ML20072A212

Nonproprietary Method for Determining Film Flow Coverage for AP600 Passive Containment Cooling Sys

ML20072A212
Person / Time
Site:	05200003
Issue date:	07/31/1994
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19304C494	List: ML20072A203 ML20072A212 NRC-94-4247, Forwards Nonproprietary & Proprietary Versions of Ltr Rept, Method for Determining Film Flow Coverage...Passive Ccs. Repts Provides Status of Activities Associated W/Validation of Wgothic Computer Code.Proprietary Rept Withheld
References
PCS-GSR-003, PCS-GSR-3, NUDOCS 9408120177
Download: ML20072A212 (19)

Text

- - . _ . .

l Wi3TINGilOOSE NON-PROPRIETARY CLAES 3 METttOD FOR DETERMININa FILM FLOW COVERAGE FOR TiiE AP600 PAhslVE COVTAINMENT COOLING SYSTEM 4

METHOD FOR DETERMINING FILM FLOW COVERAGE FOR THE AP600 PASSIVE CONTAINMENT COOLING SYSTEM July 1994 i

This is a non-proprietary version of Document IG-GSR-003.

l

/

I 9408120177 940728 ,

PDR ADOCK 05200003 l A PDR u:WWKhl199w.wpf 1b-080244 I

l

Mrtifon FOR DETt .tMINING FILM Flow CoVfRACE FOR TIIE AP600 P4ssivE CONT 4tNMEvr Coot. No SYSTEM TAllLE OF CONTENTS Section Title Pane ABSTRACT 1 Background 1

~

Film Stability Model 2 Determination of the Contact Wetting Angle for Coated Surface 2 l

Liquid Film Flow over an Ellip6 cal Dome 4 Analysis Method for Predicting Film Flow and Stability 7 Prediction of Water Coverage for the AP600 under Postulated Accident Conditions 12 CONCLUSIONS 14 REFERENCES 14 I

u:wwrAlim.wpr:1b-080294 iii

METilon ton DETEawsNING Fig.M Ftow CovERAct: tou 11tE APMH) PAwfVE CONTAINMENT COOLING SYSTEM LIST OF FIGURES Finure Pane Figure 1 Predictions of LST Water Coverage Zuber-Staub, Varying Reference R Value 10 Figure 2 AP600 Test Water Coverage Predictions Using Zuber-Staub Local Stability 11 Figure 3 AP600 PCS Operation under Postulated Accident Conditions 13 1

1 i

i l

l I

i i

l uhWX A.I 199*.wpf:It> 080294 IV

1 l

I d

Mimion roa DETERMININo Fit.M FIow CovEnAGE tor Tile APMHI PAntvE CONTAINMENT CoouSG SV5TE:M 1,IST OF TAlli,ES Table Pane Table i Summary of Test Results to Determine Contact Wetting Angle 4 ,

Table 2 Key Panuneters-AP600 Large-Scale and Water Distribution Tests 8 Table 3 Analytical Predictions of AP600 PCS Test Results 9 Table 4 Prediction of AP600 PCS Water Coverage-Design Basis Accident 12 ump 60thl199wxph1M*0294 y

METuon Fou DrTrustisisc Fitst Flow CovtRAGE FOR 11tE AP600 PAssivr CONTAINMENT COO (.ING SYSTDt AllSTRACT An analytical mod' has been developed to predict the stability behavior of a thin film of water flowing over heated and unheated surfaces. The surface consists of a steel shell in the shape of an elliptical dome, which extends down to fonn cylindrical walls. The surface of the shell is coated to resist corrosion and enhance the surface wettability. The model is compared to observed test data and is used to predict the wetting behavior of the AP600 passive containment cooling system (PCS).

llackground The AP600 PCS uses a thin film of wata that is applied to the outside of the containment shell.

During postulated accident conditions, the film is heated by the shell as it flows radially outward from the top of the dome and then downward along the vertical sides of the shell. Heat is removed from the shell due to evaporation of the film. The effectiveness of the PCS is evaluated by the extent of water film coverage that can be achieved.

Several tests have been performed to demonstrate the operation of the AP600 PCS. These tests include water film heat transfer tests using scale models of the AP600 containment,U" and non-heated water coverage tests using a full-scale,1/8-section of the AP600 dome?3 1he results of these tests and analyses have indicated that:

• The evaporating water film is an effective means of removing heat following a  !

postulated accident

• The AP600 PCS design provides sufficient water film coverage to adequately remove heat for design basis accidents, Lmiting the containment pressure and temperature below design limits

• The AP600 containment shell is coated with inorganic zine paint, which is highly corrosion-resistant and has excellent wetting characteristicsW

• Analyses of the AP600 containment have shown that the PCS design effectively removes heat, assuming conservatively bounding water coverage fractions?3 1he purpose of this analysis is to develop a model that predicts the onset of flow instability in a flowing film, leading to a set of criteria that can be used to predict the fraction of coverage for a combination of water flow rate and surface heat flux. This model will be compared to experimental results that include heated and unheated surfaces, and applied to the AP600 containment to assess the PCS water film stability and subsequent coverage.

uhp600\t 199w.wpf:1M*0294 1

MrTuon nia prTramsrw Inu trow Coven 4cc ron int AP6oo Passivr. CONTAINMENT Coouw Systru Film Stability Model To determine the water flow rate that will produce a stable film, a model is used that considers the momentum of the flowing film, the surface tension effects, the thermocapillary effects, and the potential energy. The model proposed by Zuber and Staub,A which is modified to include static pressure, is given by:

2 p 6 c(I -cos0) + _do q " cose (1)

_ gsin _p S4 + p g cos = .

15 _

p, 2 6 dT k where p is the liquid density g is the gravitational constant p is the liquid viscosity a is the liquid surface tension 0 is the contact wetting angle between the liquid and the surface p is the angle of inclination relative to horizontal q" is the surface heat flux k is the liquid thermal conductivity T is the liquid film temperature at which the properties are evaluated and S is the minimum film thickness for a stable film that is related to the minimum mass flow rate per unit perimeter, F,,, by:

in 3 F,, p g; 3,

8P' .

These equations are polynomials in F , and can be solved iteratively.

Each of the quantities in these equations are either fluid properties, such as the surface tension, viscosity, density, and thermal conductivity, or inputs specific to the application, such as the local heat llux, The only quantity that must be determined for a given problem is the contact wetting angle for .

the painted surface.

Determination of the Contact Wetting Angle for Coated Surface To measure the contact wetting angle, two samples were prepared. The first is a paint sample supplied to Westinghouse by the coating vendor Tids sample was painted by the vendor and was not subjected to weathering. The second sample is a 4 x 3 in.-section of a steel plate that was painted by Westinghouse and weathered for two years, uh imntiww.wpt:tb-oso294 2

_. ~ . - .

I l

METuon f oR DETERMWING FILM Flow COVER 4GE FOR THE AP600 PASSIVE COTTAINMENT COOUNG SYSTEM The following procedure was used to determine the contact wetting angle for both samples:

. The samples were cleaned and dried

. A drop of water was placed on the sample, which was held in a horizontal position

• An optical comparator, located at the Waltz Mill machine shop, was used to measure the angle

- between the sample surface and the drop at the interface j

= Measurements were repeated using several drops to ensure repeatability in the results l

Additional tests were conducted with the samples held at different temperatures to deterndne the effect of the swface temperature on the contact wetting angle. This was accomplished by heating the 1

samples with hot water or a heat gun. I The results of the tests are summarized in Table 1. It should be noted that the low temperature data should 'ne discounted due to condensation on the sample surface. Also, at high temperatures, the wettinll; angles started out higher than at lower temperatures. It was observed, however, that the drops quickly flattened out, reducing the angle.

It can be concluded from these tests that the contact wetting angle between the painted surface and watrr ranges from 20 to 28 degrees for weathered surfaces, and 30 to 53 degrees for unweathered surf aces. Since the AP600 containment should be well weathered prior to operation, the contact welling angle should be taken as 20 to 28 degrees. ,

l uNpNurst tW.wpf:1b-080294 3

==

METrion Hsu DrTERMINING I'!LM flow COVERAGE FOR TUE AP600 PASSIVE CONTAINMENT COOUNG MSTEM Table i

SUMMARY

OF TEST RESULTS TO DETERMINE CONTACT WETflNG ANGLE Contact Angie Contact Angle Description of Test Weathered Unweathered Sample Sample 24* 30

1. Room Temperature, T=80"F .

2. Heated, T=110"F 23 33

3. Heated, T=18(FF t=0 sec. 28" 53 23" 44 t=15 sec.

t=30 sec. 20" 35 20 28 t=60 sec.

Liquid Film Flow over an Elliptical Dome The AP600 PCS operates by applying the water at the center of the elliptical dome. 'Ihe water flows radially outward where it enco mters weirs that collect the water and reapply it in an even film. A simplified model of the film that accounts for the change in flow area due to the dome geometry and the change in the liquid flow due to evaporation is useful to determine the average radial flow per unit perimeter, F.

The dome surface is approximated by an oblate spheroid. The surface area is given by:

2 1 +C (3)

A** = n a2 .1 xb n 2 e 1 -c .

where a is the major semiaxis .

b is the minor semiaxis and c is the eccentricity of the revolving ellipse given by:

fa2 +b2 (4) g_

3 u%WRA119%.wpf:lb-080294 4

j 1

METHott FoW IDETI:RMINING FILM Flow Covt. RAGE Fou TIIE AP600 Passive CONTAt% TENT COOLING s) STEM i

Given an initial mass flow rate of water onto the top of the dome, the flow at any subsequent radius is given by:

q " , AA'

th, = th, , (5)

. h,,

where h,, is the latent heat of vaporization q", is the heat flux at the location i and AA, is the differential area given by: )

i AA, = n(x, +x,,3 )/(x,-x,,i) + (y,,, -y,)2 (6) 1 l

where:

x, = r,cosa (7) y, = r isina (8) 1 1

ab i r=

i (9) 2 2 i fa (sina)2 + b (cosa)2 l

l where a is the angle between the vector normal to the surface and horizontal.

This approach results in a conic approximation to the elliptical surface. Using angular increments of 6 degrees, the error is less than 1 percent, Finally, the mass flow per unit perimeter is given by:

th' F' = (10) 2 n r, Thus, for an initial mass flow rate, the mass flow per unit perimeter can be calculated at each radius along the dome for a given heat llux distribution.

i u:\ar41A1199w.wpf:lb-080294 5

METuon von IPETtausstso Fli.ht Ft.ow CoeurcE Fou T1tE AP600 PA%SIVE CoVT41% TENT CooWNG SYSTEht A useful quantity is the ratio of the mass flow per unit perimeter to the minimum stable value as calculated in Equation 2.

R= (11)

I[.

As discussed in Reference 7, the film is expected to split, causing a dry patch to form when this ratio is less than unity. Ilowever, this work was done for a smooth, vertical surface, and it is expected that this ratio or stability margin should be somewhat greater than unity to account for surface _

inconsistencies and/or flow maldistributions under non-laboratory conditions.

Af ter splitting, it is proposed that the coverage of the film can be reduced by an amount equal to the ratio, R; that is, the film contracts to the stability limit based on uniform thickness. Thus, the fraction .

of the surface area covered by the water film, $,is given by:

$,,, = $' R' (12) ref where R, is the local value of the stability margin

$, is the local value of the film coverage fraction

$,.i is the value of the film coverage fraction at the next point down the wall and R,a is the reference value of the stabili'y sargin that is used to determine the onset of instability and accounts for the suface inconsistencies and/or flow maldistribotion The value of $ is defined as unity before the onset ofinstability. As the film thins due to geometric spreading or evaporation, the value of R can approach R,g. At this point, the flow is assumed to split, and the coverage fraction, $ is reduced according to Equation 12. As the film splits, the water redistributes, increasing flow uniformly in the remaining wet areas. These areas remain stable until the flow thins, causing additional splitting. In this way, coverage fractions for unstable flows will decrease asymptotically until the film reaches the bottom of the surface.

This model can be correlated to water coverage fractions observed in the heated and unheated AP600 PCS tests. Specifically, the model will be applied to determine the value of Pyg that best predicts the ,

observed coverage fractions from these tests. This correlation will then be used to estimate the coverage fractions for the AP600 containment under postulated accident conditions.

u: Nap 600\ll%3pf:lb-080294 6

urmon ton nonmumw rau now Coun4cc son mE APHO Pmvr CovrowEvr Coouw sumt i

Analysis Method for Predicting Film Flow and Stability l 1

A computer program was written to solve the equations for film flow over a dome, to determine the minimum flow per unit perimeter to ensure the stability of die film, and to determine the coverage fraction should the film become unstable. The program considers a typical containment geometry, including a dome section and a cylindrical wall section. Given the surface geometry, initial conditions, heat flux distribution, contact wetting angle, and water flow rate, Equations 1 through 12 are solved to determine the value of F, F. R, and $ at each radial position down the dome. If R is greater than R,,n the average flow per unit perimeter exceeds the minimum value, and the flow is expected to be stable. If R is less than R,er, the average flow per unit perimeter is less dian the minimum value, causing the flow to split and the coverage fraction to be reduced accordingly.

Variati(ms in the surface of the shcIl will cause local flow distributions that vary around the circumference. These local variations cannot be predicted using these simple models, but the observed coverage can be predicted using the local stability model and comparing it to various tests. The model can then be used with confidence to estimate the coverage for the AP600 PCS under postulated accident conditions.

Existing data from both heated and unheated tests will be evaluated using the local stability model. 1 The heated tests include the AP600 large-scale containment tests, which include a scale model of the dome; while the unheated tests include the AP600 water distribution tests, which utilize a full-scale, 1/8-sector of the AP600 dome. Key parameters from these tests are summarized in Table 2.

DI The large-scale baseline tests utilized a serics of nozzics to apply the film in a ring near the top of the dome, it is likely that local variations in the film flow around the circumference were significant.

Observations of the tests showed that for the high heat flux / low flow tests, dry stripes were found to occur, indicating film-stability-based coverage fractions.

The unheated water distribution tests were run with a wide range of water flow rates. The film was found to remain intact at moderate to high flow rates, and steady dry stripes formed at lower flow rates. These tests included maximum weld and surface deviations, dius, providing base cold coverage fractions, accounting for full-scale geometry effects.

The stability model was used to predict the results of thirteen large-scale tests. Three different values of the reference value of the stability margin, R,,,, were used, and the results are shown in Figure 1, The reference value of the stability margin that best predicts the data was found to be:

ukptml 199m.wpf:1t-080294 7

1

! Monion rou prTruurwisc nou Iuiw covtuact tou uir APM>o P4ssive cowTrisurNT contisc MsTru pui (13)

[

In all, the stability model was applied to the large-scale tests and four water distribution tests. 'Ihese results are surmnarized in Table 3, and are shown graphically in Figure 2. As shown in Figure 2, the unheated AP600 water distribution tests are predicted by the model that was developed using the ,

heated large-scale test data.

It is also expected that this model, which relics on local film stability, is applicable to any size structure with similar geometric shape, such as the prototypical AP600 containment.

Table 2 KEY PARAMETERS-AP6001,ARGE-SCAL,E AND WATER DISTRillUTION TESTS

- a,c

'~

1,arge-Scale AP600 Water Parameter TestdM Distribution"'

Dome Major Axis (a)

Dome Minor Axis (b)

Verucal Wall llencath Dome Water Flow Rate initial Water Temperature Contact Wetting Angle Peak Heat Flux fleat Flux Distribution (i=0" (Top)

(See Note II) [1=24" (l=48*

[l=72 (i=W)"

vertical A: Note that the Al%00 water distribution tests modeled the dome and 20-ft. section of vertical wall. '

The actual AP(0) containment has a vertical wall section that is 83 ft. high.

-11: Note that the heat flus profile is due to subcooled water added at the top of the dome.

u:\apNuAli9% wpf:1 M *0294 8

Mruson n>u I>rnusumo atu trow CovruAct rou un: AlWM Passive CovrA!WrVr CooWW MMM Table 3 ANALYTICAL PREDICTIONS OF AP600 PCS TEST RESULTS Large Scale Tests Oleated) (a,c)

Test Description Predicted Coverage Measured Coverage R9L Pressure = 10 psig RIOL Pressure = 30 psig R8L Pressure = 43 psig Rl7AL Pressure = 10 psig R34L Pressure = 31 psig R27L Pressure = 40 psig R24L Pressure = 30 psig R23L Presorre = 30 psig R26L Pressure = 30 psig R21L Pressure = 31 psig R22L Pressure = 31 psig R28AL Pressure = 40 psig

_ R28L Pressure = 40 psig _

APMN) Water Distribution Tests (Unheated) _fa,c)

'VDT14 Flow = 55 GPM

, WDTIO Flow = 100 GPM l

WDT9 Flow = 220 GPM 1

• l WDTil Bow = 280 GPM L ._

uNpNrN IWw.wpf:1M*0N4 9

C I

g i Predictions of LSTWater Coverage s i, Zuber-Staub, Varying Reference R Value j k b 1 15 i  !  !  ! i i i ll; 14- i i

t

~~~+

i i, 5 s

i i i i i i n 1.3---  !  ! t i i  !  ;

V cp i i

c i- i r-i E

u 1.2-  :  :

C: -

=

3 O  :  :  : #

[

$ g'g _  !  ! ..._ _ h O I g-------

7 l------------------

W i i

' M i i 9 r5 ... .-.. a i  !  ::: :: O 8 s- 1_  : e ,

e c  :

3 E -

hr o

CD  :  :" "

1 so 0.9 i--

R"- g .

x Ett  :

F l: 5 g E  :  :  :

i m T.g Q.

0.8 i i

l-a: rTr  : g g i i i ,

: x  :  :  : .

0.7  ; i -

! i  !

i i E i i i i a 06  ! i X  :

=

4

-  :  :  :  : t

: .i . z

i ' ' '

0~5 '

A 65 70 75 80 85 90 95 100 4 Measured Value of Water Coverage (%)  !

s r

- (aJ d 4 m

E

- ~

w . e e

MITiton tua 1)ETEk%ffNING FIthi Flow CovERAcr. FOR TiiE AP600 PAMIVE CONTAl%tEvr COOLING SYSTEh!

(a,c) 9 O

I e

emman Figure 2 1

u:\ap6(ILAl199w.wpf:lb-080294 gj l

4

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

METiton rom DETERMINt%G FILM Ft.ow COVERAGE FoM T1tE AP600 PAntvE CONTAl% MENT Coott%G SYSTEM Prediction of Water Coverage for the AP600 Under Postulated Accident Conditions i

The film stability model developed from die AP600 containment tests was used to predict die coverage of the AP600 containment during a postulated accident. The AP600 passive containment cooling system (PCS)is initiated when the pressure inside the containment reaches a set value. At that time, water is applied at the top of the dome of the steel containment shell. The water flows radially -

outward and encounters a series of weirs, which distribute the flow over the majority of the dome and the vertical wall. The maximum water flow is reached early in the transient when the maximum .

energy due to the reactor cooling system blowdown is released inside containment. The water flow is reduced later in the transient as the reactor decay heat level decreases. The average wall heat flux, as calculated by the WGOTHIC containment analysis code, and the corresponding cooling water flow rates are shown in Figure 3.14 The initial value of flow and heat flux is indicative of the blowdown phase of the transient, while subsequent times represent the reactor decay heat level. The WGOTillC model that generated these heat flux values divides the water flow path into seven discrete areas; three on the dome and four on the vertical wall. The model assumes a water coverage fraction of 40 percent for the dome and 70 percent for the vertical wall.

The local stability model was used to analyze several flow / heat flux pairs from Figure 3. The results of these analyses are summarized in Table 4. These results indicate that the coverage fraction input into the WGOTHIC model should be higher in the dome region and lower in the vertical wall region.

Table 4 PREDICTION OF AP600 PCS WATER COVERAGIMDESIGN llASIS ACCIDENT  ;

Time Flow . q" Dome ('7c) Cylinder ('7c) litu - Mid Mid lir ihmh s ft 2 Top Mid Hot Top Top Bot Bot Exit 0.183 30.4 990 2.167 29.7 .728 5.167 28.7 .473 5.667 15.7 .473 9.167 15.3 .393 ,

15.17- 14.7 .331 4

21.17 14.1 .301 26.17 11.8 .288 u Anp60thlIh.wpf:1t>4*0294 12

^ ..

C "L

k 2

s ...

s.

-o

,m I l

?, AP600 PCS Operation  != ,

i Under Postulated Accident Conditions i w

30 l

i I

h: -

I i7

:  :  :  : m

:  :  :  : F

: r

---

+----------------------------------------;----------------.-._0.9 30 J -

Ni-------

^

@ i -

a m  :  :

-!  !  !  !  ! - 0.8 m,

>m C 25--- - -- ! -

t ----

--t----

--t: -t-------

3_,  :

e i i i i i $'

_ y 21 ~

i

- 0.7 -Qx s u .,

= 3 O 20-- t.-- t.-- - t.- 1 .t -

s a w 1  :  :  :  :  : 1 7

:  :  :  : s u  :  :  : - 0.6 m

- +::

e  : -:  :

• 7

~W 15--  :

+-- ---+

I a s 4 3 -

i i  !

- 0.5 =

'  : O m  : i i 3 0 10--- i.---

- r. - +-

. --------i.----

3 e,

n.  !

-0.4s y 5

4 5-r -

i  !

<  ?

=

:  : - . - m- -

^ -03 -

:  :  :  : 6

: a 0 i  ; i i i 0.2 E O 5 10 15 20 25 30 Time (hours)

METHon tou DrTEuutNING lilAt h ow Covt uact Fou 11sE APf,00 PA%IVE Co%TAIMIEVr CENH.1%4 sWitAl CONCI,USIONS The following conclusions can be drawn from this analysis:

. A set of criteria have been developed to determine the stability of water films on a coated surface for both heated and unheated conditions.

. In order to apply these criteria, the contact wetting angle between the water and the coated surface was .

experimentally measured. This angle was measured to be approximately 20 degrees for a wide range  ;

of surface temperatures.

. These criteria have been applied to several AP600 water flow tests to develop a method for calculating the film coverage for a coated steel containment structure, given a heat flux at the wall and an initial water flow rate.

1

. 'Ihis method has been shown to predict the behavior of the large-scale heated tests and the AP600 l water distribution tests. These tests cover a wide range of geometries, water flow rates, and wall heat  ;

flux values.

. The method relies on local film stability and is applicable to any size structure with a similar geometric shape to the prototypical AP600 contrJnment.

. "Ihis method has been applied to the AP600 containment to predict coverage values during postulated accident conditions. These values are somewhat different than what is currently used in the AP60()

WGOTillC analysis.

Ilased on these conclusions, it is recommended that the WGOTillC AP600 model should be revised to reflect the coverage values shown in Table 4. An evaluation should be made to determine the sensidvity of the containment pressure to these coverage values. l 11ased on sensitivity studies that have been performed with WGOTillC for the AP600,W it is expected that these recommended coverage values will not significantly af fect the containment pressure resp (mse. i i

REFERENCES i r

, 1. " Tests of IIcat Transfer and Water Film Evaporation on a IIcated Plate Simulating Cooling of the AP6(X) l 1

Reactor Containment," AP600 Doc. #PCS-T2R-011,1992.

2. " Tests of Ileat Transfer and Water Film Evaporation from a Simulated Containment to Demonstrate the AP600 Passive Contaimnent Cooling System," WCAP-13246, Rev.1,1991.

u bgMO\l 199w.wpf:l tw0tW294 ]4

a unmm ton inm uuiww nui runv connu;r i um)r P4ssm: covruwevr coouw smni

/

3. " AP6fX) 1/8 Large Scale Passive Containment Cooling System Ilcat Transfer Test Baseline Data Report "

AP6(X) Doc. # PCS-T2R-(X)3, Rev.1,1992.

4. " Phase 3 Passive Containment Cooling System Water Distribution Tests," WCAP-13817,1994.

• 5. Carboline Product Data Sheet, Carbo Zinc 11 IIS Inorganic Zine Primer, Carbolinc Inc, St. Louis, Mo.,

November 1989.

O

6. AP6fX) SSAR, Rev. O.

7. Zuber, N. and Staub, F.W., " Stability of Dry Patches Forming in Liquid Films Flowing Over IIcated Surfaces," int. J. IIcal Mass Transfer, Vol. 9, pp 897-906,1966.

8. Wills, M.E., et.al., " Effectiveness of External Cooling and Associated Studies on Westinghouse AP600 Passive Plant," INC Conference Toronto, Ont., October 1993.

a uropmnt imurt.n>ao294 15