ML20213G254

Natural Circulation Boron Mixing Evaluation Program Rept

ML20213G254
Person / Time
Site:	Beaver Valley
Issue date:	04/30/1987
From:	Bruschi H, Gagnon A, Oft R WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20213G249	List: ML20213G246 ML20213G254
References
WCAP-11461, NUDOCS 8705180206
Download: ML20213G254 (33)

Text

,

WESTINGHOUSE CLASS 3 WCAP-ll461 BEAVER VALLEY UNIT 2 NATURAL CIRCULATION BORON MIXING EVALUATION PROGRAM REPORT Prepared by: R. R. Oft .

A. F. Gagnon April, 1987 APPROVED: [ / h--

H. p Bruschi, MaVigi}d /

Systems Engineering Nuclear Technology Systems Division WESTINGHOUSE ELECTRIC CORPORATION Nuclear Technology Systems Division P. O. Box 355 Pittsburgh, Pennsylvania 15230 8705180206 870511-PDR E

ADOCK 05000412 PDR

TABLE OF CONTENTS SECTION SUBJECT PAGE

1.0 INTRODUCTION

1 2.0 SYSTEM-COMPARISON AND EVALUATION 2 4

3.0 NATURAL CIRCULATION FLOW EVALUATION 8 4.0 NATURAL CIRCdLATION BORON MIXING EVALUATION 27 5.0

SUMMARY

AND CONCLUSIONS 29

6.0 REFERENCES

30 1381q _j_

1.0 INTRODUCTION

This natural circulation boron mixing evaluation program has been developed to evaluate the boron mixing capability of Beaver Valley Unit 2 relative to the requirements of Branch Technical Position RSB 5-1, Design Requirements for Decay Heat Removal Systems (Reference 1). The program will evaluate the boron cixing capability of Beaver Valley Unit 2 under natural circulation conditions for both symmetric and asymmetric configurations. The symmetric configuration is defined as all reactor coolant loops being available for natural circulation assuming no failures in the safety grade systems. To address the case of a single failure in the safety grade systems, an evaluation of natural circulation and boron mixing under an asymmetric natural circulation configuration (i.e., one steam generator unavailable for steam release) will be performed.

The Beaver Valley Unit 2 natural circulation boron mixing program consists of the following three program activities:

1) A general comparison between Beaver Valley Unit 2 and Diablo Canyon Unit 1 of the plant systems and equipment that affect natural circulation and boron mixing.

2) An evaluation of the natural circulation flowrates for Beaver Valley Unit 2 for both symmetric and asymmetric conditions.

3) An evaluation of the boron mixing for Beaver Valley Unit 2 for both symmetric and asymmetric conditions.

1381q - - .- . _ - . .-. -._.-

2.0 SYSTEM COMPARISON AND EVALUATION This section of this report describes and compares the natural circulation and boron mixing capabilities of Beaver Valley Unit 2 with those of Diablo Canyon Unit 1, as identified in the Diablo Canyon final test report. The final report for the Diablo Canyon natural circulation test is provided in the Diablo Canyon Units 1 and 2 Natural Circulation / Boron Mixing /Cooldown Test Final Post Test Report (Reference 2).

Subsection 2.1 provides a general comparison between the system design features of Beaver Valley Unit 2 and Diablo Canyon Unit 1.

' Subsection 2.2 provides an evaluation of the applicability of the Diablo Canyon test results to Beaver Valley Unit 2.

Subsection 2.3 discusses a single failure evaluation and the resulting

asymmetric natural circulation system configuration for Beaver Valley Unit 2.

The asymmetric flow condition is evaluated in Sections 3.0 and 4.0 of this report.

l 4

1381q _ - . _ -

2.1 Ca=aarison of Diablo Canyon with Beaver Vallev Unit 2 This subsection compares the systems and equipment that affect natural circulation and boron mixing of Beaver Valley Unit 2 to those of Diablo Canyon Unit 1 in sufficient detail to evaluate systems capabilities.

Reactor Coolant System The general configuration of the piping and components in the reactor coolant loop is the same in both Beaver Valley Unit 2 and Diablo Canyon. Beaver i

Valley Unit 2 has three heat transfer loops, while Diablo Canyon has four loops for heat transfer. Each heat transfer loop contains a steam generator and a reactor coolant pump (RCP). Both plants have the same model steam gene,rators (Model 51) and reactor coolant pumps (Model 93A). Also, one loop at each plant is equipped with a pressurizer.

1 Pressure control is available at both Diablo Canyon and Beaver Valley Unit 2 using the normal pressurizer spray valves or the pressurizer auxiliary spray systems. If both the normal and auxiliary spray valves are unavailable, the pressurizer PORVs are available at each plant for RCS depressurization. At Beaver Valley Unit 2, the pressurizer spray valves and auxiliary spray valves are not safety grade, however, the PORVs are safety-grade Class 1E solenoid operated valves.

Auxiliary Feedwater System l

The auxiliary feedwater systems at both Diablo Canyon and Beaver Valley Unit 2 are. capable of supplying cooling to all steam generators using the auxiliary feedwater pumps during the natural circulation cooldown. The systems will I provide water to the SGs from large storage tanks. The condensate storage tank provides this water source at Diablo Canyon, while Beaver Valley Unit 2 uses the Seismic Category I primary plant demineralized water storage tank.

The auxiliary feedwater system at Beaver Valley Unit 2 is a safety grade system.

I 1381q -

Main Steam System-The steam generators at both plants have PORVs which are utilized for the plant cooldown. At Beaver Valley Unit 2, an additional valve, the residual heat release valve, is provided for plant cooldown. The residual heat release valve is located in a common header from the three steam generators. At t ,

l Beaver Valley Unit 2, the PORVs and residual heat removal release valve are safety grade and are powered from Class lE buses.

Chemical and Volume Control System (CVCS)

Injection of boric acid into the RCS is required to offset xenon decay and the reactivity change which occurs during plant cooldown. The Diablo Canyon natural circulation cooldown test utilized the charging pumps to charge through the boron injection tank (at 20000 ppm boron) in the Safety Injection i

System. Subsequent charging was alig'ned from the volume control tank in the CVCS. The boron concentration in the volume control tank was adjusted to 2000 ppm to simulate charging f rom the refueling water storage tank (RWST).

4.

At Beaver Valley Unit 2, four weight percent boric acid is pumped from the i

safety grade boric acid tanks (at 7000 ppm boron) by the boric acid transfer pumps to the suction of the centrifugal charging pumps. These pumps are also safety grade and are powered from Class 1E buses. A backup source of boric I

- acid is available from the RWST (at 2000 ppm boron). The borated water is then injected to the RCS via the normal charging line and the RCP seals.

Either of the redundant throttling flow paths in the SIS provides a safety i

grade backup means for injection. Each of the SIS paths contains a safety l grade Class 1E solenoid operated throttling valve that permits variable control of the makeup flow rate.

To accommodate the borated water addition to the RCS, letdown capability is normally provided by the non-safety grade normal and excess letdown lines to i

i the CVCS. If both the normal and excess letdown lines are unavailable, l letdown is provided by the safety grade reactor vessel head vent letdown line to the pressurizer relief tank. Throttling control of the head vent letdown is provided by two redundant parallel safety grade Class 1E solenoid valves.

1381q l

k i

2.2 . Anolicability of the Diablo Canyon Test Results to Beaver Vallev Unit 2 2.2.1 Natural Circulation i

The Diablo Canyon natural circulation test evaluation verified that RCS natural circulation flow could be established, thereby permitting boron mixing and RCS cooldown/depressurization to RHR system initiation conditions. This l

phase of the test had no specific acceptance criteria and it was evaluated based on the results of the boron mixing and cooldown/depressurization phases of the natural circulation cooldown test.

The Diablo Canyon test results indicated that natural circulation flowrates were adequate to ensure that core decay heat removal, boron mixing and plant cooldown/depressurization were maintained throughout the test. The response of the RCS temperatures indicated stable natural circulation conditions throughout the test.

j -The Beaver Valley Unit 2 plant and Diablo Canyon Unit 1 have been compared

, (Section 2.1) to ascertain any differences between the two plants that could potentially affect natural circulation flow. The general configuration of the piping and components in each reactor coolant loop is the same in both Beaver

! -Valley Unit 2 and Diablo Canyon Unit 1. The elevation head represented by these components and the system piping is similar in both plants. Steam i

generator units were also compared to ascertain any variation that could I- affect natural circulation capability by changing the effective elevation of the heat sink or the hydraulic resistance seen by the primary coolant. It was concluded that there are no differences in the design of the steam generators in the two plants (both plants use Model 51 steam generators) that would adversely af fect the natural circulation characteristics. Therefore, because

! of the similarity between the plants, it is concluded that the natural circulation flow capabilities would be similar; and, therefore, the results of l prototypical natural circulation cooldown tests conducted at Diablo Canyon are

! representative of the capability at Beaver Valley Unit 2.

I I 1381q l

2.2.2 Boron Mixine ,

p The Diablo Canyon boron mixing test evaluation demonstrated adequate boron nixing under natural circulation conditions when highly borated water at low temperatures and low flow rates (relative to RCS temperature and flow rate) was injected into the RCS. It also evaluated the time delay associated with

+ boron mixing under these conditions.

c The acceptance criterion for this phase of the Diablo Canyon test was that RCS hot-legs (loops 1 & 4) indicate that the active portions of the RCS were borated such that the boron concentration had increased by 250 ppm or more.

Boron injection was conducted at the Diablo Canyon test using the 20000 ppm i boron solution contained in the boron injection tank (BIT). The BIT's contents were flushed into the RCS and within 12 minutes, natural circulation had provided adequate mixing to increase the boron concentration in the RCS by i 340 ppe. Following injection, makeup to the VCT was set to provide 2000 ppm boron. This simulated suction of the charging pumps aligned to the RWST. The j charging pump discharge was aligned to provide seal injection flow to each RCP

' and charging flow to one RCS loop. This alignment was continued throughout the remainder of the test causing the boron concentration to further increase.

For the Beaver Valley Unit 2 plant, boron would be injected into the RCS from l the 7000 ppm boron solution of the BATS through the RCP seals and the normal charging line, if available. Also, as noted previously, a safety grade backup means of boron injection is provided by the SIS flow paths. This boron concentration (7000 ppm) at Beaver Valley Unit 2 is less than that used for j the successful 0,iablo Canyon test requiring the addition of a larger quantity of borated water over a longer time period for Beaver Valley Unit 2 to achieve a similar change in boron concentration. Thus, the Beaver Valley Unit 2 j boration is less concentrated than that tested at Diablo Canyon. However, because natural circulation flow at Beaver Valley Unit 2 is expected to be f very similar to the flow obtained at Diablo Canyon, adequate mixing of the i

boron would also be provided for Beaver Valley Unit 2.

t A

}

13814 - -__ - - _ __ - - - - -

2.3 Sinale Failure Evaluation The following is a brief description of the steam generator cooling capability for the Beaver Valley Unit 2 plant.

Each motor driven (MD) auxiliary feedwater pump and the turbine driven (TD) auxiliary feedwster pump are capable of supplying feedwater to all three steam generators. Each steam generator is provided with its own individual PORV.

Also, an additional atmospheric dump valve, the residual heat release valve, is provided on a common header from all three steam generators. The residual heat removal release valve has adequate capacity to permit cooldown to RHRS initiation conditions. The electric power for the three individual steam generator PORVs is provided by one Class lE bus while the common residual heat release valve is powered from the other Class lE bus. This valving arrangement provides the capability to tolerate a single failure without reduction in the number of steam generators available for cooldown, since the residual heat release valve is redundant to the three individual steam generator PORVs.

With the above described arrangement, all three steam generators would normally be available for a natural circulation cooldown even assuming that a single failure has occurred. However, if a steam generator is postulated to be unavailable for cooldown due to any reason, an assymetrical natural circulation configuration i.e., one steam generator unavailable for steam release would exist.

As noted in the introduction to this report, boron mixing under natural circulation will be evaluated for both the symmetric and asymmetric configurations. This evaluation is performed in Sections 3.0 and 4.0.

1381q 3.0 NATURAL CIRCULATION FLOW EVALUATION This section performs an evaluation to determine the expected natural circulation flowrates for Beaver Valley Unit 2.

Subsection 3.1 presents a summary of applicable industry natural circulation tests and discusses the applicability of these tests to Beaver Valley Unit 2.

Subsection 3.2 calculates expected flowrates for Beaver Valley Unit 2 for-symetric and asymmetric conditions.

O i

4 1381q -

8-

3.1 Aeolicability of Natural Circulation Tests to Beaver Vallev Unit-2 L

A series of natural circulation tests have been performed by various utilities with Westinghouse designed plants. A typical list of the tests performed by

~

the utilities are:

Natural Circulation Verification I

Natural Circulation with Loss of Pressurizer Heaters Natural Circulation at Reduced Pressure Natural Circulation with Loss of Offsite Power Effect.of Steam Generator Isolation on Natural Circulation Simulated Loss ci All.Onsite and Offsite Power Cooldown Capability of the Charging and Letdown System i

For the purpose of this report, two of the typically performed tests have been evaluated and summarized. The two tests reviewed were the Natural Circulation ^

Verification Test and the Effect of Steam Generator Isolation on Natural Circulation Test. As shown in Table 3.1, these two tests were performed at a ainimum of five Westinghouse designed plants (North Anna 2, Sequoyah 1, Salem j 2, McGuire 1 and Diablo Canyon 1). .

I Natural Circulation Verification Test i

4 The objective of this test was to establish, maintain and recover from natural i circulation conditions while at low power. The following description applies to the five Westinghouse plants that performed the natural circulation l verification tests, which are summarized by Table 3.1.

l The initial conditions of the tests were'as follows. The reactor was critical l at approximately 3% reactor power. All reactor coolant pumps were in ,

operation. The RCS was at normal temperature and pressure. Pressurizer level was at normal no-load conditions and steam generator narrow range levels were normal.

I i

l 1

1381q -g- '

4 With the reactor stabilized at approximately 3%, all reactor coolant pumps were tripped in rapid succession. Plant conditions were monitored and f recorded during the subsequent stabilization of plant parameters. Adjustments were made to RCP seal water flow rate, charging rate and auxiliary feedwater flow rate in order to maintain stable plants conditions. After stabilized conditions occurred, natural circulation was verified by monitoring the wide ,

range hot and cold leg temperatures and the resultant loop AT as shown in Table 3.1. Recovery from natural circulation was achieved by restarting all RCPs at the end of the test.

i Effect of Steam Generator Isolation on Natural Circulation r

The purpose of the test was to determine the offect of steam generator l isolation on natural circulation conditions to verify that natural circulation

, can provide sufficient flow to remove decay heat after a partial loss of heat sink. Table 3.1 summarizes the natural circulation steam generator isolation tests performed at the five Westinghouse designed plants.

The following is a general description of the natural circulation steam l generator isolation tests, which applies, in general, to all five plants listed in Table 3.1. (Note that for North Anna 2 only one steam generator was isolated during the test, while the other plants isolated two steam generators.)

l The reactor was critical at a low power level (e.g.1%). All reactor coolant pumps were running. The auxiliary feedwater system was in service supplying water to all steam generators. The primary system was at normal temperature I and pressure. i

)

! With the reactor coolant system stabilized at the low power level, all reactor

! coolant pumps were simultaneously tripped. The system was allowed to come to equilibrium conditions. Natural circulation conditions were verified in all loops and natural circulation was considered stable when the AT between T

COLO and T H0T for all loops became constant. The loop AT was noted and recorded.

I i

1381q - 10 l i

With natural circulation established in all loops, a first steam generator was isolated by closing its main steamline isolation valve and isolating all blowdown and auxiliary feedwater to this steam generator. After conditions stabilized, natural circulation was verified and confirmed stable by observing the loop AT for the unaffected loops. Next, a second steam generator was isolated. Again, after conditions stabilized, natural circulation was verified and confirmed stable by noting the loop AT for the unaffected loops. It is important to note that a small loop AT was observed in the isolated loops, which indicated that a weak natural circulation flow was being supported by ambient heat losses in the isolated loops. Finally, the steam generators were unisolated and natural circulation was re-verified in all reactor coolant loops.

Test Results and Conclusions The test data was reviewed for the natural circulation verification and steam generator isolation tests that were performed for the Westinghouse designed plants identified in Table 3.1. Based upon this review the test results for all of these natural circulation tests were found to be very similar.

Specifically, these tests demonstrated that natural circulation could be established and maintained with all reactor coolant loops in service and also following isolation of one or more steam generators. Also, all plant responses were consistent with expectations and results seen at other similar Westinghouse designed plants.

Since Beaver Valley Unit 2 is very similar in design to the Westinghouse plants that were tested (Table 3.1), it is concluded that adequate natural circulation flow is expected to occur at Beaver Valley Unit 2 with all steam generators available and when a steam generator is removed or is not available for heat removal.

1381g - 11

h TA8LE'3.1 SUfG4ARY OF NATURAL CIRCULATION TESTS i

4 NATURAL CIRCULATION VERIFICATION TEST SUfG4ARY North Diablo Anna 2 Seouovah 1 Sales 2 McGuire 1 Canyon 1

Test Number ST-8 1 90.1 7.2 1.1 4 . Reactor Power 3% 3% 35 35 3%

N.C. Flow Verified Yes Yes Yes Yes Yes Loop AT 36-40*F 36*F 40'F 28'F 35-40*F 4

i NATURAL CIRCULATION STEAM GENERATOR ISOLATION TEST SUMNARY North Diablo Anna 2 Seouovah 1 Sales 2 McGuire 1 Canyon 1

, Test Number 2-ST-11 4 SUP 90.4 7.4 1.5 Reactor-Power 1% 15 1% 1% 1%

All RCPs Tripped Yes Yes Yes Yes Yes N.C. Flow Verified Yes Yes Yes Yes Yes Loop AT - 21*F 24'F 20'F - 1 First SG Isolated SG B SG 3 SG 3 SG C SG 3 Loop AT - 26*F 30'F 20*F -

Second SG Isolated N/A SG 4 SG 4 SG D SG 4 Loop AT N/A 42*F 55'F 25'F -

i Key:

N/A - Item is not applicable

- Data is not available i

I-f l

4 13814 .

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

3.2 Natural Circulation Flow Calculation The attainment of transient equilibrium in the natural circulation mode can be confirmed by observing certain RCS and steam generator (SG) parameters. The parameters of interest are:

1. RCS subcooling based on core exit TCs should be greater than instrument '

inaccuracies.

2. The core exit TCs, RCS hot leg temperature and SG pressures should be decreasing slowly with time, as core decay heat falls off.

3. With SG pressures held relatively constant, the RCS cold leg temperatures should remain relatively constant at or slightly above the saturation temperature for the SG pressures being maintained.

4. Hot to cold leg temperature differences should be approximately equal to the full power forced convection temperature difference with all loops available for heat removal.

5. The core exit average temperature, based on averaged core exit TC readings, should be higher than the average cold leg temperature (assuming all loops in operation). This averaged reading should also decrease with time as the core decay heat input falls off, in step with hot leg temperatures and SG pressure readings in all active loops.

Continuous recording of these parameters will provide trending information in order to eliminate the effects of pointwise variations in the readings and einimize the chances of misinterpretation of any one set of readings.

variations in the same parameters in different loops can result from:

Asymmetry in the heat transfer and heat transport process between the loops.

Instrument inaccuracies.

1381q .-- --

- Differences in instrument sensing elements placement between loops.

- Variations in feedwater flow and steaming rates between the steam generators.

- Differences in the location of various auxiliary system connections to the 4 reactor coolant loop piping (e.g., charging line, letdown line, pressurizer surge line, etc...).

i In order to support natural circulation flow, the presence of a loop delta-T is necessary. Calculated delta-T's for Westinghouse PWRs in a stable free convection mode are in the range of approximately 120% of full power loop delta-T and downwards, with the higher delta-T's obtainable only at the highest expected decay heat levels. These higher delta-T's are probably not cbservable on RCS instrumentation since stable free convection heat transfer will not be set up until core decay heat levels have dropped significantly from their highest theoretically attainable values.

Westinghouse and the Westinghouse Owners Group (WOG) have developed a methodology to _ calculate natural circulation flows as a function of core decay heat levels. The details of this methodology can be found in the Emergency Response Guidelines (ERGS) Executive Volume, Generic Issues subsection related to Natural Circulation (Reference 3). A brief summary of the methodology follows.

Assuming quasi equilibrium, simple analytical expressions for mass flow rate and loop temperature differences expected during natural circulation operations assuming various core decay heat levels can be obtained. The expressions were derived in Reference 3 by equating the available thermal driving head to the total core and loop resistances under natural circulation flow conditions. These expressions are not intended to replace detailed analyses of the natural circulation process and should be treated as approximate relations for describing system performance under quasi steady-state heat removal conditions. These expressions, however, are useful for comparing the predicted conditions for natural circulation decay heat j removal.

l

- 14

~

13814

The equations to estimate the natural circulation flow rates and delta-Ts for any core decay heat level are as follows:

• (8.02pn)2 * (-5ql) 1/3 (l k/hr) (1) 2 Cp * (Kh + n g ) _

"(8.02En)2 , (_gqt) 1/3 g , q ,

(*F) (2) 2 Cp Cp * (K +n g) where n = number of active loops.

k = coef ficient of thermal expansion (*F~I)

Cp = average specific heat (Btu /lbm-Deg. F) based i on P and T K'g = sum of all flow corrected volumetric flow resistance factors for loop components (ft/gpm**2) q = core decay heat output (8tu/hr)

K'c = sum of all flow corrected volumetric flow resistance factors for core components (ft/gpm**2)

L = vertical separation between the hot and cold leg thermal center locations (f t) i 1

138fg _- _ _ - _ _ . - _ . - _ _ _ - _ - - _ _ _ _ _ __

p = average density (lbm/f t**3)

=

P Pg + Pc 2

p = cold leg fluid density (1bm/ft**3) p = hot leg fluid density (lbm/ft**3)

The fully developed natural circulation flow under quasi steady-state conditions is seen to be a function of the average fluid properties (p, Cp, 5),systemgeometry(L,K{,K*,n)andsystemheatinput(q).

c Significantly higher core and loop temperature dif ferences are developed in the free convection heat transfer as compared to the normal forced convection mode under the same decay heat assumptions. This becomes obvious by looking at the relation, under quasi steady state conditions:

i ATg = ATc" ._

Kp where AT = the temperature difference between the core inlet and outlet c

conditions.

i l

I

')

l 13819 1

Since the mass flow rate is much greater under forced convection compared to f ree convection conditions and Cp is about the same in both cases, a higher coolant average temperature is required to remove the same amount of heat under free convection conditions assuming constant steam pressure.

In addition to the verification of the methodology performed during the ERG development program, equations 1 and 2 above were independently verified during this effort by comparing the calculated results to the initial low power startup tests performed at Diablo Canyon Unit 1 (Reference 4). These tests were performed to verify the existence of natural circulation flow under conditions of full steam generator availability and 1 and 2 steam generator isolated. Table 3.2 presents the comparitive results between the calculated values and the test predictions for the loops with active steam generators.

In all cases, the natural circulation flow rates and delta-Ts predicted using the methodology were conservative with respect to the test results.

Therefore, this methodology can be used to conservatively predict the expected plant response under natural circulation conditions. It should be noted however, that this methodology does not attempt to determine conditions in loops with inactive SGs.

Specific calculations were perforned for both Diablo Canyon Unit 1 and Beaver Valley Unit 2 using the appropriate plant specific information (hydraulic resistances, system elevations, normal operating conditions, etc...) for each plant. The following assumptions were made for each plant specific calculation performed.

1. Steam generator pressures were held constant at all decay heat levels to maintain ,the cold leg fluid conditions equal to the normal full power conditions.

2. Iterations were performed on average fluid conditions based on the predicted delta-T values.

1381q  ;

3. The decay heat levels considered were:

a. 65 of full power.

b. 5% of full power.

c. 4% of full power.

d. 3% of full power.

e. 25 of full power.

f. 15 of, full power.

Figure 3.1 presents the comparison of the Beaver Valley Unit 2 and Diablo Canyon Unit 1 natural circulation flow predictions under full steam generator available conditions. As can be seen, the predicted flows versus core decay heat responses are nearly identical when the same methodology is applied to both plants. These results demonstrate that Beaver Valley Unit 2 should exhibit- similar natural circulation and boron mixing performance as was observed in the Diablo Canyon Unit 1 plant with all steam generators available. Table 3.3 presents additional plant specific results for the Beaver Valley Unit 2 cases analyzed assuming full steam generator capability.

Figures 3.2 and 3.3 demonstrate the predicted ATj and Tavg versus core decay heat level responses for Beaver Valley Unit 2 assuming all steam generators available.

In addition to the full steam generator availability cases analyzed, evaluations were also performed assuming one steam generator isolated. These additional evaluations were performed to address the single failure evaluation performed in section 2.3. The postulated assumption of one steam generator not available for steam release results in the necessity to evaluate the ability of the plant to achieve natural circulation under this asymmetrical configuration. The methodology used to predict natural circulation flow rates under asymmetric conditions is identical to that previously discussed above with the exception of the value for the number of active loops (n). This

! value was reduced by the number of steam generators which have been isolated.

[ For Beaver Valley under the worst single failure assumption,1 steam generator

[ will be isolated (i.e. n=2). Table 3.4 presents the specific results for the asyneetric evaluation performed for Beaver Valley (Figures 3.4 through 3.6 present the results graphically). As can be seen, decreasing the number of 1381q - 18 '

active steam generators results in both a decrease in the predicted natural circulation flow and a corresponding increase in the predicted loop delta-T as compared to the all steam generators available case. This is as expected under asymmetric loop natural circulation flow conditions and consistent with the steam generator isolation tests previously performed at numerous Westinghouse plants and discussed in the previous section. Although no calculations were performed to determine flow conditions for the loop with

! inactive steam generator (s), test results have indicated that delta-T conditions suitable to support reduced natural circulation flow in the inactive loop tave been measured (see Section 3.1). Therefore, the isolation cf one or more SG doesn't result in complete loop stagnation. As a result, some mixing of borated water in loops with inactive SGs is expected to occur.

In summary, the Beaver Valley Unit 2 and Diablo Canyon Unit 1 predicted i natural circulation flow responses assuming full steam generator capabilities are similar (as seen in Figure 3.1). In addition, even under postulated asymmetric core decay heat removal conditions, natural circulation flow was still predicted to occur. Therefore, it is expected that if tests were to be performed at Beaver Valley Unit 2, similar to those performed at Diablo Canyon Unit 1, natural circulation and boron mixing capabilities should be similar to those seen in Diablo Canyon Unit 1.

i i

i b

1 F

l 1381q - 19

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

TABLE '3.2 Methodology Verification versus Diablo Canyon Startup Tests POWER ACTIVE FLOW OT TAVG CASE (1) LOOPS (1) (Dee-F) (Dea-F) 1 Data 3. 4 -

39.0 562.0 i Pred. 3. 4 5.26 48.7 567.3 2 Data 1. 3 2.68 27.0 517.0 Pred. 1. 3 2.29 31.0 518.6 3 Data 1, 2 2.01 35.0 528.0 Pred. 1. 2 1,81 39.3 528.9 i

TABLE 3.3 Beaver Valley Unit 2, All Loops in Operation i

POWER POWCR FLOW FLOW OT TAVG

(8tu/hr) (1) (Ibm /hr) (5) (Dee-F) (Dea-F) l 5.431E+08 6. 5.346E+06 5.66 76.05 580.53 4.526E+08 5. 5.014E+06 5.26 68.32 576.66 3.621E+08 4. 4.638E+06 4.82 59.75 572.37 2.715E+08 3. 4.200E+06 4.33 50.08 567.54 1.810E+08 2. 3.656E+06 3.73 38.86 561.93 9.051E+07 1. 2.891E+06 2.92 24.94 554.97 l

j -

l TA8LE 3.4 1

Beaver Valley Unit 2,1 Steam Generator Isolated 4

POWER POWER FLOW FLOW OT TAVG (8tu/hr) (1) (1bm/hr) (1) (Dee-F) (Oeo-F)

< 5.431E+08 6. 4.207E+06 4.56 94.00 589.50 4.526E+08 5. 3.935E+06 4.21 85.00 585.00 3.621E+08 4. 3.634E+06 3.84 74.75 579.87 2.715E+08 3. 3.285E+06 3.43 63.01 574.01 1.810E+08 2. 2.855E+06 2.94 49.16 567.08 9.051E+07 1. 2.255E&O6 2.29 31.75 558.38

}

1 l

13814  ;

. , ~ , , - . . . . . , - , _ , , _ - , _ , - _ . - - - . - . . , , , .r,,.. - - - ~._ m mm ',- ,_. - . - _ _,.r._...__

1 Figure 3-1. Natural Circulation Flow Comparison

- Beaver Valley Unit 2 Versus DiablJ Canyon Unit i DNW _ _ , pg 6.0 5.5 5.0 >s g

k -

. m 4.5 j /

5 g h

/

I' 4.0 ,, /

y /

/

% . e E

o E 3.5 f

/

/

3.0  !

ff 2.5

- 3.0 4.0 5.0 6.0 1.5 2.5 3.5 4.5 5.5 Power Level (% of Nominal)

o.

to an.

sa

\ o.

sa m

a 0 gn.

J L

I m

e o e L

• C o 5 -

u a n

o E e +

b o

> Ma s-I m 0 to M

r"9 W

8 ~'

o.

o. m b o z O o J o.

N so A ed L

o cm

.e4

' \ .

m nn.

N o.

o. o. o. o. o. o. o.

• o o o o o o  ! o e s m o v m m (3 6a0) 1-94Isa

i 0

6

- 5 5

/ 0 5

l e

v e

L

/ f 5

4 r

e w

o )

P .

l 0 a e n r 4 i o m C o N

s u f o

s r 5. E e 3(

V g l v e

. a v T e 0 L p

o 3 r o e L w o

. P 3

3 5 a 2 r

u g

i F

0 2

/

/ 5 1

0 1

0 0 0 0 0 0 0 5 0 5 0 5 0 5 8 8 7 7 6 6 5 5 5 5 5 5 5 5

- h9 ?

i l

I \  ! ]l)

=

0

' 6

[ 5 5

l e

v e

L

/ 0 5

r e

P w

o e

r

/ '

5 4

oe Cs a )

sC l u 0 a sd re n 4 i et Va m

o wo l N os f lI F .

o no r 5.

3( X ot ia l tr e ce v l n e ue 0 L cG r 3 r im e Ca w e o lt P aS r

ui 5 t

a 2 N

4 3

f 0 a

r 2 u

i F

g

/

/ 5 i

/

0 0 5

~

0 5 0 5 01 5 4 4 3 3 2 2 8 8h.a as E

11 l

\ ,

0 6

5 5

l e

v

/ 0 5

e L

r 5 e

w 4 o

P e e r s Co a sd ue sta r

C

/ 0 4

)

l i

a n

m o

el N Vo s f T- I o 7 5.

e r t o 3( X l ta l e

Dr e pn oe oG L

e a v e m

/ 0 3

L e

v e

r e

w o

i t P S

t A

5 c-i

/ 5 2

3 i

F e

r u

g

/ e 0

2 5

1 0

0 0 0 0 0 0 0 01 0 0 0 0 0 0 0 0 0 9 8 7 6 5 4 3 1

!b.s

['

!l

1 l

l o.

m nn. .

no o.

na m

e e

J c_ in, I v o

o. e a ee p c- u o

u T3 o e

• C e v -

eu a 3m em @

cO em .

>s na a

4 m c. *M

>O F3 w ma na c.

m as O. a3 y ac =

sa \ e a

m. - e

>e x

• = c aa a.

<u tt)

. nn i N N m

G c.

D o.

k N nn.

\ e

o. *

o. o. o. o. o. o. o.

o nn a no o nn o nn m o m s o m to in na nn nn nn na nn nn nn (d *680) S^eJ.

7-4.0 NATURAL CIRCULATION BORON MIXING EVALUATION This section presents the boron mixing evaluation results for Beaver Valley Unit 2. By combining the results of the natural circulation flow evaluation from sections 3.1 and 3.2 with several small scale mixing tests performed in support of Pressurized Thermal Shock (PTS) resolution, sufficient information is available to predict boron mixing under natural circulation flow conditions. Specific industry data reviewed for this effort included both the CREARE 1/Sth scale tests performed in New Hampshire and the tests performed at the Imatran Voima OY facility in Helsinki Finland. A brief description of the facility and the tests will be presented in the following paragraphs.

As a result of concerns arising due to the potential for reactor vessel failure occurring as a result of PTS, the CREARE 1/5th scale test f acility was constructed to address these concerns. The PTS concern arises since the toughness of vessel weld material deteriorates after being irradiated, the introduction of cold high pressure injection (HPI) water to the cold leg fluid under low flow and stagnant conditions may not suf ficiently mix. This causes the welds in the reactor vessel to undergo thermal stresses. The combination of the thermal stresses and a rapid repressurization of the primary system could result in the propagation of pre-existing vessel flaws leading to ultimate vessel failure. The CREARE test facility, consisting of a single cold leg model with various HPI injection locations and downcomer were intended to determine the degree of mixing of the cold HPI water with warmer loop and downcomer water assuming various loop and HPI flow rates. The results of the tests and the exact dimensions and layout of the facility can be found in Reference 5.

One important parameter used during this series of tests was the ratio of loop flow (Q)) to HPI flow (Qhpi). From the test report (Reference 5), it was determined that for a ratio (Q)/Qhpi) in the range of 10 or greater, excellent mixing of the HPI and loop flow exists. Eight tests were performed f 10. These include tests 41,42,53,54,59,63,65 and at a ratio (Q j/Qhpi)

66. Higher ratio values were examined in test numbers 47,48, and 58. Again, these data indicated excellent mixing between the HPI and loop fluid. For the cases analyzed in the natural circulation flow section, the predicted loop 13814 ___

flow at 1% core decay heat was conservatively estimated to be approximately 2600 gpm per loop. Therefore, the maximum charging flow which would not exceed this ratio was determined to be approximately 260 gpm per loop. This value exceeds the normal charging flow capacity at Beaver Valley Unit 2. Thus it is clear that good mixing of the borated injection water with the RCS fluid will be maintained. However, the CREARE tests did not examine the potential interaction resulting f rom asymmetric flow configurations for a mutli-loop plant.

Additional tests performed at the Imatran Voima OY facility in Helsinki Finland further examined HPI and loop fluid mixing interaction under asymmetric loop flow and HPI injection conditions. The tests were sponsored in part by the USNRC to further clarify the expected mixing phenomenon under these conditions. The major differences between this facility and the CREARE facility are the existence of multiple cold legs and a semiannular downcomer.

This test configuration is more representative of a typical PWR configuration. Tests considered various combinations of HPI injection and loop flow conditions ranging from HPI injection only, to HPI injection in conjunction with loop flow. It should be noted that all but one of the test cases (Test case 111 being the only exception) were performed under asymmetric HPI injection conditions (i.e. injection into one or two of the cold legs only). A comparison of the test results indicate that the modified Froude number and density ratios used for this test program, closely match typical PWR conditions. Therefore, the test results obtained f rom this program can be applied to PWR geometries and conditions. In all cases, even when HPI injection into an all stagnant loop configuration was considered, mixing of the cold HPI water with the warmer fluid in the cold leg and downccmer regions was still predicted to occur although somewhat delayed. The details of the test matrix analyzed, facility description and results can be found in reference 6.

In summary, by comparing the test results from the two facilities with the expected flow responses at Beaver Valley Unit 2, it is expected that the borated HPI fluid will thoroughly mix throughout the system under either symmetric or asymmetric conditions and provide the required shutdown capability.

13814 - 28 '

5.0

SUMMARY

AND CONCLUSIONS A general comparison of the plant systems and equipment that affect natural circulation flow and boron mixing has been made between the Beaver Valley Unit 2 and Diablo Canyon Unit 1 plants. This comparison has demonstrated that the Beaver Valley Unit 2 natural circulation flowrates and boron mixing system capabilities are comparable to those of Diablo Canyon Unit 1. Available industry natural circulation test results for several Westinghouse designed plants, which are similar to the Beaver Valley Unit 2 design, have been reviewed. These tests included tests where all RCS loops were available for natural circulation flow conditions as well as cases where steam generators were isolated. Also, two Beaver Valley specific natural circulation flow calculations were performed. The first case assumed all steam generators were available for heat removal, while the second case was based on isolation of one steam generator. Based upon the industry natural circulation test results reviewed and the natural circ'ulation calculations performed, it is concluded that natural circulation flows will be established for Beaver Valley Unit 2 for both symetric and assymetric conditions. Furthermore, the natural circulation flow calculation results show that the Beaver Valley Unit 2 response is nearly identical to that predicated for Diablo Canyon Unit 1.

Finally, based upon comparing test results from several small scale mixing tests with the expected natural circulation flow responses for Beaver Valley Unit 2, it is expected that borated fluid injected into the RCS will thoroughly mix throughout the system under either symetric or assymetric conditions.

Therefore, it is concluded that boron mixing under symetric and assymetric natural circulation flow configurations will occur at Beaver Valley Unit 2, and testing at the Beaver Valley Unit 2 plant to verify boron mixing under natural circulation conditions is not required.

13814 . . . _ _ _

6.0 REFERENCES

1. Branch Technical Position RSB 5-1, Design Requirements for Decay Heat Removal Systems, Revision 2, July 1981.

2. WCAP-11086, Diablo Canyon Units 1 and 2 Natural Circulation /8oron Mixing /Cooldown Test Final Post Test Report, March, 1986.

3. " Emergency Response Guidelines - Revision 1", Westinghouse Owners Group, Septembe r, 1983.

4. Diablo Canyon startup test, "Special Low Power Test Program - Natural Circulation (T.P. 44.1)", May 1984.

5. NP-2312, " Fluid and Thermal Mixing In A Model Cold Leg and Downtomer With loop Flow", Paul'H. Rothe and Margert F. Ackerson, January 1982. ll

6. NUREG/IA-0004, " Thermal Mixing Tests in a Semiannular Downcomer With Interacting Flows From Cold Legs", H. Tuomisto and P. Mustonen, October

_1986.

13814 _ _ _ _ _ _ _ _

_ __