ML20196E175

Addendum 2 to Vol 2 to WCAP-12071, Westinghouse Large-Break LOCA Best Estimate Methodology,Vol 2: Application to Two- Loop PWRs Equipped W/Upper Plenum Injection,Addendum 2: Pbnp Plant Specific Analysis

ML20196E175
Person / Time
Site:	Point Beach
Issue date:	12/31/1988
From:	Dederer S, Hochreiter L, Schwarz W WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19295G789	List: ML20196E157 ML20196E162 ML20196E175
References
WCAP-12071, WCAP-12071-V02-ADD02, WCAP-12071-V2-ADD2, NUDOCS 8812090285
Download: ML20196E175 (183)

Text

.

WESTINGHOUSE CLASS 3 WCAP-12071

=.

0 WESTINGHOUSE LARGE-BREAK LOCA BEST ESTIMATE METHODOLOGY VOLUME 2: APPLICATION TO TWO-LOOP PWRs EQUIPPED WITH UPPER PLENUM INJECTION ADDENDUM 2: PBNP PLANT SPECIFIC ANALYSIS S. I. Dederer F L. E. Hochreiter W. R. Schwarz

\ *' * >

C. K. Tsai l,

M. Y. Young December 1989 Westinghouse Electric Corporation Westinghouse Energy Center P.O. Box 355 Pittsburgh, Pennsylvania 15230 l

8012090285 s 1130 i

PDR ADOCK 05000266 P PDC

I

. TABLE OF CONTENTS EXECUTIVE

SUMMARY

vii

~

e 1-0. INTR 00VCTION AND BACKGROUND 1-1 1-1. Introduction 1-1 1-2. Background 1-2 1-3. References 1-4 2-0. WCOBRA/ TRAC MODEL OF PBNP. 2-1 2-1. Introduction 2-1 2-2. Analysis Method 2-2 2-2-1. Steady-State Calculation 2-2 2-2-2. Transient Calculations 2-4 2-3. PBNP Vessel Description 2-4 m 2-4. PBNP Loop Models - Primary and Secondary 2-12 2-4-1. Steady-S'. ate Loop Model 2-13 2-4-2. Cold Leg Guillotine Break Loop Model 2-13 2-5. References 2-15 3-0. REALISTIC PWR CALCULATION AT NOMINAL CONDITIONS 3-1 3-1. Introduction 3-1 3-2. PGNP Model and LOCA Boundary Conditions for Nominal 3-1 Calculations 3-3. Results of PBNP Realistic Calculation at Nominal Conditions 3-9 3-3-1. PBNP Nominal Condition Steady-State Results 3-9 3-3-2. PBNP Nominal Condition Transient Results 3-9 3-4. References 3-12 4-0. REALISTIC PWR CALCULATION AT SUPERBOUNDED CONDITIONS 4-1 4-1. Introduction 4-1 l ~ 4-2. PBNP Model, and LOCA Bounding Conditions for the 4-1 Superbounded Calculations 5 .

se 1816v:t o/112888

. 4-3. PBNP Superbounded Calculational Results 4-9 4-3-1. PBNP Superbounded Steady-State Results 4-9

. 4-3-2. PBNP Superbounded Transient Results 4-9 4-4. References 4-13 5-0. SENSITIVITY STUDIES 5-1 5-1. Applicability of Lead Plant Sensitivity Studies to PBNP 5-1 5-2. Operating System Pressure Sensitivity Study 5-7 5-3. Hot Assembly Location Sensitivity Study 5-7 5-4. References 5-10 6-0. PBNP APPENDIX K CALCULATIONS 6-1 6-1. Introduction 6-1 6-2. PBNP Appendix K Model and Transient Results 6-1 6-2-1. PBNP Appendix X Steady-State Results 6-2-2. PBNP Appendix K Transient Results 7-0. COMPARISONS OF THE APPENDIX K CALCULATIONS AND THE 7-1

REALISTIC CALCULATIONS AT 95% PROBABILITY LEVEL 7-1. Introduction 7-1

- 7-1-1. WCOBRA/ TRAC Code Uncertainties 7-1 7-1-2. WCOBRA/ TRAC Bias 7-2 7-1-3. WCOBRA/ TRAC Uncertainties 7-3 7-2. Conservative Estimate of the 95th Percentile PCT Using 7-8 the "Superbounded" Calculation for PBNP 7-3. Realistic Estimates of the 95th Percentile PCT for 7-10 the PBNP 7-4. References 7-13 8-0.

SUMMARY

AND CONCLUSION 8-1 8-1. References 8-2 8

'* w 1616v;10/112888

. EXECUTIVE SU> MARY

. This report describes the methodology used to perform s large break LOCA

• analysis of the Point Beach Nuclear Plant (PBNP), two loop PWRs with upper plenum injection (UPI). This analysis conforms to G 3ECY-83-472 LOCA analysis approach. The analysis toe tsed for the.c <r.lculations is the Westinghouse version of the COBRA / TRAC code (WCOBRA/ TRAC), which was originally developed at Battelle Pacific Northwest Laboratory. Westinghouse has modified, improved, and validated the WCOBRA/ TRAC code against separate effects and system effects thermal-hydraulic' test data which cover the range of thermal-hydraulic conditions expected for a postulated loss of coolant accident.

The PBNP WCOBRA/ TRAC computer model, which is discussed in detail in this report, is a four-channel core model which includes separate core channels for assemblies under control rod guide tubes, assemblies under support columns, open holes, and free standing mixers, and a het assembly. The location of the ,

hot assembly was determined from sensitivity studies which showed which assembly type would have the highest peak cladding temperature (PCT). In addition, a fourth core channel was modeled to simulate the low powor region on the edge of the core. The structures above the low powered fuel assemblies were also modeled to account for the geometry differences. The power level in these assemblies is substantially less than the average power in the core due to the low leakage core leading used for these reactors. This is particularly important for two-loop PWRs with upper plenum injection, since the flow enters the reactor vessel at tne upper plenum on the outside of the core (or the inside of the core barrel) and can penetrato into the core through the lower power fuel assemblies on the outside of the core. The UPI penotration behavior was demonstrated in the Upper Plenum Test Facility as well as the Japanese CCTF-UPI experiments, which confirmed that the UPI water would directly penetrate the core under the injection port. The four-channel core model WCOBRA/ TRAC calculation 3: elds a more realistic treatmont of the UPI behavior than core models with fewer channels.

• ~

The basic approach used to establish an estimate cf the 95th percentils PCT, as required by SECY-83-472, was to evaluate all se rces of uncertainty including those arising from code development, cc::s assessment, and code 1616v:1o/112888

application to the PWR. To determine the uncertainty associated with the code .'

development and assessment, extensive comparisons were performed with single ,

and integral effects experiments. The comparisons between WCOBRA/ TRAC and the ,

experimental data included void fraction distribution, pressures, pressure ,

drops, flows, and heater rod or nuclear rod temperatures. The peak cladding temperature from the experiments was used to quantify the degree of agreement between the code and data. To quantify the uncertainty associated with the PWR application, a large number of sensitivity studies were performed on the PWR boundary and initial conditions, as well as specific models used in the PWR calculation. These results were reported for a lead two-loop PWR with UPI and are applicable to the Point Beach units.

There are three types of WCOBRA/ TRAC PWR calculations presented in this report:

1. A four-channel, realistic core model calculation at "nominal" PWR conditions (a nominal calculation contains nearly all plant input and code model parameters at their nominal conditions).

2. A four-:hannel, realistic core model calculation at a "superbounded" condition (where all plant input and code model parameters are placed at values which yield higher PCT, as determined from one-at-a-time sensitivi:y studies).

3. A four-channe' core model calculation which complies with the required features of the 10CFR50.46 Appendix K rule.

The four-channel, realistic core model calculation at the nominal condition is an estimate of the 50th percentile peak cladding temperature a two-loop PWR could have for the worst postulated LOCA and assuming a single failure. The calculated peak cladding temperature for the nominal case was (

ja.c To bound all the uncertainties, a four-channel, realistic core model .-

. "superbounded" calculation was performed. In this calculation, input and 1616v-10/112888

. model parameters for which sensitivity studies from the lead plant had shown a

., peak cladding temperature penalty, were at their bounded condition. In t:,ls

. fashion all the PWR calculational uncertainties were considered in one

- calculation to yield a peak cladding temperature which wculd exceed the 95th percentile peak cladding temperature. The WCOBRA/ TRAC codo uncertainties which had been derived by comparing the code to test data were also considered. The resulting PCT for the "superbounded" calculation was (

a .C j

The third calculation performed using WCOBRA/ TRAC was a 10CFR50.46 Appendix K calculation. For this calculation the Appendix K required features and models were included in WCOBRA/ TRAC. Those models and features included; fuel rod swelling, burst, and blockage, ANSI /ANS 1971 decay heat standard plus 20 percent, Baker-Just cladding reaction rates, accumulator bypass and other Appendix K requirements. Using these assumptions and requirements, the resulting PCT was found to be ( Ja.c

~ ' '

The SECY 83-472 methodology for LOCA calculations stipulates that best estimate thermal-hydraulic calculational methods can be used if the uncertainties in the computer code, reactor parameters, and accident boundary  !

and initial conditions are considered such that a 95th percentile can be calculated. This 95th percentile is then compared to the PWR calculation using the required features of Appendix K. The 95th percentile peak cladding temperatur) should be less than that obtained from the Appendix K calculation, which inJicates that the Appendix K requirements have sufficient conservatism to cover uncertainties at or beyond a 95th percentile.

[

lac. All the estimates of

>* the 95th percentile peak cladding temperature were below the value calculated with the required features of Appendix K. Therefore, the objectives of the

"* SECY-83 472 LOCA calculational approach have been acnieved for the PSNP two-leep PRs with upper plenum injectien.

1816v.l e/112888

l

. SECTION 1 INTRODUCTION AND BACKGROUND

. 1-1. INTRODUCTION This report contains the analysis and application of the SECY-83-472 (1) X approach to the Wisconsin Electric Power Company (WEP) Point Ber.ch Nuclear Plant (PBNP) Units 1and2. The PBNP Urits are Westinghouse designed two-loop PWRs equipped with uppar plenum injection (UPI) eeergency core cooling systems (ECCS). The cocputer code used for this analysis is the ECOBRA/ TRAC ode (2) which has been approved by the NRC for LOCA analysis of PWRs equipped with UPI (3). This raport contains the PBNP WCOBRA/ TRAC model description and the results of three WCOBRA/ TRAC plant specific calculations.

The calculations include:

9

1) A nominal plant condition calculation where several cf the plant parameters are placed at nominal conditions to estimate the 50th

. percentile peak cladding temperature.

2) A superbounded plant calculation where all inputs and conditions are placed at their bounded values in the same calculation such that the resulting plant calculation is an estimate of the 95th percentile peak cladding temperature.

3) An Appendix K calculation for the Point Beach units using only these required features of 10CFR50 Appendix K as stipulated by SECY-83-472.

These calculational results support the conclusion that the UPI system effectively cools the core for the postulated case of a large break LOCA.

3*

s e .

1614v:1o/112948 1*1

\

I l

1-2. BACKGROUND .

Westinghouse has submitted two WCAP reports which provide the basis of the NRC ,

Safety Evaluation Report (SER) approving WCOBRA/ TRAC for LOCA calculations .

using the SECY-83-472 approach. The first volume of WCAP-10924 (2) and the acsociated supplements [4,5,6,7) documented the WCOBRA/

TRAC code and the code uncertainty calculation. Volume 2 of WCAP-10924 (Reference (8]) and its supplements (9) documented the application of the }{ COBRA / TRAC code to two-loop PWRs equipped with UPI using the SECY-83-472 approach.

The SECY-83-472 document states that there are three areas of conservatism in the current licensing models: the required Appendix K consarvatism, the conservatism added by both the NRC staff and industry to cover uncertainties, as well as conservatism imposed by the industry in some cases to reduce the complexity of the analysis. Based on a review of the available experimental data and the best estimate computer code calculations, the NRC staff has concluded that there is more than sufficient safety margin to assure adequate performance of the ECCS, and that this excess margin can be reduced without an adverse effect on plant safety. In the new approach, the NRC staff is suggesting that the licensee could utilize a realistic model of the PWR to cliculate the plant response to a LOCA at the most realistic or most probable level (50 percent probability) and at a more conservative 95 percent probability level. The calculation at the 95 percent prebability level would account for uncertainties in such things as power level, fuel initial temperature, nuclear parameters, and computer code uncertainties. The parameters, which would have to be examined, and the methods of how the uncertainties would be combined (either statistically or as a one-sided bias) would have to be justified. The realistic PWR model and the uncertainty analysis can be performed on a lead plant PWR model which is representative of a class of similar plants, that is, two , three , or four-loop PWRs so that lead plant uncertainties are applicable to the individual plants.

The third calculation the licensee must perform is a plant-specific realistic 'c best estimate calculation, which includes the required Appendix K features, such as 1971 ANS decay heat plus 20 percent, Moody break flow model, no return 161stlo/112:s s 1-2

. to nucleate boiling during blowdown. and so forth. This new calculation would be acceptable, provided that the peak cladding temperature is greater than the

. peak cladding temperature calculated at the 95 percent probability level but

. belov; the licensing limit of 2200'F.

The SECY-83-472 interpretation of these results would be that the required features of Appendix K have sufficient margin to cover all unce.tainties inherent in a LOCA analysis combined at a 95 percent probability level. Such a series of calculations would provide an acceptable licensing basis for the NRC and is expected to result in peak claddin'g temperature margin which the licensees can use for improved operational flexibility, low leakage loading patterns to address PTS concerns, and to accommodate more economical fuel designs.

The lead plant chosen was the Northern States Power Prairie Island Units which had the highest ratio of core power to emergency core cooling flow.

Therefore, it was expected that the Prairie Island Units would have a greater

~**

peak cladding temperature sensitivity to different plant assumptions and bounding conditions such that sensitivity studies performed on Prairie Island

~"

would bound the PBNP Units. For the PBNP analysis, the magnitude of the peak cladding temperature changes, from the lead plant sensitivity studies, are not used; rather, the direction of the peak cladding temperature change is used in developing the conditions for the superbounded PBNP calculation. In this manner, a plant specific superbounded calculation is performed for PBNP at the limiting conditions as an estimate of the 95th percentile peak cladding temperature. Also, a PBNP plant specific Appendix K calculation is performed using the required features of 10CFR50 Appendix K to compare with the superbounded calculation, including WCOBRA/ TRAC code uncertainties.

! In addition to the conservative estimate of the 95th percentile, another, more realistic estimate of the 95th percentile was made using the lead plant sensitivity studies and the PBNP four-channel core nominal calculation. This estimate is given in Section 7 and indicates that the use of the superbounded l

1stsotomasas 1-3 l

L

I calculation gives a 95th percentile PCT higher than more realistic estimates, .

and substantially higher value of the PCT compared to the nominal or 50th ,

percentile PCT. , ,

The comparison of the Appendix K calculations and the estimates of the 95th percentile PCT, as calculated in Section 7, indicate that the Appendix K calculation yields a PCT even higher than the conservative estimate of the 95th percentile PCT using the superbounded calculation as a basis. The Appendix K calculated PCT is also less than the licensing limit of 2200*F, such that the intent of the SECY-83-472 calculational approach is satisfied.

s' 4

e f

. e 4

isiscion12 :: 1-4

- 1-3. REFERENCES

. 1. NRC Staff Report, "Emergency Core Cooling System Analysis Methods;"

. USNRC-SECY-83-472, November 1983.

2. Hochreiter, L. E., Schwarz, W. R., Takeuchi, K., Tsai, C-K, and M. Y. Young, "Westinghouse large Break LOCA Best-Estimate Methodology, Volume 1: Model Description and Validation," WCAP-10924-P, June 1986 (WESTINGHOUSE PROPRIETARY).

3. Letter from A. C. Thadani, Assistant Director fer Systems, Division of Engineering and Systems Technology of the office of Nuclear Reactor Regulation, to W. J. Johnson, Manager of Nuclear Safety "Accrptance for Referencing of Licensing Topical Report WCAP-10924 "Westingh>vse large Break LOCA Best Estimate Methodology," August 29, 1988.

4. Letters from W. J. Johnson to M. W. Ho.!ges, Reactor Systems Branch USNRC, dated November, 1987 "Responses to NRC Question on Westinghouse large Break LOCA Methodology, WCAP-10924, Volume 1," NS-NRC-87-3279.

5. Hochreiter, L. E., Schwar:, W. R., Takeuchi, K. Tsai, C-X, and M. Y. Young, "Westinghouse Large Break LOCA Best-Estimate Methodology, Volume 1: Model Description and Validation, Revised Code Predictions and Code Uncertainty " WCAP-10924-P, Volume 1, Addendum 1 April 1988.

6. Hochreiter, L. E., Schwarz, W. R., Takeuchi, K. Tsai, C-K, and M. Y. Young, "Westinghouse Large Break LOCA Best-Estimate Methodology, Volume 1: Model Description and Validation, Revised Appendix B: Heat Source Models," WCAP-10924-P, Volume 1, Addendum 2 July 1988.

7. Hochreiter, L. E., Schwarz, W. R., Takeuchi, K. Tsai, C-K, and M. Y. Young, "Westinghouse Large Break LOCA Best-Estimate Methodology.

Volume 1: Model Description and Validation, Supplemental Information on

. WCOBRA/ TRAC Uncertainties, CCTF Four-Channel Core Analysis, and Responses

• ~

to NRC Ouestions," WCAP-10924-P, Volume 1. Addendum 3, July 1988.

1:1sv.1o/112988 1*5

8. Dederer, S. I., Hochreiter, L. E. , Schwarz, W. R. , Stucker, D. L. , Tsai, .

C-K., and M. Y. Young, "Westinghouse large-break LOCA Best-Estimate *

,f Methodology, Volume 2: Application to two-loop PWRs Equipped with Upper ,

Plenum Injection," WCAP-10924-P, Volume 2, Revision 1 April 1988. .

9. "Responses to NRC Ouestions on WCAP-10924-P, Volume 2," WCAP-10924-P,  ;

L Volume 2, Revision 1 Addendum 1. July 1988.

i

[

I l

I l

t l

l i

l l

l l

l t

f l

f f

.. I f

1816v 1o/112848

. 1*6

. SECTION 2

.. WCOBRA/ TRAC N00EL OF P8MP 2-1. INTRODUCTION The best estimate thermal-hydraulic computer code which is used to calculate realistic fluid conditions in the PWR during blowdown and reflood of a postulated large-break LOCA is WCOBRA/ TRAC. The code was orginally developed for the NRC at Battelle Pacific Northwest Laboratory (1) under the name of COBRA / TRAC and then modified at Westinghouse. The code name was changed to WCOBRA/ TRAC (2).

COBRA / TRAC is basically a combination of two codes, COBRA-TF and TRAC-PD2.

The COBRA-TF computer code uses a two-fluid, three-field representation of two phase flow. Each field is treated in three dimensiens and is compressible. The three fields are a continuous vapor field, a continuous liquid field, and an entrained liquid drop field. The conservation equations for each of the three fields and for heat transfer from and within the solid structures in contact with the fluid are solved using a semi-implicit, finite-difference numerical technique on an Eulerian mesh. COBRA-TF features extremely flexible noding for both the hydrodynamic mesh and the heat transfer solution. With this flexibility, the wide variety of geometries encountered in components of nuclear reactor primary systems can be modeled.

TRAC-P02 is a systems code designed to model the behavior of the entire reactor primary system. It features special models for each component in the system. These include accumulators, pumps, valves, pipes, pressurizers, steam generators, and the reactor vessel. With the exception of the reactor vessel, the thermal-hydraulic response of these components to transients is treated with a five-equation drift flux representation of two phase flow.

The TRAC-PD2 vessel module has been removed and COBRA-TF has been implemented instead as the new vessel component in TRAC-PD2. The resulting code is

, CCBRA/ TRAC. The vessel component in COBRA / TRAC has the extended capabilities of the three-field representation of two phase flow and the flexible noding.

isistio m ass: 2-1

The code has been assessed against a variety of two phase flow data (2) from .

experiments which simulate important phenomena anticipated during postulated ,.

accidents and transients and has been approved by the NRC for the analysis of .

PWRs equipped with Upper Plenum Injection (UPI) (3). .

2-2. ANALYSIS WETH00 2-2-1. Steady-State Calculation A WCOSRA/ TRAC PWR LOCA calculation is initisfized from a point at which the flows, temperatures, powers, and pressures are at their approximate steady-state values before the postulated break occurs. Steady-state WCOBRA/ TRAC calculations are run for different time increments (0 to 10 secondu, 10 to 20 seconds) to verify that the calculated conditions are indeed steady and that the desired reactor conditions are achieved.

The values which are used were obtained from calculated steady-state PBNP conditions from the System Engineering Departo.ent at Westinghouse. These calculated plant conditions reflect the input parameters such as reactor ,, ,

ccolant pump flens, core power, and steam generator tube plugging levels for the superbounded calculation as well as the nominal conditions so that a j different set of c2ncitions and resulting steady-state conditions were j obtained for each calculation.

_ a,e l

i i

i .

I -

1 _

1 1sisolo/11 ses 2-2 ,

a i

7-While the above mentioned fluid and core conditions are likely to differ

., slightly from plant to plant, the degree to which these parameters are matched

- in the WCOBRA/ TRAC simulation must remain consistent from plant to plant.

- Table 2-1 shows the criteria used in WCOBRA/ TRAC for acceptable simulation of plant conditions. The conditions are ordered in terms of importance to the i LOCA transient (Reference (5) Section 3). First level variables are required to match or exceed, in the conservative direction, desired values to tighter tolerance than second level variables. The basis for the choices made is as follows: a,e l . ..

.l l

l j

t e .

Il16v:1D/112 :: 2-3

Once the vessel fluid temperatures, flows, pressures, loop pressure drops, and .

fuel rod parameters are in agreement with the desired input parameters and ,

conditions and are steady, a suitable initial condition has been achieved for ,

the LOCA transient. .

2-2-2. Transient Calculations Once the steady-state calculation is achieved and found to be acceptable, the transient calculation is initialized. ,

a,c 9 e g 4 %

examm 4

e5 0 iii.<io,s uii. 24

I I

. 2-3. POINT BEACH VESSEL DESCRIPTION An Upper Plenum Injection (UPI) plant basically has the same general design as most other Westinghouse PWRs with the exception of the location of the injection point for the residual heat removal (RHR)/ low head safety injection (SI) flow. For this class of plants, the low head SI injects directly into the upper plenum through two injection ports, rather than the cold leg, hence j the name UP! plant. Figure 2-1 showr, a schs:natic of the PBNP vessel. The RHR  :

I pumps have separate injection lines (not headored) such that the assumption of j a single failure loss of an RHR pump results in asymmetric injection into the upper plenum.

! The detailed description for the PBNP vessel model is very similar to the ,

four-channel lead plant vessel model description given in Section 2-3-4 of Reference (5)onpg.2-20. The changes which were made in .the vessel model from the lead plant to PBNP, reflect the hardware differences between the two

! plants. One of these differences is in the upper internals configuration.

] Figure 2-2 shows the upper internals for PBNP with a flat upper support plate. The support columns are tubes directly above the upper core plate with flow slots between 13 and 76 inches above the upper core plate. Similarly, some of the open holes have free standing mixers (FSM) above them, which are cylindrical collars for 13 inches above the upper core plate. These structures restrict the comunication between the uppe* plenum and core

]

region, and are modeled in detail with WCOBRA/ TRAC. Other open hole locations l on the upper core plate have no structures above them and thus no restriction l to the flow from the upper plenum to the core region below. The guide tubes l

i have an open flow area above the upper core plate which allows for direct 8

j comunication with little restriction between the upper plenum and the top of I the core. Figure 2-3 shows a cross section of the upper internals, at the upper core plate, with each type of structure identified over each fuel assembly.

l 1

l..

1stsolo/112988 2*5

The lead plant (Prairie Island) has an inverted top hat apper support plate -

design as shown in Figure 2-4. (The design of the support columns allows ,.

direct comunication from the bottom of the upper plenum to the top of the .

core region with little obstruction. The guide tube assemblies have slots in .

the lower portion and bolt holes at the upper core plate level which allows liquid to enter or exit the guioe tubes from the top of the upper cors plate level to the top of the laat slot. Other locations in the upper core plate above the fuel assemblies are taken up by open holes, which again allow direct communication between upper plenum and the top of the core.]a,c Other plant differences are listed in Table 2-3. PBNP has a lower power and smaller steam generators, reactor coolant pumps, and accumulator tanks. The basic loop piping is identical for each plant. Figure 2-5 shows the PBNP high head S! injection connecting directly to the cold leg, separate from the accumulator injection. The low head SI (RHR) pumps for each plant connect directly to the vessel, each with their own line. PBNP has upflow in the barrel-baffle region compared to downflow in the lead plant. .. .

. The ECOBRA/ TRAC vessel model is separated into horizontal sections and each "'

section contains one or severai vertical flow channels. Gaps are used to model horizontal flow paths between channels. EachsectionoftheECOBRA/ TRAC vessel model for PBNP is discussed and conpared with the lead plant vessel model in the following paragraphs. Figure 2-t' gives the PBNP vessel noding scheme, and the channel and gap locations. Channel numbers are enclosed in squares and gap numbers are ench sad in circles with arrcws at channel boundaries. Table 2-4 lists each channel number, what it represents, and the verticle connections between channels. Table 2-5 gives gap numbers and the channels that are connected horizontally, a,c neo istertoniassa 2-6

Esc O --

O e

e g 84 e e, O

t e =

and mano 99 tats <.torti ::: 2-7

a,C l l

l

- ) =.

1616v 10/112:33 2-8 ,

i

soc M

0 9

• O 6

O e e, 1

1e i

M e

• mesa l

ista .to/itass: 2-9

S,C lN e

e t

e e

1 1

e 6

e* 4

,e e aumma 4 e l l

ee I

e ists,.to/itress 2-10 l

8,C .

- i N >

e 4

6 I

4- g e

L 1

i I

(

6 7

9 l

i t

i i

r f

t

. *e  !

P L

I e e.

I f

I P

I O

f e a m

46 interio/tinus 2 11

a.c

~~ ---

2-4. PBNP LOOP MODELS - PRIMARY AND SECONDARY l

The primary and secondary systems of the PSNP two-loop PWR are modeled the i

..-1 same as the lead plant with a combination of WCOBRA/ TRAC one-dimensional loop I components which include:

Accumulator Component The ACCUM module medals an accumulator is a pressure vessel filled with ECCS water and pressurized with nitrogen gas. During normal operation each accumulator is isolated from the reactor primary coolant system by a check valve, if the reactor coolant system pressure falls below the accumulator pressure, the check valve opens and water is forced into the primary coolant system.

BREAK Component - The BREAK component imposes a time dependent pressure boundary condition one cell away from its adjacent component. The void fraction and fluid temperature associated with the break are specified.

This is used to model the containmen; oressura transient.

ists..ioetiass: '!-12

• FILL Component - The FILL component imposes a time-dependent velocity beundary indition at the junction between tra FILL and its adjacent

. component. FILLS are used to model ECCS injeck.en and steam generator

. feed water injection.

P!PE Component - The PIFi; component is used to model thermal-kydraulic flow in various sections of the reactor coolant system piping.

Pressurizer Component The pressurizer is modeled by the PRIZER module. The surge line is modeled by a section of a TEE component.

PUMP Compunent - The PUMP component describes the interaction of the reactor coolant system fluid with a centrifugal pump. The component calculates the differential pressure and its angular velocity as a function of the fluid flow rate and the fluid properties. The model also accounts for two phase flow effects.

Steam Generator Component - The STGEN component models the primary side and secondary side hydrodynamics separately, with the wall heat transfer

~

the only coupling betwecn the two sides.

)

TEE Component - The TEE component models the thermal-hydraulics of three piping branches, two of which lie along a common line with the third entering at an angle from the main axis of the other twe.

VALVE Component - The VALVE component is used to model the fluid f.ow in a valve such as an accumulator check valve. The valve action is modeled by controlling the flow area and hydraulic diameter at a given cell boundary in the component.

5

$-4-1. SteaJy-State Loop Model 9

0 m

m e,

toisv.tontants 2-13

~ - - _

l

4 2-4-2. Cold Leo Guillotine Break loop Model a.c ,

f l

i i

l 6

l

=

l l

l 2-14 1816v:10/' 4284

f 2-5. REFERENCES

. 1. Thurgood, M. J., et al., "COBRA / TRAC - A Thermal-Hydraulics Code for

. Transient Analysis of Nuclear Reactor Vessels and Primary Coolant Systems," NUREG/CR-3046, Mar. 1983.

2. Hochreiter, L. E., Schwarz, W. R., Takeuchi, K. Tsai, C-K, and M. Y. '

Young, "Westinghouse large Break LOCA Best-Estinte Methodology, Volume 1: Model Description and Validation," WCAP 10924-P, June 1986 i (WESTINGHOUSEPROPRIETARY).  ;

3. Letter from A. C. Thadani, Assistant Director for Systems Division of Engineering and Systems Technology of the office of Nuclear Reactor Regulation, to W. J. loSnson, Manager of Nuclear Safety, "Ac u ptance for Referencing of Licensing Topical Report WCAP-10924-P, "Westinghouse Large

Break LOCA Best Estimate Methodology," August 29, 1988.

... "Improved Fuel Performance Models for Westinghouse Fuel Red Design and 4.

Safety Evaluations," WCAP-10851-P. June 1985 (WESTINGHOUSE PROPRIETARY).

. 5. Dederer, S. I. , Hochreiter L. E., Schwarz. W. R. , Stucker, D. L. ,

Tsai, C-K., and M. Y. Young; "Westinghouse Large-Break LOCA Best-Estimate I Methodology, Volume 2; Application to Two-Loop PWRs Equipped with Upper 4 Plenum Injection," WCAP-M924-P, Volume 2, Revision 1, April 1988 (WESTINGHOUSE PROPRIETARY).

j'

6. Letter ' rem P. Coddington to TRAC distribution dated April 29, 1983.

l 7. Letter from T. R. Charlton, EC&G Idaho to J. E. Solecki dated May 12, 1982.

I isisaenizes 2-15 i

t i

TABLE 2-1

• CRITERIA FOR ACCEPTABLE STEADY STATE -

l 1

e* ,

e'

• 9

• . l 1

l ee isistiomises 2-16

4

. 1 TABLE 2-2

• PUMPED ECCS ASSUNPTIONS f 1

Offsite l Single Pumps Containment Sprays Case Power Failure Operatina Operatina a,c  !

~ l 1 1 I

I i

i o

b

[

!=

t f

, l r

f f

i.

l 1

I

~

r F

t

' I r

! b

) l l .

j t l

. se i

l 1siscionitses 2-17 i

i i

TABLE 2-3 CURRENT TWO-LOOP UPI PLANT DESIGN AND OPERATING PARAMETERS Lead Plant PBNP Parameter Prairie Island (NSP) Point Beach (WEP)

UpperSupportPlate(USP) Inverted Top Hat USP flat USP l

Typ Fuel 14x14 Optimized Fuel 14x14 Optimized l Fuel Assemblies Assemblies -

Power (MWt) 1650.0 1518.5 l

l Steam Generator Model 51 Model 44F--Unit 1 1 Model 440--Unit 2 Reactor Coolant Pump 93A 6000 93 6000 System Prassure (psia) 2250. 2000, or 2250.

6 6 Containment 1.37x10 1.065x10 3

Volume (ft ) l Accumulater: ...

3 ka:wr Volume (ft ) 1270.0 1100.0 P. essure (psig) 700.0 700.0 Flow Dire: tion in Down Up Barrel-Baffle Region O e 1sts, ion 12:ss 2-18

. TABLE 2-4 8

- CHANNEL DESCRIPTIONS FOR WCOBRA/ TRAC FOUR-CHANNEL PBNP UPI MODEL O9 e 4e e

t 90 ios.:io/11:ssa 2-19

l l

i TABLE 2-4(CONT.)

CHANNEL DESCRIPTIONS FOR WCOBRA/ TRAC FOUR-CHANNEL CORE UPI *) DEL a,c .

l

~

• .I i

K i

1 i

.1  ;

i l f i

1 J l 2 ,

l I r 1

I 3

1

.* - l t s J i i t 9 f l l i

l

, f I

l i f l

1 a

i f -

i e.

e isisv; ten 12 ss 2-20

. TABLE 2-5

. GAP CONNECTIONS FOR WCOBRA/ TRAC FOUR-CHANNEL PBNP UPI N00EL ,,e

• 4

. *e 4

l e

• m uma isisvaonizess 2-21

i l

l l

___ f _,, l

..-,g '

M N I( ,,, j, I  !

1 x

.. - g s .,. .

t

, '/ .  % ,

• / %q 1,( 2 (

.e/

! ), * '

it l'

i b

i

'h a _ d lil 1 .

J...! !,. _.

7,h s--l  ;

~ ' s

~I I

) ,', ,

e e% **

- j I,' i i= r

,7< p1 , M~, _%

p

-s 4., .= _

S ,  ; h'

? I' -,' '

4 !l R / 3- .

O $, u n- ,

, [

.< , e ..p L

i e l .

'I '-

WW I

. N

. id b^M $

Dririgf5McTfr[h

5 7 - _

m [=

{l , ,,

,, , b) !I ,.

] i i TW 7 LLJ l ,r 2 4 .

'/ -

N 19 /

N 5 s @. Hb- ,,

Figure 2-1. Schematic of Point Beach (WEP) Vessel ,,

i 2-22

p 7gl7s{c-%,

. _.I _) Il T) l jfp'I' FLAT UPPER SUPPORT PLATE 5  %}d (

cz . __

~

g 4=  : =

j

-hf 1 y

n

,s.i , = ,  ;,

r" 4 3 .

f, ),:

-- m -

[, h, 9=i $N e - ,

SUPPORT COLUMNS N

N 1,Tt- -

ll r-ll

%s l =

N imi = l ..

GUIDETUBES .[ _

p + .l ,i J e

N

[

fr. s s 4

)

N imi

)l~ : s Y.

j

=r ,

'[ i h

i

" I ~

1:  :

j -

,ini 2

i i ,

,, .e I r- p.

J l r HOT LEG N0ZZLE  : ,ius  : e i !q y

. . . l %'l

! Ei t : 1 o r _

a-

ph Iu n str ' ' h,jk ;W

. a h- -

s

' / . -- _- .

6 .gb' L i

\ ' -

d  ! r i.3, .

a

' i m . . u! i 6 --

s - -  :

FREESTANDING MIXER- .g

.~.

eL r.

laq JL M 'M(

f

, s ,

I ._4- . - . h. y (l

..m.

i UPPER CORE PLATE #E

• v L"7 r- , ,i US l \ {, c

E , , , ,

' ., ,, g p r r . vm yc i n o, ,j h.

" MMMithin&0ijH!JE'jg.l l l

, .~

Figure 2-2. Section View of Upper Internals for Point Beach (WEP) 2-23 l

l

i

/ -

-l N

~'

/ . '

/ .,H.3*

{rg}) \'

\

/

~

9 T p '

, h*, ,

. .... 1. ':

-~ -

I - -

f. - - - / -

x s .. .

M(og,'tQ[ 4ge M A = GUIDE TUBES (37)

B = OPEN HOLES (34)

C = SUPPORT COLUMNS (32)

D = FREE STANDING MIXERS (18) ,

Figure 2-3. Cross-Section of PBNP Upper Internals -

2-24

g . .g . . .

/ // <

o<- d- - --

. / // l- ['"I

]

, / . .

o ,n . - u: h INVERTED TOP HAT q - / ' 11 G _

UPPER SUPPORT PLATE S K. r , p , e n ,, , cQ;6 , . .

WiA% 'ip v Q\ ,

/s  ; o , 4

y. .-

c ,p

\M '

N ' , ,8 "uu t _ N5 ' ' ..h.t. I E SUPPORT COLUMNS i ', , ,

GUIDE TUBES di i ,, y ', l}, - d D-h MN N N-s s l

% . l Q l d '

a: b,:

NN i ,i a .

I I?x x x i , l l

., k Q ' + +'hl ';'

- i

'ff HOT LEG el l l . . ,

N0ZZLE . ,

m , ,

1 , , .

I  : , i 1 i ,

e i I' m ' = '

(y N 'l ,

i l l  :

N i=

k l, f 4L =

V UPPER, CORE PLATE -

\\ x' #n3 -*

d- '; '1}' &'

u s, ap'4 .,,

Q .%m; (' W %gfy a s c )F!FipirFl6mr.181[' i

$ pip h, l If ,  !! II Figure 2-4. Section View of Upper Internals for Prairie Island (NSP) 2-25

b F 3 l l

i L

ACCtM)LATOR t TANK ,

HIGH HEAD SAFETY INJECTION f t

v > L

~  % i

1. ,N  !

i i

r  ;

i COLD LEL

[ [

L I

s i

i1 l

i I+

Figure 2-5. PBNP Cold Leg Safety Injection Piping

. i f

2 26 1

a,c e

~

Figure 2 6. PBNP 4-Channel Model Vessel Noding I

2-27

a.c . .

Figure 2-7. Section 1

_, a c .,

l Figure 2-8. Section 2 ,,

f 2-28 l

4C

= -

f Figure 2-9. Section 3 i

Four Channel Core Model

.ac i, . .

I i

4

' Figure 2 10. Section 4 (CCFL Region)

.e 2-29

8,C 0

4 8 e

O l

I e

e#

• I l

I 1

f I

i I

I f

I I

\

l

= , l 1

1 \

<_. a Figure 2-11. Detail of the Low Power Channel CCFL Region (Section 4) f I

2-30  ;

I t i i-

t p

ac f

2 I

I I i I

t I

i Figure 2 12. Section 5 1 r

1 a.c  ;

1

< - I i

.. l t

r il 4 . ..

i i

r l

i n

l

l t

] I

~

I L

I t

i, P

r

- r i

Figure 2 13. Section 6  !

g i 2-31  !

i i

i

l l

l ac ,

l i

.t i

Figure 2 14. Section 7 ..

, isC

~ .

Figure 2-15. Section 8 .

2-32

- - a.c I

l Figure 2-16. Section 9 e

e **

t I

l l

B E

e w

e So 2-33 i I

-_,3---.7.,.__--_____

5

- 4eC 9

e 1

. . j l

l

\ \

t

.t t  !

a ,

?

N 3

+ i e

t

. I J

i  !

N r I

Q V l

t L

i i

}

. . , mb L

• l l

essor.tenasses 2 34 r

L r

l 4.c l

i i

J r

i 4

h F

i i

! I i

... k t

!' [

l Figure 2-18. Intact Loop Hot Leg Ceeponent 2

., a , c ,

4

.. l 2

I i t

a f

l i i ,

I E

I e

L i

l 4 i ,

Figure 2 19. Steam Generator Components 3 and 15 f

> t

.. t i

> sektonuses 2-35 .

\

, [

i i

f f

m . - - - . - . - - - _ - - - - - - - - ----_---- _ -.----_

1 1

1 l , a.c l

I i

l r

i Figure 2-20. Loop Seal Components 4 and 16

- "'C ..

t e

.. . ! i i

i i

l 1

Figure 2-21. Reactor Coolant Puse Components 5 and 17

~.

aswaonuses 2-36 l

l i

a.

Figure 2 22. Intact Loop Cold Leg Component 6 a.c

~

(.

<- 9 Figure 2 23. Broken Loop Cold Leg Components 18 and 20 een .tonanes 2-37 l

- a ,t  ;

i L

e .

L l

r t

rigure 2-24. Pressuriser. Component 14 [

a.c  !

L e* ,

i, t

t i

k r

I ri,,.2.n. ., . t ,woit.,.a ,,...,..,so u c o a ,n -

~

l 1

l eswetonaams 2 33

{

yy - - - - - - - - - _ _ . _ _ . . _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _

f 4C

[

?

i e I o e s O

e 4

h I

F a

I f 1l

.t  :

I P

• e

- 3 '

l a

I l e **

Y t w

. +

i l

f

, I

- l t

f Y

[

2'39 I f

00h10/183904 $

I

-m _*g'..y% *,--

c.

Figure 2-27. Intact Loep Cold Leg Cold Leg Guillotine Break a.

r .'

l Figure 2 28. Broken Loep Cold Leg Cold Leg Guillotine Break ons, sontassa 2-40 1

ac Figure 2-29. Area Expansion at Last Cell Before Dreak *

.. 2~4's 00281.10nstsis

• SECTION 3

, , REALISTIC PWR CALCULATION AT NOMINAL CONDITIONS

. 3-1. INTRODUCTION As discussed in Section 1, the SECY-83-472 (1) document recommends two

r. 'stic calculations of a PWR LOCA transient. A nominal calculation at the E, , probability level and a more conservative calculation at a 95% probability level. The real'stic calculation at nominal conditions represents the more probable operating conditions for the reactor at the time the postulated LOCA oCCdrs.

3-2. PBNP MODEL AND LOCA BOUNDA3Y CONDITIONS FOR NOMINAL CALCULATIONS Unlike a traditional LOCA analysis in which all of the uncertainties were compounded in the analysis, the rsalistic calculation at nominal conditions

. attempts to use the nominal or best estimate plant operating and initial conditions with some models and code inputs restored to their nominal or best estimate values. However, there are still some model, plant, and accident boundary condition parameters which are kept at a more bounding condition.

(The reason for using the more bounding assumptions in the nominal calculation is that not all parameters have a frequency distribution that permits use of the mean or nominal value, and also, selection of more bounding assumptions in selective cases permits direct comparisons between the nominal and bounded cases. One example of a more bounding assumption used in the nominal calculation is the loss of one RHR (UPI) pump. Table 3-1 list; the key input model parameters and their values for the realistic calculation at nominal conditions.]a,e 3-2-1. System Pressure a,c o

1816r10/1128s8 3-1

Core Fluid Temperature a,c 3-2-2. ,

3-2-3. Weactor Upper Head Temperature a,e i

I l

i 3-2-4. Reactor Coolant Loop Flow Rate ac i

1 i

+

l

.i "..

.i t

1 l

=

1 4 M l

]

[

t I

isist.ioniasse 3-2 i

I i., . - .. ._ . - . . . . , . . _ _ - . _ . - _ . _ _ - _ _ _ _ _ - . , , _ - - - . . _ . _ - - _ - . . _ _ _ _ _ . _ . __

(

l

• 3-2-5. Accumulator Water Volume a,c

l. .

I

• 1 1*

i I

3-2-6. Accumulator Gas Pressure a,c 3-2-7. Accumulator Water Temperature a.c 3-2-8. Safety injection Flow a.c f

s.

1stevnentases 3-3

a,e i

(

t i

l 1

j i

r t

i  ;

i i .

I 4

4 - 1 J

)

I 3-2-9. Safety injection Delay Time a.c l

? ._,

)

i 5 I

,I t

. i i

a  :

i .t

_. t

[

4 {

+ tsisotoniasse 3-4 r 1 i Y

I (

i

- - - - - . . -t - _ . - _ _ _ , . - . . , _ _ , - . _ _ . _ , . . . _ - - .

ac e

3-2-10. Safety Injection Water Temperature a,c -

l l -

l 3-2-11. Steam Generator Tube Plugging Percentage a,c l '~'

l I

3-2-12. Containment Pressure a,c m

4 t -

e So isist.io/11:ss: 3-5

a.c l

t 3-2-13. Initial Fuel Rod Temperatures a.c I 1 ,

1 1

l 3-2-14. Peak linear Heat Rate (PLHR) and Hot Rod Power Values a.c M

D e.E ge isis .ioniance 36

a.c

\

+.

,, 3-2-16. Core Power Uncertainties a.c 3-2-17. Rowet Option and Uncertainties a.c MI 3-2-18. Decay Heat and Uncertainties a.c

=

3-7 il16v;10/112088 s

1 3-2-19. Cladding Reaction Rates and Uncertainties ac.

3-2-20. Break Size and location a,c 3-2-21. Steam Generator Conditions ac M

9

- i e

~~

l totstiontassa 3-8

3-3. RESULTS OF PBNP REALISTIC CALCULATION AT NOMINAL CON 0!TIONS

~

.- 3-3-1. P8NP Nominal Condition Steady-State Results The nominal condition steady-state calculation was performed to achieve the desired vessel fluid temperature, flows and pressure drops before starting the transient calculation. Using this steady-state plant condition as an initial condition, the PBNP nominal conditions transient was performed.

~

I t

3-3-2. P8NP Nominal Condition Transient Results l

^ '

The transient calculation for the 0.6 DECLG calculation at nominal conditions has been divided into three phases. The time periods for each phase are shown on the cladding temperature plot in Figure 3-1 and are as follows:

Phase 1 - Blowdown --- 0-22.5 seconds of transient Phase 2 - Refill --- 22.5-43. seconds of transient Phase 3 - Initial Reflood --- 41-63 seconds of transient Phase 4 - Final Reflood --- 63-100 seconds of transient Phase 1 is characterized by rapid depressurization as seen in Figura 3-2.

Rapid voiding in the core results in an early departure from nucleate boiling (DNB) and rapid cladding heatup. The core power shuts down on voiding in the

+ a, without use of control rods. During this period, some of the upper head water drains down through the guide tubes (GT) and a portion of this goes directly to the core, keeping the sverage rod (Red 3) cooler than the average rod (Red 4) under the combined support columns, free standing mixer, open hole channel resulting in a lower blowdown peak temperature for GT assemblies. The lower power channel, which generates less steamflow, receives a larger fraction of the downflow during this period. The flows into the top of the individual core channels are shown in Figures 3-3 to 3 6.

- By 5 seconds, the water in the upper head has drained past the top of the e guide tubes, but the upper head continues to drain through the gaps in the

~

side of the guide tubes. By this time the break has dominated vessel flow and e.

Illtrio/112 ss 3-9

caused core flow reversal, as seen in Fiy es 3-3 through 3-6,* which turns .

around the peak cladding temperature (PCT). This results in a hot rod , ,

blowdown PCT of 1312'F at 7.75 feet and 5.75 seconds. .

Up to 6 seconds in the transient, the flow down from the upper plenum to the core fuel assemblies under guide tubes (channel 11) continues to be proportionally higher on a flow / assembly basis than the flow from upper plenum to core assemblies under free standing mir.ers/ support columns /open holes (FSM/SC/0H) (channel 10). This is a continuation of the flow momentum begun by the draining of the upper head through the guide tubes. The break flow rate continues to decrease, which ir turn decreases the negative core flow l

rate and causes the hot rod and hot assembly fuel rods to begin a gradual heatup after 15 seconds. The average rods initially begin to heatup until the flow reversal which cools the red down to the saturation temperature at 8.5 seconds into the transient. During this time the accrmulator and high head safety injection flows have begun, by 7.5 seconds, but do not reach the vessel downcomer until about U seconds. By 15 seconds, the RHR injection to the uoper plenum has begu'4. The high head safety injection flow is shown in -

l Figure 3-7 and the accumulator flow is shown in Figure 3-8. Tho UPI flow is

~

l shown in Figure 3-9 with one RHR pump injecting.

In Phase 2, the accumulator water reaches the downcomer and begins to flow down as well as bypass around the downcomer and out the break. During this j time both the average power rods and low power rods remain quenched (in nucleate boiling), while the hot assembly and hot red slowly heatup. UPI water coming into the upper plenum continues to preferentially flow down the l low power channel (channel 13), which has less steam generation. Some UPI l water flows down the average channels to maintain cooling of these rods. By l 25 seconds the lower plenum begins to fills up with liquid, as seen in Figure j 3-10, signifying the end of bypass. The lower plenum is filled by 41 seconds. I l

The accumulator and high head safety injectio. flow fill the downcomer, the j UPI flow also penetrates the ca o primarily through the low power channel and .

helps to fill the lower plenum. A liquid level builds up in the upper plenum

~

frem the UPI injection. There is sufficient liquid and entrained downflow -

I isis,iomasse 3-10 i

l l

l 1

down the average power charnel 11 to keep this rod in nucleate boiling. The

. - average power rod in channel 10 (FSM/SC/0H) is at slightly higher wall temperature. There is councerflow in the hot assembly which provides

- sufficient cooling to limit the temperature rise of the hot rod during this time period.

At 43 seconds, the core begins to fill from the combined accumulator /high head safety injection, and UPI flow and the initial phase of reflood begins. Plots of the core liquid levels are shown in Figures 3-11 to 3-14, and show the core beginning to fill. The downcomer level is shown in Figure 3-15 and indicates it fills at the same time. The UPI flow creates a liquid level in the upper plenum as shown in Figures 3-16 and 3-17 as the UPI water accumulates on the I

upper core plate and then penetrates into the core.

Towards the end of the initial reflood phase, the downcomer has filled to a high enough level that it can force water into the core. This results in

.. oscillations between the core and the downcomer as the core refloods. The oscillations promote improved cooling as the water surges into the core. The

. .. , reflooding behavior terminates the temperature rise of the hot rod and hot assembly rod such that the reflood peak cladding temperature is 1382'F, at 6.0 feet and 47 seconds into the transient. The core average rods which have begun to heatup slightly during this time period are quenched and remain in nucleate boiling. The lower power channel rod also remains in nucleate boiling as seen in Figure 3-1.

At 63.0 seconds the accumulator empties and the nitrogen injection from the accumulator forces even more water into the core. The core end downcomer levels oscillate as the core continues to fill. As the co fills, the het rod and hot assembly rods completely quench.

SG tamiomass 3-11

_ . . _ 9

1 3-4. REFERENCES

1. NRC Staff Report, "Emergency Core Cooling System Analysis Methods," .

'USNRC-SECY-83-472 Novether 1983. ,

2. De1 Signore, T. A., et al., "Westinghouse ECCS Two-Loop Plant Sensitivity Studies (14x14)," WCAP-8854-P-A, May,1977 (WESTINGHOUSE PROPRIETARY).

3. Dederer, S. I., Hochreiter L. E., Schwarz, W. R., Stucker, D. L.,

Tsai, C-X., and M. Y. Young; "Westinghou'se Large-Break LOCA Best-Estimate Nethodology, Volume 2; Application to Two-Loop PWRs Equipped with Upper Plenum Injection," WCAP-10924-P, Volume 2, Revision 1, April 1988 (WESTINGHOUSE PROPRIETARY),

4. Cathcart, J. J., et al., "Zirconimum metal-water Oxidation Kinetics IV -

Reaction Rate Studies," ORNL/NUREG-17, August,1977.

.)

I 1

l l

'l 1

.l isis <.to/tians: 3-12

. , 1

, J 4

TABLE 3-1

. . REALISTIC TWO-LOOP PWR a,c CALCULATION AT NONINAL CONDITIONS m e

s to 0

9 e

m g e,

1stev;1oniass 3-13

,y i

t i l

TABLE 3-1(CONT.),. -

3 ,

REALISTIC TWO-LOOP'PWR ,

• I t

CALCULATION AT NONINAL CONDITIONS a.c ,

t l

i i

k I

i i

'i

[

t i/

1 W

I

}

i e* ,

t l

l l

l

[

I i

i e~

h

[

}

\

es t

i Islev:10/112:ss 3-14 l l

'i

. l s v.

[, .gs ., ,

4

~

8sc g e 6

f h>

g 3

BO 8

d W

e L d 6

E 3

2 4

b 2

5 I

A 2

8, C ..

e b

4 3-15

4 e

4,C

)

e e

t 3

e t

3 .

I .. .

9 N

e M

8, b

I i

eG G

mas W l

1 l

l t 3-16 l.

O O

4,C

. e - -

e a

P to 4

W C

5 r.M

• p=

E

\

F 1

2 m

i.

t b

E E

l 8

a b

1 a

2, O

se C.

E C

e m

8.

b D

M

4. I l

l 3-17

4 8:C O 9 e

O h "??

4 4 0

W M

A 1 bD i W  !

T.

or D

4.3 0

C .e 4J

• O u

C e

3 o

W b

Y e

M W

L 3

CD e

6 e4 e

M h

3-18

h e

4,C O

I 8

t P

1 a

a . .

1-S 3

E c

u; A

i

=

D

1. 9 hue 3 19

I~

4 .,

J i.

Y, aC r,

y ,

i, '

k

~

C 2

E F

1 3

m m

!e .

eb g

a b

E a

a 2

B c

A E

8, c *

& 9*

w 3 20

I' e

O 8,C

. e -

mm e

6 e

a T

w a

a og

• 9 O e  % 8 c

m E

b I

M b

8 a

i 4

I i

l i-I r .

• r

.I

e.

• f l

3 21 l

4 l- .

.'i,.p!

4 nC e . .

N 4 e

a T

8 a

U 3E m

O

+d a.Ee e*

• e B

Ide b

O so 4

U N

. t 2

8, b

l

. 1 1

a e l

a w

N 3-22 a- _ _ _ _ - - - - - - - - _ _ -

J a 4 ~ c&,s-_+ $ w u aa-- 1 &

4.

T i

f 6

4,C e

0 e

4 1

m 3

i. -

b I

a i *e b O

i e C

^

w S

J w

.) e i

e l M 1

t*

i l**

l

! W l

)

1 J 3 23 1

1

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

v-w -w------------

E

( m o

e Bec I . .

o 4

i

=

i 1

w e P

W w

3 O

Y m

k C

e 4

L. .

3-24

a 4 9

9 8C

• O 9

+

e F'

I 8*

a" b

Y a<.

6-

~g a~

b.

=e E

r=

23 c1

'3 3*

a8u

.J W e

pug m

b C

am 6

M to b 3-25

I.

s G

e SeC

~

~ . ,

e D

6 O

I 8

e b

1 a

b M

4 9

2 *

.8 C ** e w

> g W

J

'O l 3 N

Y m

+

8, I

4 i

se e

• oe 3 26 i - -

C)

'f  !

~

1 n,

e soc e 8 j I

e l

l.  ;

-/ ,I f

(

1 g  ;

31  !

1. 3 i Et l t8 >

l2

~

8C

• e i e

• t 2n

~<  :

C@

e-

~

.2

>C as -

S 2.i O$

O M

i M

G b

b 6

M O

to m

3 27

n. ,

1-1 s

4

(*.

4,C 4

4 4 ,J' j' 5' r G

e W

• 4 4

t t.

s I

2 C

• e W

h **

> . r G 1 i i T i e

t 4

e t

' i j g M

1 M

i h r

' I

_E i N

I 1

, I f

f i

w at I

3-28

h 4

4,C

?

t E

e

.J a

.e.

J

• e e b

~

e N

m O

a M

b 1

1 i

e

!.- i f

.r

e.

• a a

3-29

mm S

4,C O

e

'i e

8 I

N 4

I L ,s 9

i .a g

1 J

m M

t .

l e

~

f a, e .I I N f I & I I >

J i I

i I

h t

a huse .e i  !

i 3-30 i ,

! I v---- + -- ,. .

ww- e n-- , ,,---g,.,,-e- w--- -e-,,m-.m---.e,--em-,m,gw-,v.-o--, -c -m.a.-n--.w., _.e,c-- w.-++-----.,----y- , - - - - ,n-~--

S 4C 6

P*

ia

.G.

.. z l

i m

e

'a e

b C

e M

9e W 3-31

SECTION 4

. . REALISTIC PWR CALCULATION AT SUPERSOUNDED CONDITIONS

. 4-1. INTRODUCTIOh As discussed in Section 1, the Westinghouse approach to the SECY-83-472 (1) realistic LOCA analysis is to use a more conservative, superbounded but realistic calculation which would yield a peak cladding temperature, PCT, which is higher than the 95% probability point. This calculation will yield PCTs greater than the 95% probability limit Secause the uncertainties in the key LOCA parameters such as hot rod power, stored energy, total core pewer are combined together in a single calculation rather than in a statistical fashion. The frequency of plant operation at these conservative bounding conditions is negligibly small.

4-2. PBNP, MODEL, AND LOCA BOUNDING CONDITIONS FOR THE SUPERBOUNDED

., CALCULATION Unlike the nominal calculation, the superbounded calculation places many plant

~ #

parameters which have a known penalizing effect on the calculated peak cladding temperature at their upper bound. Also, selected models and correlations are placed at their bounding values so as to penalize the calculated peak cladding temperature. Table 4-1 lists the key WCOBRA/ TRAC input and model parameters that were used in the superbounded calculations.

Each parameter, its value and uncertainty will be discussed below. a,c W

System Pressure a,c 4-2-1.

m a

O eM m.m 1sisv ien12:ss 4-1

r 4-2-2. Core Fluid Temperature a.c '

\.

L 4-2-3. Reactor Upper Head To u rature, a.c t

4-2-4. Reactor Coolant loop Flow Rate a.c t

a toisv:tonites: 4-2

a,c 4-2-5. Accumulator Water Volume a.c 4-2-6. Accumulator Gas Pressure a.c 4-2-7. Accumulator Water Temperature 4-2-8. Safety In.iection Flow a.c

~

r 1818 v.1D/112848 4*3

--_..----- .__________________ _ ____J

e,C Safety injection Delay Time a,c 4-2-9.

l 4-2-10. Safety injection Water Te m rature ac

.* l

'l

~

l 4-2-11 Steam Generator Tube Pluccina Percentage ac

{

f 4-2-12. Containment Pressure ,,e taiscioniassa 4-4

a.c 4-2-13. Initial Fuel Rod Tosceratures a.c e

t l

1814v;10/112888 4-5

a,c e

4-2-14. Peak linear Heat Rate (PLHR) and Hot Rod Values ,,e e

m e

(

s eiso om a.. 4.s f

t 1. .

ac 4-2-16. Core Power and Uncertainties a,c 4-2-17. Rowet Option and Uncertainties ac

~

i J

i i 4-2-18. Decay Heat and Uncertainties a,c 1

l i L i

, 4-2-19. Claddino Reaction Rates and Uncertainties ,,e l

l

}

$ w f

I 1 ~

l iii.a onia.. 4-7 l

1.-_. .

ac 4-2-20. Break Size and Location a.c 4-2-21. Power Shape aC 4-2-22. Steam Generator Conditions a,c 4-2-23. Pump Characteristics a,c em

~

l j 1616v:10/112taa 4-8 l

l

. 4 2-24. Accumulator Nitrogen a.c 4-2-25 Upper Plenum De-entrainment ,,e 4-3. PBNP SUPERB 0VNDED CALCULATIONAL RESULTS 4-3-1. PBNP Superbounded Steady-State Results The superbounded conditions steady-state calculation was performed to obtain

• the desired vessel fluid temperatures, flows, and pressure drops before starting the transient calculation. The steady-state criterion established in Section 2 were compared to the PBNP WCOBRA/ TRAC output to insure that a valid steady-state had been achieved. Table 4-2 compares the desired values of the key parameters with the WCOBRA/ TRAC calculated value for the PBNP superbounded conditions. As indicated in the table, the key parameters are within the accaptable steady-state uncertainty range for PBNP. Using this steady-state plant condition as an initial condition, the PBNP superbounded transient was initiated.

4-3-2. PBNP Superbounded Transient Results a.C

~ ~

Il16rlo/112848 4*9

_ _ _ _ _ _ _ ______________A

, t 1

4,C 1

" i

. l T

. . [

i i,

4 I

l r

I s\

h p

i i

I 6

I t

i P

l l 0 1 \

l .*

\

I t

l  !

I f

t f

I r

i i I e

L t

i i ,

l m

I en

~

i

, ** t i r i f I

f 1616v;10/112644 4*10 l i t a

F-4 e

4,C 9

• e e

3

+

9 e

e G

O m

1sisv.1on12:ss 4-11

f. i 1

r. .

4,C l

\

I .

f t .

l i,

a iI

,f f

.\

m l l

l

~ l

<1 1816v.10/112944 4*12

. 4-4. REFERENCES

. 1. NRC Staff Report, ' Emergency Core Cooling Systems Analysis Methods,"

USNRC-SECY-83-472, Novesber, 1973.

2. Dederer, S. I., Hochreiter, L. E., Schwarz, W. R., Stucker D. L.,

Tsai, C-K., and M. Y. Young; "Westinghouse Large-Break LOCA Best-Estimate Nethodology, Volume 2; Application to Two-Loop PWRs Equipped with Upper Plenum Injection," WCAP-10924-P, Volume 2 Revision 1. April 1988 (WESTINGHOUSE PROPRIETARY).

3. "!mproved Fuel Performance Models for Westir.ghouse Fuel Rod Design and Safety Evaluations," WCAP-10851-P, June 1985 (WESTINGHOUSE PROPRIETARY).

4. Carthcart, J. J., et al., "Zirconium metal-water Oxidation Kinetics IV

- Reaction Rate Studies," ORNL/NUREG-17, August, 1977.

b

~

1sisv:1o/1129sa 4-13

. . . _ . _ . . . _ _ . . _ __ _ _Q

TABLE 4-1 ,

PLANT SPECIFICATIONS FOR REALISTIC TWO-LOOP CALCULATION AT SUPER 800NDED CONDITIONS a,c * .

9 l ..

9 luus S4 1sisv;10mtsu 4-14 l

1

, TA8LE4-1(CONT.)

PLANT SPECIFICATIONS FOR REALISTIC TWO-LOOP

,. CALCULATION AT SUPER 800NDED CONDITIONS ,,c e

s

• e o

ese h

e to isiscioninus 4-15

i. .

l TABLE 4-2 '

STEADY STATE RESULTS FOR FOUR-CHANNEL SUPERB 00NDED CiLCULATION l 8C ,

l ,

m e

l l

6 9

4 e.

eunes 9

9 ee isier.ioniassa 4 16

5et er O e

O 8

z*

5 a

i !

, l t, ,

e

.. i 8

4

~

.h 3

.F a

e 8,

c 4

e emuumW w

44 4-17

aC e

4 f

b

$ g e

• L l

(

b r

a f

I I

f 6

Y l .

1 b f

r l

l .I.

t j

. i l >

g k 1 o ..

t I

l

.I I

. I N t e i

• l r

r I

ep.

I I i

?

I I

t l

es a . .

.. l 4-18 .

\

, , - - - -- ., . - . , . , - . . , - - , - . , - - - . . _ - , , . _ . , - . , , - . . . ,,-- _ - ,_,.-- , ,~...----,--,n--. --,..,.-r,--

• 4,C

, ='

b

• e

.9 3

c >

z-

_a l ,

g

$b E1 x

2 2

- 43 y,4 h

l tj 2-88 et 4

E 5,

C 9

w humusse em 4-19

P i

4C .

I e

e i

f h  !

4 '

r 3 . .

C *

<ee b

1  ?

4  !

b, i

% l I L cg l E

S.3 v W

e Y e 8

b We t

W em 4-20 i

4 8sc e

' 6 O

l 3

t g

. 2 5

~

W a

b c

a 4

=

i e

9 huumuue W

6 4-21

___g.

W l

' 4C .

M

' e P

a I

1 a

8 .-

k ..

^

1

• i b l 5 I 1

B C

1 4

=

1 e

I

)

• O 4-22

F

. 4eC De um e

• 9 9

W 4

ed

=

- 8 p

z 1

4

=

i G

D w

G.

4-23

9 soc O

e 4

i D

1 5n.

4 N

e W

L b

I i

l i

1 e !

l M M to 4-24

' .L L

, ':e .

. ~n l

3 e

e .

r F

(

)

0s 1 e l

l o eH n

nn ae hp CO

/

en rm ou Cl nC o

i t

l r eo vp

• , ep i L u S

d/

is ur qe i x Li M

d eg sn pi ad l n l a ot CS 9

4 e

r

_ g u

i F

i s _

l (

t

• lo .v

, 1I i! j f1

O le a

e K

B,C lj -

I e e

h e

O W

.O 3

? Y e

e  % 1 e  !

3 t.D l

~ 1 l

W m

W b l C

C m

.C U

W b

O

, y C

.e.

p e*

. ., U q k U A

g .*

c.-

T C

e CL m

m M

O U

e O

m 0

W W

L 3

CD w

E

/

> r

. i -

M ed

%, 4-26 s _.

0

&,C 9 g e

b A

b t

W 4

9 N

m W l e

4 T

V 6

8 C

.e.

i. ** ,

W .

> i W

d i ea Y

3 cr t Y

CL 4

m O

V e

M M

i W

W L

3 m.

b e

M 4-27

1 e

0 4

8,C e

4 8

C

.o=

1.

O 3

O 3

O a

w O

,==

P O

C C

m

.C U

0 L

O .

U l C

• w l 0

0 J ..

4 D

CT ew V

O e

O.

m r==

.="

O U

e N

m 1

T 0

L D

.e b

94 t

64 4-28

l -

l e

6:

I' IOS

i. .' -

t l

i i

8 O

b W

e C

l W l U W

J T

e O

e

• O

. 4 C

U C

W L

=

O M

0 W

W 6

E e

f h

W M M

4-29 s

.-....is .

8,C e

e e

W D

,C-L 0

se D

O w

P-*

O 0

J V

ese 1 O l

.cr e

\

1 J  !

T '

G l w **

C. l

@ . I 7"O U

E 3

e.. C O

L 0

c.

L 3

Y m

9 4

=

C sr D b

.4 e

SI Q*

4-30

e 0

e 8 s i' = Sec

==

e w

i b

b 3

9 M

t "E b

O

• e W

.. 4

, e I ,s l

I 4 v l

i I

G N

m 1

T W

6 h

4 e

W

1. 9 4-31

t 8,C

• e 6

4 E

b 3

a M

.c.

I a h

T O

U l

C t

.w ,

• 2 .

=

W Q ..

N U

m M

e m

0 W

l l

E l

e e4 h

mummme 9

4-32

_. e a 6 .L-Am+M'

)

• a.c O

e W

3 1

• 4 t

I a

T e

g 5

1 8

~

e O C

t Y

8 1

a j

e b

J 4 "9 W

l i

a f b a

4 k O I

I I*

I

=

• b M

4-33

y -_.

9 O

l;'

• 5eC

• 1

• e e

e I.

b b

O e

b g -

m 9

T b

8, E

e to 9-SS 4-34

e

• ~ &eC e

b 0

J T

e O

(J M

U 90 M

C m

C a

o b

~

s-~ W-m 1

T 0

L D

Cn w

h I

I i

O to e

to 4-35

. SECTION 5 P8NP SENSITIVITY STUDIES 5-1. APPLICABILITY OF LEAD PLANT SENSITIVITY STUDIES TO PBNP a.c s

e 0

a.

W ll2Cr10/112ss: 5-1

0 9 O aC

~

m e

e 0

e e

o e a e

t%

9 0

• m 4 b

iszov. ion 12s:: 5-2

8,C

. m e

e

.4 1620v:10/1129: 5-3

&sC e

e 0

e e*

0 0

se 44 istov:ioniassa 5-4 I

1

I i

e I 1

aC s

• e e

.4

\

IIl0v:10/112988 55

s Boc l

F t nm _

es 1820v:10/112888 5-6

5-2. OPERATING SYSTEM PRESSURE SENSITIVITY STUDY a,e

- M I 9 4

e a

h e

4 m

5-3. HOT ASSEMBLY LOCATION SENSITIVITY STUDIES

..c i _ -

l I

a ,

l "

Y m M

(

tuor.to/itasu 5-7 I

l

&sC e

e 4

O e I e

mO 8

4 7 W*%a 0

m

$ N .e e

-S istodoniassa 5-8 I

(

4 0

8,C m

e e

8 e

O O e

G

'We a e

e 6h is2ov:io/112ss 5-9 s

a,c w - ,

I ,

5-4. REFERENCES

.e e

1. Dederer, S. I., Hochreiter, L. E., Schwarz, W. R., Stucker D. L., Tsai, C-K., and M. Y. Young; "Westinghouse Large-Break LOCA Best Estimate -

Methodology Volume 2: Application to Two-Loop PWRs Equipped with Upper Plenum Injection," WCAP-10924-P, Volume 2, Revision 1, April 1988 (WESTINGHOUSEPROPRIETARY).

2. Debranich, D. and L. D. Buxton; 'Large Break LOCA Analysis for Two-loop PWRs with Upper Plenum Injection", Sandia Report Sand 84-0040 (NUREG-CR-3639) May 1984.

i  !

e 0

r S.

isaotio/iiases 5-10 I

TABLE 5-1

. PRAIRIE ISLAND AND PBNP CORE POWER AND FLOW COMPARISONS a,c r

I L

I

(

i d

l t

4

~

i .  ;

i  !

' b 1 l l

[

i.

f l

. i I

l I j* [

i i 1

4 l

h m m... 5-11 i 1 ,

1 i 1

-, - , __,__,,.-,,,___,__y .y,, , ,,,, ,. ,,,__ ,, _ , _ , _ , , , , , . . , _ _ _ _ _ _ _ _ _ _ . , , _ _ _ , , . , _ , , , , _ _ _ _ _ _ _ _ , . , .._y_g,

- _4mh_ - -

e _ _ - _ ;is-La,=- ___-.# x-w_ as___._,LL_ .,

1. . .,A_ c__ _# t .A,_,__,.mt. 2,h4 .A a.,, p, A- -

• P e

S

.g TA8LE 5-2 SuleNY OF LEAD PLANT SENSITIVITY STUDY RESULTS [1]

F 1 1

+

)

4 I

-i

i I

4 t

i l

9 i

4 is2or.iomzses 5-12 'n l

i l

,J a - .

t

• # 8 . . g h

. ___ _ _m-, . , . - . . - - - - . - . - . . . - - - - - , . - . . . - - - - - , - - - - - . . , , .

-.. ._--- ... - - - - - , - -. . .- . ..n-.. .-.. - - - .- . m - - - .

- , y ,g + a- - .- - -. . . . u - r~ . ..

aya--- . ...

)

f aC s

9 .

r i .  ;

[ a 4

4 M

w d

4 4

w S

M w

N em

.. V.

d M

" On -

t W

k w

I 4

N 5

o w

N s

W 4

89 I

• B 2

m a

l a,c' [

L- i I

i i

l 6

t i

I l

.* r I

i e

%_ ~

ge Figure 5-1. Axial Power Shape Used in PBNP Analysis 5-14

= ,

e.c

. ~\

,i

't t,

d i

J ji i

4 L *.

i-

]

^

i 4

t l

1 i

l

{

i i

l i

1-Figure 5-2. Lead Plant Axial Power Shapes Used l inSensitivityStudies[1]

i l

i i 5-15 l

~ " * - - - - - . n nn. _ , , , , , . _ _ _ , , _ _

a,c -

Figure 5-3. f PBNp Axial Power shape Used joPn Ana ys s with Power Shape 1.imits '.

5-16

i l

i l

C a,c I

)

~

J

=.

4 Figure 5-4. PBNP 95th Percentile PLHR J

5-17

a,e .

l 1

i 4

1 1

t ee Figure 5-5. PBNP RCS PressurJ Sensitivity Study L

5-18

. ac o

• l l

~ Figure 5-6. WCOBRA/ TRAC Plenum Model Used to

- Determine Hot Assembly Location e

5-19

ac Figure 5 7. PBNP Hot Assembly tocation Sensitivity Study '.

1.0 DECL Guillotine Break 5-20

l

. SECTION 6 POINT BEACH APPENDIX K CALCULATIONS 6-1. INTRODUCTION As part of the SECY-83-472 (1) Core Cooling Calculational approach, a calculation must be performed using the required features of the 10CFR50 Appendix K requirements. This Appendix K calculation is then compared to the 95th percentile PCT to assess the margin in the Appendix K requirements. The description and compliance of the WCOBRA/ TRAC UPI model with the Appendix K requirements was presented in detail in Reference (2). WCOBRA/

TRAC was accepted as a valid model in compliance with t.oth the Appendix K rule and the SECY-83-472 approach for two-loop PWRs in the NRC Safety Evaluation Report given in Reference (3). (There were specific Appendix K compliance items which were found not to be applicable to the UPI class of PWRs. Those nctions are: 1.D.3:

Calculation of the Reflood Rate for Pressurized Water Reactors, (the

- carryover portion of this requirement), and I.D.5: Refill and Reflood Heat Transfer for Pressurized Water Reactors. As required by the NRC Safety Evaluation Report, an exemption will be submitted which will provide a technical basis for excluding these specific requirements.

However, the methods used in }{ COBRA / TRAC comply with the intent of the requirements. The other specific requirements of the 10CFR50 Appendix K rule are fully complied with as discussed in Reference (2). The Appendix K compliance developed for the lead two-loop UPI plant (Prairie Island) is also directly applicable to the PBNP Units since the reactors are of the same configuration and design.)a.c 6-2. POINT BEACH APPENDIX K WODEL AND TRANSIENT RESULTS The UP! model discussed in Section 4 was modified to include the Appendix K model changes discussed in Section 7-1 of Reference (2). This revised model was then used to run a transient calculation using the Appendix K requirements for the PBNP.

~ ,

se tszov.tomass 6-1

Table 6-1liststhekeyMCOBRA/TRACinputandmodelparametersthatwere ,

used in '.he Appendix K calculations. (The hot assembly was modeled as a channel under support column type of structure since this was the most ,

limiting channel type, as described in Section 5-3. This is consistent with the four-channel core superbounded model described in Section 4.]a,e .

6-2-1 PBNP Appendix K Steady-State Results As described in Section 2, the steady state calculation must seet specified fluid and core conditions within an acceptable tolerance. The l ,

conditionsandcriteriarequiredforeachMCOBRA/TRACcalculationwere presented in (Table 2-1.)a,c (Table 6-2 compares the desired Steady State Conditions for the Appendix K calculation with the calculated

conditions for the four-channel core model from WCOBRA/ TRAC. The third column shows the amount of variation in the calculated values when compared to the desired steady state condition. This comparison shows that the criteria listed in Table 2-1 are met.Ja.c 6-2-2 PBNP Appendix K Transient Results

)

l e4

=

1820v:10/112988 6-2

t 4,C

\ .

l l .

l l

l t

e

• e e

1820v:1D/112988 63

S,C l

l 1

1 1

1 1

1 1

l l

• l l

• ~ 1 1

' j l

l l

l 1620v.10/112988 6-4

SeC 0

e 4

• e O

e

1. 9

- e-W istov.ie/itasi: 6-5

6-3 REFERENCES

1. NRC Staff Report, "Emergency Core Cooling System Analysis Methods," ,' .

USNRC-SECY-83-472, November 1983.

2. Dederer, S. I., Hochreiter, L. E., Schwarz, W. R., Stucker, D. L., Tsai, C-K., and M. Y. Young, "Westinghouse large-break LOCA Best-Estimate Methodology Volume 2: Application to two-loop PWRs Equipped with Upper Plenum Injection," WCAP-10924-P, Volume 2 Revision 1. April 1988 (WISTINGHOUSEPROPRIETARY).

3. Letter from A. C. Thadani; Assistant Director for Systems, Division of Engineering and Systems Technology of the Office of Nuclear Reactor Regulation, to W. J. Johnson, Manager of Nuclear Safety, "Acceptance for Referencing of Licensing Topical Report WCAP-10924 Westinghouse Large Break LOCA Best Estimate Methodology," August 29, 1988.

182ov 10/112944 6-6

. . TABLE 6-1

• . *. Pl. ANT SPECIFICATIONS FOR REALISTIC TWO-LOOP C.41.CULAT10N AT APPEN0lX K CONDITIONS a,c

~

V 4

0 e

)

4 i

1 1

i

\'

! e l- _

N i.mwmn. s-7

TABLE 6-1(Cont) a.c ,'

i

) _

1 l

l J

I \

l l i

l

\

.o 1 i t

5 1820v.1D/112933 6-8 e

^

. TABLE 6-2

. P8NP APPEN0!X K STEADY STATE RESULTS

. a,e l

.i I

• e G

baum M em e

E4 isaov:io/t n ess 6-9

7*4LE 6-3 TIME SEQUENCE OF EVENTS FOR 0.4 DECLG APPENDIX X LARGE BREAK LOCA .

Event Time (Sec)

Start 0.0 t Reactor Trip ~0.1

• SafetyInjectionSignal 2.1 l

l High Head Safety Injection Begins 7.5  ;

i Accumulator Injection Begins 8.2 ,

i Blowdown Peak Cladding Temperature Occur 8.5 Low Head Safety Injection Begins 15.0 t i I End of Bypass 18.14  !

l i Hot Rod Burst 28.4 -

l l

Bottom of Core Recovery 32.3

~

\ '

l Hot Assembly Average Rod Burst 35.3 ,

Accumulator Water Empty 63.5 l

l System Mass Inventory Equilibrates 83. l Accumulator N Injection Ends 86.

2 Reflood Peak Cladding Temperature Occurs 104.5  !

l I

t I

L I

r

' I t

m istov.io m ass: 6-10 .

l

TABLE 6-4

, FOUR-CHANNEL CORE N00EL APPENDIX X RESULTS Blowdown Peak Cladding Temperature (*F) 1724.

Time (Sec.) 8.5 ,

Location (Ft.) 7.75  !

i l

Reflood Peak Cladding Temperature ('F) 2023.

Time (Sec.) 104.5 i Location (Ft.) 6.25  :

Hot Rod Burst Time (Sec.) 28.4 Location (Ft.) 6.25 f Hot Assembly Burst Time (Sec.) 35.3 Location (Ft.) 6.25 l

% Blockage 54.2 Maximum Not Spot Cladding Reaction (%) 13.24 l Namimum Core Wide Claddi1g Reaction (%) 0.1094 x 10'3 j I

l 1

i i

% i i

e L

isacaonizsas 6-11  !

i

. - . . , . ._,n, - - - - _ _ . _ - - , . _ . _ - . - , _ _ - - - - . , - - _ _ _ , - _ . . - __ -- - - . - _ - - -

CLADDING TEMPERATURE WEP 0.4 APP K-24 ASSMBY-257. SGTP-FO=2.5 ELEVATION 6.25 FT. I176.251HI

-~ ,, ,

.22E-04< gi g gg gg!!

I Sl I

• lI*

.2E 04-g .18E l8 g

I g

q l Ah4g e

m '

8 16E 04- I I

9 I I I -

E f

<r I I E .lE 04- I

[ I

% i '

M 800. hg I I I 600. b ,_ 8 l l 400.

W* I r-_ _ l -

r-200 25.

B. 25. 50. 75. 100. 125. 150. 175. 200.

TIME ISEC0f0SI Figure 6-1. PB80P Cladding Temperatures for 0.4 DECL Guillott e Break

, s .

4 i .N 6 4C 9

)<

t l

I I

k

.E - ,

j 1

~ 4 te  !

1  !

w l

I l

i I

M S W 6-13

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

i 1

J t

l .

I d

t i

t

,i i

i 9

i i

i

)

. om i e I W l &

I i

l i

4 i

i  :

I i i-t i

I i

i i

i l 1 1

)

I Figure 6-3. Flow into the Support Column / Free Standing Miner /Open Mole Assemblies l *

! S t t ,

• e , ,

. - . - - _ ,-._._- - ..- - - - - _,. - ,,r--,w,.- - ,me-,,., m~.. - . , --.--. . --,,. ,- - - - , - . - - - - - - - - . - ,. - , - ..m -,,. , -- - - , . * - - - - - . .

t O

e B

4

. * - AsC

" ou a

E

[

l r

6 l

" \

. {

4 i

l

*e k i

- g ,

~

2<a  ;

o en C

I

' 6 i e

. T, i

+.g)

W >

i i

b I

- t

\ k l t I

r I

t i

i I

l f

1i

) 6 9

1

% J I I I  ?

I l .

i h

I i.

l t 6-15 l l t

_ . _ _ _ . , _ . . . _ . _ _ _ _ _ _- . - _ -f

e i

G e

m 4,C

• i P

I ,

b '

6 b

N sa

~

1 O [

• J +

C

• i t

E l 6

{

l

@ I I

B, N

i r

4

?

i I l

t I'

{

-f

= l i

r b

.. 4 i

O >

l r

6-16 i

i

..a_. 4 + 4 _ L 3-- - R _u--+ .h_. 2 $. a:-----__A,. - ++%e __ m_____. 4 a m __.i_.u._A-,m __w_ A 4

l

. I l I dec .

I I e '

. ,e i e

F"" b 4

I 8

W l sb <

t i g i i

i x 1

E '

i M

1 M [

j *e i . l 6

I N i

• e .

i c-  !

1- a i 1 .: -

I [

L I

l B i a t t

O r G

E.

- t

" l 4

i

• i E [

8.

b

• I l

Ep'WD

+

6-17 ,

I

1 i

e

~

- . Sec  ;

4 g

. i t

b n

P i

e r

a F

4 S

E e.=

O i

e E *t t=4 8

e

==

• 1 N E

l. s e

I @

l k

8.

. N 1

I I

r tl l **

i i

l,

6-18 i

1

e e

4C

, e 4

e a

9 m

8 e

U b=

==

O e

C me

. I c

M t

S r=

I.

N e

k N

I k

+

l i

I r

6-19

- - ----- --,,y_

4

~

~~ _

&sC e

e i

E 8

I z

u i .

~ .

B

=

p.e E

c e.e b

O d

E 8,

c e*

e M

W

• 6-20

2 - - -. A.- - -._.- - #

1 1

I i

. I e

4ec  ;

e e  :

i e

P b

I b \

. h o (

4 N

X  !

- t C  !

t

• e N

e. O t i

- 1 O '

~ \

E

w. t l f b f N

1 o ce f

1 6 f b f 8, i

{

l 4

• e

+

usund humumma o

e 6-21

V O

. &sC

= -

O 9

e 4

4 1

b b

m C

pm ,*

m g .

5 o

e 984 94 4

t 3.

b 9'

4 m W 6-22

h a

h l

4C I

W \

3 .

4 6

=  !

.iE F

1 4

'f

• l 8

i .

t i

i I

l i, -

' Y J

e N

m 0

W b

i

+

o M

e 6-23

9 4ec .

m e

9 e

't I

=

4 y ..

E

.e.

W *

.J

.E 13 M

9 O

M b

8, c

e8 h

enmf em 6-24

r

• 3,C l n

- M I

I e

i b

k I  !

I t

w 6 b

a t i

M ,

g 1 t

F E

. t

• P* f a  :'

w f

& I a .

I T

W 6

)

9 t t E

i I

t I

i 1

. I s

= M 0

6-25

O

. &eC _

- vi e

W f"*

4 ett 4 .

W e

a s

N.

e E

F 3

m m

E 1

e ..

L a .

L m

E a

W 5

w

=

W W

4

'O w

m 9

W

=>

• Ch 6 -

a in ^

=

b 6-26

2 a s -u- .ma ,

t 4

0  !

e e

et e 9 b. ';

_ j

,- e -e -

l r

• 4 i

L 6

I M

G

.,J 9

e I

N b

t 5

t

,-s I

f

[

b r 3

e m

b t

l

\ l s  ;

W L 6-27 l

I l 7

aC

- s m -

g E

t

=

l 2u m

O e

0 1

l b

I l

l l.

8b u

g .-

E e

W W

W T

e 3

C' e

e b

v=e I  !

e i

E c

b 1

• I j w l .

l l

6-28

.- )

r *

. e

• I I I i

1

.I t

1 1

t i.

1 i

i i

1 .

e 1

3 i

a i

4 i

i i i i

i o I

Figure 6-18. Liquid Level in the Upper Plenum GJter Global Channel f

4 5

SECTION 7

, . COMPARISONS OF THE APPENDIX K CALCULATIONS AND THE REALISTIC CALCULATIONS AT 95% PROBABILITY LEVEL 7-1. INTRODUCTION Detailed comparisons and calculations of the estimated 95th percentile PCT were presented in Reference (1) and the sup9 4 mental reports to Reference [1] using the lead two-loop plant results. Similar' calculations will be presented in this section for the PBNP Units using the plant specific superbounded and nomiral calculations. (The superbounded calculational approach was the result of the sensitivity studies performed in Reference (1) for the lead plant. That is; the numerical values from the sensitivity studies in Reference [1] are not used in the suparbounded calculation, but the impact of the parameter variation on the direction of the PCT is used such that the PCT is

,, maximized. The approach will be to show that the superbounded calculation is a conservative estimate of the 95th percentile PCT while at the same time, the 95th percentile PCT is less than the Appendix K calculated PCT. Therefore, the intent of SECY-83-472 (2) is satisfied.)a,e 7-1-1. WCOBRA/ TRAC Code Uncertaintios The uncertainties for WCOBRA/ TRAC code were described in Section 8 of i

Volume 1 of WCAP-10924 (3), suppleaents to th"is report (4), as well as Section 3 of Reference (1) and are defined below for completeness. The WCOBRA/ TRAC code uncertainty includes calculated values for (a code bias, uncertainty in the code bias, une)rtainty in the data, uncertainty in the measured test initial and boundary conditions, and the uncertainty in modeling test facilities. Each component of the coue uncertainty will be I given below.)a,e s.

1620v:1o/112988 7*1 l

7-1-2. WCOBRA/ TRAC Bias .

P *

.s e 4 L

1820v:10/112988 7-2

e 7-1-3. WCOBRA/ TRAC Uncertainties a,c e

e n.

4

~

1620v:10/1122tl

~

8,C

" e e

4 4

e e P

=

M g

use E

9 h

a s

b istov 10/1122:s 74 ,

)

O e

aC s e

O b

e, e

e I

f b

" m o

l l

l 1820v.10/112688 75 l

i

8,C

~~ - .

e

}

.=

e i

j l

l I

f 4 6

1820v:10/112988 7.g

&,C n

e 4

e c.

d il20v 10/111988 7-7

a.c i

v t

t t

7-2. CONSERVATIVE ESTIMATE OF THE 95TH PERCENTILE PCT USING THE "SUPERB 0VNDE0" i CALCULATION FOR P8NP a,c l

t k

h r

i I

. r h e4 6

b l

1820w1D/112944 7-8 l l

r t

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

1 1

i 8,C '

C e

o e

O I

e. >

4 e

F b

1820v:10/112288 7.g

a,c .

e

+ ,

I 7-3. REALISTIC ESTIMATES OF THE 95TH PERCENTILE PCT FOR PBNP a ,c

• e 9 8

e*

Emung b

ishioninu 7 10

I i

&,C

- - l 0

e S

0 9

O t

e e

N asumu O

"4 e

1620v.10/112988

- 8,C e

e 9

1 r

nomm 1

+-

6 1820v:10/112333 7-12

7-4. REFERENCES

1. Dederer, S. I. , Hochreiter, L. E., Schwarz, W. R. , Stucker, D. L. , Tsai,

, C-K., and M. Y. Young; "Westinghouse Large-Break LOCA Best Estimate Methodology Volume 2: Application to Two-Loop PWRs Equipped with Upper Plenum injection" WCAP-10924, Volume 2, Revision 1, April 1988 (WESTINGHOUSE PROPRIETARY).

2. "Emergency Core Cooling Systems Analysis Methods," USNRC SECY 83-472, November 1983. ,

3. Hochreiter, L. E., Schwarz, W. R., Takeuchi, K. , Isai, C-K. , and Young, M.

V., "Westinghouse Large-Break LOCA Best Estimate Methodology Volume 1:

Model Description and Validation," WCAP-10924-P, Volume 1 June 1986 (WESTINGHOUSE PROPRIETARY).

,, 4. Hochreiter, L. E., Schwarz, W. R., Tsai C-K., and M. Y. Young, i "Westinghouse Large Break Loca Best Estimate Methodology, Volume 1: Model

, Description and Validation, Addendum 3, Supplemental Information on WCBR.4/ TRAC Uncertainties, CCTF for Channel Core Analysis, References to NRC Questions on Decay Heat Model," WCAP 10924-P, Volume 1, Addendum 3.

l July 1988 (WESTINGHOUSE PROPRIETARY).

l l -

~

l .

l l

, b l

l l

182ov 10/112888 7-13 L

1 0

SECTION 8 4 3UlWARY AND CONCLUSION

, This report describes an analysis carried out to assess the performance of the Emergency Core Cooling system of Westinghouse PWRs equipped with upper plenum injection for the Point Beach Units owned and operated by Wisconsin Electric Power Company. The basis of this analysis is a realistic approach (Best Estimate Methodology) in the treatment of core and system thermal-hydraulics, to establish the ECCS performance during a postulated LOCA. Corpliance with current regulations in Appendix K is achieved by additional calculations and analyses using the models and assumotions that are required by Appendix K.

As described in the introduction, more recent NRC staff positions enbedied in SECY 83-472 (1) allowed reductions in calculated peak cladding temperature by using more realistic models (with the exception of those models required by Appendix K), provided the margin of conservatism of these calculations was

,, quantified by comparison with a suitable estimate of the true 95th percentile

. PCT.

The analyses in this report addressed the above requirement in the following manner:

1. A thermal-hydraulic model was chosen which incorporated all the phanomena deemed to be of importance for the system to be analyzed. The model used in this analysis was WCOBRA/ TRAC (2). -

2. An estimate of the accuracy of the code was established by comparing the code prediction to several experiments representative of the reactor system being analyzed and which included the thermal-hydraulic phenomena expected for a postulated LOCA. From these comparisons a code uncertainty wasdetermined(2). At this point the stateuwnt could be made that, f or a known set of initial and boundary conditions, for any system calculation the true PCT was within some value of the calculated PCT, with some probability level.

.I f

istorioniass: 8-1

r

3. A model of a two-loop PWR was developed with specific attention given to ,

those geometric features unique to the two-loop PWR with upper plenum injection. A lead plant concept was used in which the plant with the ,' ,

highest core power was analyzed.

4. Two types of analyses were performed. A bounded calculation was performed which contained initial and boundary conditions at biased values, so as to yield high peak cladding temperatures. A nominal calculation was performed which contained initial and boundary conditions'at their nominal values. The models used for these calcul,ations were the same except for a small number of heat source related models which were set at their biased (i.e., high peak cladding temperature values in the bounded calculation).

These cases would estimate the peak cladding temperature at the 95 percent probability level, and at the 50 percent probability level, respectively.

5. A comparison of the peak cladding temperature calculated using the model with Appendix K required features and the peak cladding temperature l calculated using the "superbounded" model confirmed that the licensing temperature exceeded the 95th percentile peak cladding temperature. It also confirmad that a sufficient level of safety and a margin of ,

l conservatism existed in the licensing calculation.

1 In conclusion, the PBNP analyses presented in this report, indicate that substantial safety margin exists in the ECCS performance of the upper plenum injection system. This analysis also providas a basis for the continued licensed operation of these plants.

8-1. REFERENCES 1

1. "Emergency Core Cooling Systems Analysis Methods," USNRC CECY 83-472, November 1983.

2. Hochreiter L. E., Schwarz, W. R., Takeuchi, K., Tsai, C-K., and Young, M. Y., "Westinghouse Large-Breat LOCA Best Estimate Methodology Volume 1: Model Description and Validation," #

WCAP-10924-P, June 1986 (WESTINGHOUSE PROPRIETARY).

8 Istoriomass: 8-2