G

G 0

d N

a 0

o S

S 3

M p

p d

o o

o o

L L

s

-r i

n t

ve d

e c

R n

k a

e o

t g

r n

00 k

i e

B I

i 5

r a

L 2

er B

2

~

t uf 4

0 00 c

i i

0 e

c s,

0 2

e e

m r

s m

i u

T s

r e

r 0

P 0

i 5

y 1

ra r

dnoce m

S 0

r G

0 i

S 0

1 m

5 7

er wi u

0 g

0 i

5 F

m_

0 r

0 0

0 0

0 0

0 0

5 0

5 0

5 0

2 1

1 0

0 9

9 1

1 1

1 1

m r

5 2a De8o$

L.

• . C T

-r

~b

\\

l

2 Figure 7-6.

Inner Vessel Liquid Volume, 0.04-ft Break, Revised Model

3000, i

i 2500 2000 1

---

~------'

1500 e---

f Core Top o 7

m C

U ii."

1000 -

O v

,Er Bottom of Core 500 0

8 8

I I

0 500 1000 1500 2000 2500 3000 3500 Time,sec Note: Liquid volume is indicative of collapsed water level. To determine mixture height, the void fraction must be considered. The mixture height is used to determine core cooling.

l l

t

-r t

-r m

00 5

3 M

00 i

l a

e 0

d 3

a o

M c

des i

m

' ve 0

r R

i 0

e 5

2 ka a

er r

'B 2tf 0

a 0

a 4

0 r

2 0

0

ces, mt e

e a

m R

i 0

T w

0 i

a a

o 5

l 1

c F

kae m

L r

0 i

7 a

0 0

7 1

m e

r rug i

x F

-c 0

i i

0 5

n u

0 r

0 0

0 0

0 0

0 0

0 0

0 0

0 2

0 8

6 4

2 1

1 m

ENE. -.{e $c s3 c

i m

m_

~E

(_

l 1

2 Figure 7-8.

RCS Pressures, 0.04-ft Break, Comparison of Results of Early and Revised Models 2500 i

i i

i i

i Legend Resised Model 2000

- - - - - - - Early Model e

i 1500

~a

\\

g

\\ ~ -

1000

~

500

~~

• ==.

0 0 500 1000 1500 2000 2500 3000 3500 4000 Time,sec me a=

==

num um em um en am um em um em um que um en am um

l M

M M

M M

M M

M M

M M

M M

M M

M M

M M

l 2

Figure 7-9.

SG Secondary Pressures, 0.04-ft Break, Comparison of Results of Early'and Revised Models 1150 i

i i

i i

i i

1100 m;

Revised Model, Intact ~

[

L p SG, SG-2, (no AFW) 1050 E

i N

i

=>

1 0

\\

~

L.

E

\\

e g

x

/-Revised Model, Broken Loop SG, SG-1,

\\

Ai 1000 B

g (with AFW) ~

O k

E k

U T

e*

\\

900 g

I Early Model

\\

f L------------2------------

800 8

8 0

500 1000 1500 2000 2500 3000 3500 4000 Time sec

2 Figure 7-10.

Inner Vessel Liquid Volumes. 0.04-ft Break, Comparison of Results of Early and Revised Models 3000 s

i s

l Legend 3

2500 l

Early Model k

Revised Model

\\

2000

\\

\\

'~~--~s s

\\

+

s s

N C

8 2

" ~ Top of Core

~~~

' ~ ~ - - ~ ~ - ~ ~~~

=

1500 g

3 1000 Bottom of Core 500 O

i i

e i

n 0

500 1000 1500 2000 2500 3000 3500 Time,sec Note:

Liquid volume is indicative of collapsed water level. To determine mixture height, void fraction must be considered. The mixture height is used to g

g g

g g et g e g co g g. g g

g g

g g

g g

g g

g

l l

lll 0

~ '

~

0 5

l 3

e d

l o e M d o

d M e

f s y 00 o

3 i

l v r 0

s e a 3

t R E lu s

e R

d n

0 f

e 0

o ge 5

3 2

n L

o s

i ra pmo C

0 8

0 k

0 a

2 er s l

B e

d c

2 o e

tM s,

f d

0 e

4 e 0

m

0. is 5

i 1

T 0v e

,R swd on l a F

y 0

kl 0

i ar 0

ea 1

LE

\\

A 1

I 1

7

\\

\\

e 0

i r

0 u

5 g

p i

F Q

1 tiI

\\l I !

[ l $

0 0

0 0

0 0

0 0

0 0

0 0

0 0

2 0

8 6

4 2

1 1

8

- iE %3

[m 1lll

I 2

Figure 7-12.

Integrated Flows to Containment, 0.04-ft

' Break, Comparison of Results of Early and Revised Models 700,000

~

lI

/

600,000 p

/

/

/

/

\\

/

500,000 I

/

Revised y

3 Model 5

400.000

/

Early Modei l

/

l

/

8

/

/

I O

300,000

~

j 3

/

l

/

2

/

h200,000

[

~

/

/

I

~

100,000 1

/

1

/

l 0

i t

i 8

0 500

.1000 1500 2000 2500 3000 3500 Time,sec I

I 7-19

I I

8.

CONCLUSIONS I

Qualitative and quantitative assessments have been made of whether the existing SBLOCA analysis spectrum that

applies, either directly or I

indirectly, to the B&W 177-RL plant meets the requirements of NUREG-0737,Section II.K.3.31.

This spectrum consists of specific 177-RL plant analyses (References 1,. 3 and 5),177-LL plant analyses that conservatively apply to

~

the 177-RL plant (Reference 4), and the evaluation provided herein.

This l

spectrum is considered to bound the complete SBLOCA size range (up to 2

0.5-ft ), and the evaluation is considered to fulfill the requirements of section II.K.3.31.

In addition, to more fully respond to the II.K.3.31 requirements, a 0.04-ft2 break at the RC pump discharge location was reanalyzed.

This break is representative of the break sizes within the Category 2 to Category 3 transition regime.

The reanalysis was performed with an EM that was revised I

to conform to the requirements of NUREG-0565. To quantify the effects of the EM changes, the results of the analysis of the 0.04-ft2 break were compared with the results of an 0.04-ft2 break analysis performed using an earlier EM.

This earlier analysis is documented in Reference 5.

The comparison of results from the two cases showed some differences in significant parameters throughout the transients.

However, the core was not predicted to uncover in either case.

Furthermore, at the completion of both transients, long-term cooling had been established (at approximately the same time in both' cases),

and the primary and secondary systems were in essentially the same condition.

The qualitative assessment and analysis results presented herein address the entire SBLOCA spectrum.

For evaluation purposes, the spectrum was reduced to four separate categories indicative of similar behavior trends within each category.

Various ems have been used for analytical assessment of SBLOCA phenomena and ECCS performance.

EM modification has continued historically 8-1 J MCDermott Comparty I

I as a result of regulatory and technology changes.

This report presents l

evidence concluding that both previous and current SBLOCA analyses are conservative for licensing (design basis) application.

Although specific l

transient paramete'rs may differ dependent on the EM utilized, the overall analytical result.s with respect to the acceptance criteria of 10 CFR 50.46 has not changed.

For category 1 and 2 break sizes (up to -0.02 ft ), the 2

core remains covered with' a mixture thereby maintaining fuel cladding temperatures within a few degrees of the RCS fluid saturation temperature.

For category 3 and 4 break sizes (-0.02 ft2 2

to 0.5 ft ), the revised EM will predict much less core uncovering, if any at all, than that of the early EM, thus resulting in lower peak cladding temperatures than those previously reported (reference 5).

The results confirm that the revised EM as well as the previous ems for the 177-RL plant are valid analytical tools that meet the requirements of 10 CFR g

50 Appendix K and NUREG-0737 and will produce conservative results relative to the criteria of 10 CFR 50.46.

Furthermore, the existing SBLOCA spectrum for the 177-RL plant, including those applicable results from 177-LL plant SBLOCA analyses and the analyses discussed in this report, is considered to be a complete SBLOCA spectrum analysis that complies with the requirements of NUREG-0737,Section II.K.3.31..

I I

I I

I I

I

• -2 I

a ucoermeer company

[

[

9. REFERENCES b

1.

N. Savani, J. Paljug', and R. Schomaker, "B&W's Small Break LOCA ECCS Evaluation Model," BAW-10154A, Babcock & Wilcox, Lynchburg, Virginia, l

July 1985.

l

[.

2.

J. Cudlin, et al., " CRAFT 2 Fortran Program for Digital Simulation of a Multinode Reactor During Loss of Coolant," BAW-10092A. Rev. 3, Babcock

& Wilcox, Lynchburg, Virginia, July 1985.

3.

" Evaluation of Transient Behavior and Small Reactor Coolant System Breaks In The 177 Fuel Assembly Plant," Volume III, May 7,1979 via

(

Letter from J. H. Taylor to R. J. Mattson (NRC) dated May 7, 1979, NRC Public Document Room, Accession #79051901C4.

'. H. Taylor (B&W) to S. A. Varga (NRC),

Letter, July 18, 1978.

f 4.

J L

Subject:

"B&W's SBLOCA Spectrum Analysis."

r 5.

L. Cartin, J. Hill, and C. Parks, "Multinode Analysis of Small Breaks L

for B&W's 177-Fuel-Assembly Nuclear Plants with Raised loop Arrange-ment and Internals Vent Valves," BAW-10075A. Rev.1, Babcock & Wilcox, Lynchburg, Virginia, March 1976.

6.

J. H. Taylor (B&W) to Dr. Ernst Volgenau (NRC), Letter, April 14, 1978.

Subject:

" Evaluation of 177 FA Lowered Loop ECCS Concern."

7.

" Evaluation of SBLOCA Operating Procedures and Effectiveness of Emergency Feedwater Spray for B&W Designed NSSS," B&W Doc. Id. 77-1141270-00, Babcock & Wilcox, Lynchburg, Virginia, February 1983.

8.

G.

Anderson, and J.

Paljug, "Small Break Loss-of-Coolant Accident Analyses for B&W 177-FA Lowered Loop Plants in Response to NUREG-0737,

{

Item II.K.3.31," BAW-1976, Babcock & Wilcox, Lynchburg, Virginia, September, 1986.

[

[

9-1 E

e.

a McDermott company

- - - -