ML20214P396

Rev to Yankee Atomic Electric Co Steam Cooling Model for Reflood Rates Less than 1 Inch/Second

ML20214P396
Person / Time
Site:	Maine Yankee
Issue date:	04/30/1987
From:	Jensen R INTERMOUNTAIN TECHNOLOGIES, INC.
To:
Shared Package
ML20214P379	List: ML20214P378 ML20214P396
References
MN-87-59, NUDOCS 8706030356
Download: ML20214P396 (26)

Text

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

M AINE YANKEE ATOMIC POWER COMPONV ATTACHMENT B MN-87-59 Revision to YAEC Steam Cooling Model for Reflood Rates Less than 1 In/Sec R. T. Jensen

Intermountain Technologies, Inc.

i 1

8706030356 870521 PDR ADOCK 05000309 P PDR a 8650L-HFJ

, :y 5

REVISION TO YAEC STEAM COOLING MODEL IVR REFLOOD RATES LESS TEAN 1 IN/SEC April 1987 1

Prepared for

YANKEE ATOMIC ELECTRIC COMPANY 1671 Worcester Framingham, Mass 01701 i

e Prepared by R. T. Jensen INTERMOUNTAIN TECHNG 03IES, INC.

1400 Benton - P. O. Box 1604 Idaho Falls, Idaho 83403-1604

NOTICE This report was prepared by the organization (s) named below as an account of work sponsored by the Yankee Atomic Electric Company (YAEC). -Neither YAEC, members of YAEC, the organization (s) named below, nor ag person acting on behalf of av of them: (a) makes av waranty, express or implied, with respect to the use of av information, apparatus, method, or process disclosed in this report or that such use may not infringe privately owned rights; or (b) assines any liabilities with respect to the use of, or for damages resulting f rom the use of, any information, apparatus, method or process disclosed in this report.

Prepared by INTERFOUNTAIN TECHNOLOGIES, INC.

Idaho Falls, Idaho -

9: ,

CONTENTS Section f.agg SJHMARY AND CONaUSIONS 1 1 INTRODUCTION 2

-2 DISCUSSION OF EECHT-SEASET BLOCKED BUNDLE TEST RESULTS 3 3 YAEC MODEL DESCRIPTION 8 4 REFERENCES 14 APPENDIX A EECHT CORREATION BENCHMARKING 15 4

i 4

l l

4 . _ _ _ . , _ . . . . . , _ ,

SUMMARY

AND CONCLUSIONS

-The Yankee Atomic Electric Company (YAEC) steam cooling model for core heat transfer with reflooding rates less than 1 in/sec has been revised. The objectives met by the new steam cooling model are to:

1. Remove the excessive conservatism present in the existing YAEC steam cooling model.

2. Utilize information from RECHT-SEASET in developing and justifying the new model.

3 Satisfy the intent of Appendix K by conservatively computing the effect of the blockage upon the channel flow and heat transfer.

4. Assure that the model is always conservative compared with the EECHT correlation.

1 A review of the results of the FLECHT-SEASET test results for blocked and unblocked bundles has shown that, even for blockages with flow bypass, the net effect of blockage upon the core heat transfer is beneficial. The heat transfe enhancement observed in EECHT-SEASET was due to the combined effects of droplet breakup and single.-phase steam turbulence. The Appendix K requirenent that heat transfer be assumed to be only due to steam cooling precluded accounting f or the effect of droplet breakup but allows the inclusion of a steam turbulence model. The revised model thus assumed steam flow with the dominant heat transfer to be as calculated using the YAEC version of the EECHT heat transfer coefficient correlation. The detrimental effect of flow bypass in the blockage region and the heat transfer enhancement effect of steam turbulence were conservatively modeled. In addition, since the YAEC EECHT correlation serves as the basis for the heat transfer coefficient, the correlation was benchmarked against six test runs from two of EECHT test series which were not included in the data base used in developing the correlation. This benchmarking showed the YAEC l

EECHT heat transfer correlation to be conservative in the range of interest for Maine Yankee.

Sensitivity studies performed for Maine Yankee with cosine and top skewed power l.

profiles have shown that the use of revised steam cooling model results in lowe peak cladding temperatures (PCI) than were computed with the old steam cooling model. The computed PCTs were, how ev er, higher than those computed assuming no t

steam cooling requirement (RECHT correlation).

i l

I 1 l

Section 1 INTRODUCTION Since the Yankee Atomic Electric Company (YAEC) steam oooling model for reflood heat transfer at reflooding rates of less than 1 inch /sec was developed in 1975, a considerable amount of new low reflooding rate data have become available. These data include test assemblies with and without flow blockage.

The data from EECHT-SEASET(2,3,4) have shown that: (1) no discontinuity in heat transfer occurs at a flooding rate of 1 in/sec as postulated in 10CFR50 Appendix K, and (2) that the core heat transfer is not degraded under blockage situations expected for a FWR LOCA. The requirements of 10CFR50 Appendix K are, however, still in effect, and it is necessary for the revised steam cooling model to meet the intent of Appendix K. A revised steam cooling model has thus been developed which assumes that cooling is by steam only and which accounts for the effect of blockage upon both the heat transfer and steam flow. The new model has been defined to remove most of the excessive conservatism in the existing model but such that the calculated heat transfer cannot exceed that which would be calculated if the steam cooling requirement were not imposed, i.e. , that calculated using the YAEC version of the EECHT correlation.

The justification of the new model consisted of using a fccmulation consistent with the data trends and models from the EECHT-SEASET 163-rod blocked bundle test s. In addition, the YAEC version of the EECHT correlation was benchmarked against data from the EECHT Low Flooding Rate Cosine (5) tests and the FLECHT Low Flooding k.e Skewed tests (0) .

Section 2 of this report discusses the results of and the applicable analytim1 models develcred in the EECHT-SEASET blocked bundle tests. The new YAEC steam cooling model and typical results obtained with the new model are presented in Section 3 *1he EEWT carrelation benchmarking results are presented in Appendix A.

2

.s *

, Section 2 DISCUSSION OF EECHT-SEASET BLOCKED BUNDLE TEST RESULTS The EECHT-SEASET blocked bundle tests consisted of tests in a 21-rod bundle and a 163-rod bundle. The 21-rod bundle tests were used to determine effects of various flow blockage configurations on reflooding behavior and to screen flow blockage configurations for 163-rod flow blockage tests. The 163-md bundle provided the means to determine the effect of blockage upon heat transfer with the worst case blockage configuration in a bundle large enough to allow flow bypass around the blocked region. The overall result was a blockage heat transre benefit relative to no blockage.

The relative benefit due to blockage can be seen in Figures 2-1 and 2-2.

Figure 2-1 shows the ratio of heat transfer coefficient with blockage to that without blockage as a function of time and elevation for configurations with and without flow bypass. The maximum temperature rise difference or the temperature rise observed in an unblocked bundle minus the corresponding temperature rise in a blocked bundle is shown in Figure 2-2 as a function of flooding rate and elevation. Dio important conclusions can be reached based upon these data. The first is that even with flow byptss, the effect of blockage upon heat transfer is beneficial. The second conclusion is that for floodinc rates of less than 1.5 in/sec, as the flooding rate decreases, the relative benefit due to blockage generally increases. This is important because the maximum benefit occurs at the flow rates for which Appendix K requires a penalty to be taken.

Analytical studies performed using the COBRA-TF computer program (3) showed the heat transfer enhancement to result from two effects. These were (1) breakup of the liquid droplets due to acceleration and increased impact in the blockage zone, and (2) increased turbulence of the steam phase due to boundary layer separation and reattachment as the flow accelerates in the blockage region.

A model for the heat transfer enhancement due to the steam phase turbulence was developed in EECHT-SEASET(3,4) . Since the steam phase phenomena are allowable with1.7 the Appendix K requirements, that model will be described here. The

, model exprems the local Nusselt number in terms of the free stream (unblocked)

Nusselt number and the maximum Nusselt number which occurs at the point of 3

l1ll1 3

1 9 A

\ -

0 2

1 1 e g

N-1 a

k S c S o l

A P S O E 0 1 B Y

B S

A

- A_ r a

P n W, a Y 9 l O B p L o C

F W ,o 1 T O h L - t U F 8 o O B H

T I

H TI u A r o

f W W 7 t

D n e .

- /

Z i) g 0 E N A d-6 5

i c4 fe fc en oe Cr rf ee e

\

, fR s

nm ao

- 4 rr TF

(

t

- as

d. en Ho 3 i dt ea

- zr i u

$ E A 2 l g ai C C C mf rn E E E S S S oo NC 0 0 0 4 6 0 l 1 .

1 1

2

- - _ - - e

- 0 r

~

u g

7 6 4 3 2, 1 0 9 i 1 1 1 1 1 1 0 F o

E[

ll l

• e 12s= A LAmot sumoLE o 25 aco ausoLL

,, _z.u O LARGE suMDLE D 21 aoo su~utt 2.cs. O Lenot stacit O 2 eioo suuutt -son ai ax runutt o n . -,. .n . -

L 25 a mie

_ 2s i- no 20 2 mis

!. l'"'-

I i 1 100 -

pL000

• RA11 7 ,,,, #

i c

CD ~ p/

m g

s f ,,,,, -

\

n f

a - /f , ' '

b D lf a -

"D , ' b* % , ,/

g f #

E n

/ .

, -o

/

A /

eE I I l l l I I o.1 a.2 oJ os os os o.7 os o.s l Pont RIF L OW R Afl0 i

l Figpre 2-2. Fhximta Tanperature Rise Difference Between Blocked and Unblocked Bundles as a Ebnction of Flooding Rate and Elevation (From Reference 4).

l 5

a boundary layer reattachment. The ratio of the maximum-to-free stream Nusselt utanbers is given as 0.2Re2 /3

-E**=-----**kg773 ( 2.1 )

% 0.0797 Re p 0

where the expression in the numerator was obtained from abrupt expansion experiments performed with air by Zemarick and Dougall(I) and the denominator is a correlation for fully developed flow from the FLECHT-SEASET 161-rod bundle steam cooling test. The rndisturbed diameter is D , and the Mundary layer 0

separation point diameter, D3 ,p, is assumed to be the diameter at the blockage plane. Pbr blockage without flow bypass, the mass flow is constant and Eq. 2.1 can be approximated by

- f* = 2.5 (D 0 sep Comparison of 14 2.2 to data frun several experiments, including the FLECHT-SEASET 21-rod bundle steam cooling tests (2) showed good a5reement if the constant 2.5 were changed to 1.88, thus

- f* = 1.88 (D 0 sep The Nusselt number returns to its fully developed value as the reattached toundary layer develops and the free stream turbulence decays. The decay was f ound to be exponential and independent of Reynolds number. The exponential relationship is Nu - Nu

• 0 (2.4 )

g ax--

m g0 = exp [-C( z- z,,) /D0 ]

where C = 0.6 - 0.45(D3 ,p/D0 ) 2.5) 6

The tern z is the distance above the blockage plane and z is the distance from the blockage plane to the reattachment point.

Equation 2.4 can te rewritten as Nu gO = 1. + ( Nu, /Nu 0

~ * * ~("* max 0 3 (*

In all of the above equations the diameter was asstuned to be defined as D = ( g$)1 2 2.7) 4 l

4 4

7

a Section 3 YAEC MODEL DESCRIPTION The objectives in developing the new YAEC steam cooling model are as follows:

1. Remove the excessive conservatism present in the existing YAEC steam oooling model.

2. Utilize information from EECHT-SEASEf in developing and justifying the new model.

3. Satisfy the intent of Appendix K by conservatively computing the effect of the blockage upon the channel flow and heat transfer.

4. Assure that the model is always conservative compared with the EECHT oorrelation.

The modeling assumptions used to develop a new steam cooling model which meets the above objectives are:

1. The coolant flow is saturated steam with a constant temperature of T .

2. The dominant heat transfer is as calculated using the EECHT correlation.

3 Flow bypass in the blockage region will be computed using the existing YAEC flw diversion model.

4. The change in heat transfer due to flow bypass will be calcun'ted.

5 Heat transfer enhancement due to single-phase turbulence will be conaervatively calculated.

9

6. Heat transfer enhancement due to droplet breakup will be neglected.

Fquation 2.1 can be used to compute both the effect of flow bypass upon the heat transfer and the single-phase heat transfer enhancement. The YAEC ficw diversion model provides the ratio of the local mass velocity, G, to the free stream mass velocity, Go, as a function of blockage fraction and distance from the blockagp pl ane. If Fq. 2.1 is written in terms of mass velocity 8

. Nu' 3-

" O.2( OD**f/p)

(3.1)

"o 0.0797(0 D0 0 P}

or Nu o 2/3 D 2

- E*f = CONS (g-) (D82E) /3 (3.2)

"o O O where the constant, CONS, will be defined to assure conservatim. The dianeter, D3 ,p, is the diameter at the blockage plane, and DO is the unblocked channel diameter. Thus, for diameters as defined by Eq. 2.7 D

g ffE = (1. - BLK) (3 3) 0 where BLK is the blockage fraction.

Substituting Eq. 3.3 into Eq. 3.2 gives

- E = CONS (--)2/3(1. - BLK)1/3 (3.4)-

"O O Equation 3.4 is used to solve for the Nusselt number at the blockage plane. To obtain the heat transfer coefficient from the Nusselt number, a further dianeter correction is needed. From the definition of Nusselt number (Nu = hD/K Nu h D /K

__E H = fax __sep_,

Nu 0 0 0' or hEE = - Ea3 (1 - BLK)- (3.5) 0 "O where h, is the heat transfer coefficient at the blockage plane. The heat transfer coefficient for an equivalent unblocked coolant channel, g, is obtained from the YAEC FLECHT correlation.

9

'T For the unblocked nodes in the channel, Eqs. 2.5 and 2.6 are solved fcr the Nusselt number. The diameters for all nodes except the blockage plane are asstaned to equal the diameter of the unblocked channel. The heat transfer ratio is, thus, equal to the Nusselt number ratio for all nodes except the blockage

! ' plane. In addition, the heat transfer coefficient computed in the steam cooling model is limited to be less than or equal to that computed using the FLECHT correlation.

Computed heat transfer coefficient ratios for blockage of 20, 30 and 60% are shown in Figure 3-1. The heat transfer returns to the free stream or unblocked value af ter 1.625 feet for blockages of 20 and 60% and af ter 1.875 feet for 30%

blockage. The potential effect of the new model upon the PCf in the Maine Yankee

.I.

reactor is shown in Figures 3-2 and 3-3 for power shapes that peak at core mid-plane and 855 core height, respectively. The new model can be seen to result in lower PCfs than the grevious steam cooling model, but is definitely conservative with respect to the FLECHT correlation.

e i

f I

o 10

%{. .

.4 .

8 ,. . .,

s 3 t >

's ,

s t

! o

~

o

/ e i

. i 1 l e t.

. //

l I

//

O ,  !// \'-

g ,' V l I ' '.

g- /.

,i I

P, o o i

'l ,

s s

s< .m .. g ,

~ , hg Jr mP 1 g

s,.

g

'.r ~

.z ,1 g , ,.

, w , ,,1 8

~ ,

s o I -

i s m I ,t ,

u. l -

i a

. u. o _1 >I w8  ! _

o . I oo I t

I e Ii+ < ,.

w  ; e, ( '

s

- u. I , ,s m .

z . ,

lI " lI a, m

• i- o I s, s

, m _i , _

i F- .

l (y

1 -<ow I '

~~

+ e.. I y) I ,.

q a\

.. )t, g s l - $

{; ' :

I .

.: 1 ,,

u -4 o

m

'N

.,)1 ; 'f

'i s

. 20% , BLOCKAGE y.'

5' d I - 430% BLOCKAGE i

.. v; I ,

r I s

I - - - -

60% BLOCI' AGE I

\;I 1 I i s *C5 l r

\

e o' l l I r I l

' ~

0. 0 0. 4 - 0.' 8 1. 2 1. 6 2. 0 DISTANCE. ABOVE BLOCKAGE (FT)

. . ., i

/ \ -

'[ FIGURE 3-1. HEAT TRANSFER COEFFICIENT RATIO ABOVE THE BLOCKAGE i

r

.. k,

~

t 3, ,

11

q. .,_

_l ..

\ .

7 m ,

8 i i i i 1 g

OLD STEAM COOLING MODEL

- NEW STEAM COOLING MODEL f

o -

l

---

NO STEAM COOLING N

h \

w Oo f

n8

/

g- E E

Z i

\

! \

\

\-

l 8

$ 1 I I I I

\ '

5 6 7 8 9 10 11 ELEVATION (FT)

FIGURE 3-2. STEAM COOLING MODEL COMPARISONS,

COSINE WITH 60% BLOCKAGE 12

\

8 '

s o

g - _

N b

u Oo / \

I8 -

/ ks -

E~ /.  %

• T s \

o g - _

OLD STEAM COOLING NEW STEAM COOLING NO STEAM COOLING 8

3 I

9. 5 10.5 11.5 ELEVATION (FT)

FIGURE 3-3. STEAM COOLING MODEL COMPARISONS,

, TOP SKEWED WITH 60% BLOCKAGE o

13 7

n- esw~n --,a m ~ a ~ -- .-o- , <ww.- ,, ,-m m - -

9 Section 4 REFERENCES

1. Exxon Nuclear comoany WREM-Based Generic PWR ECCS Evaluation Model.

Sunclement 1: Further Definitions and Justifications to Reflood Heat Transfer Models, XN-75-41, August 14, 1975.

2. N. Lee et al. , PWR FLECHT-SEASET Unblocked Bundle. Forced and Gravity Reflood Task Data Evaluation and Analysis Reoort, NRC/EPRI/ Westinghouse-10, February 1982.

3 L. E. Hochreiter et al. , Analysis of FLECHT-SEASET 169-Rod Blocked Bundle Data Usinst COBRA-TF, NRC/EPRI/ Westinghouse-15, April 1985.

4. L. E. Hochreiter, FLECHT-SEASET Prostram Final Reoort, NRC/EPRI/ Westinghouse-16, November 1985.

5. E. R. Rosal et al. , FLECHT Low Floodinst Rate Cosine Test Series Data Recort, WCAP-8651, December 1975.

6. E. R. Rosal et al. , FLECHT Low Ploodinst Rate Skewed Test Series Data Reoort, WCAP-9108, May 1977 7 P. P. Zemaneck and R. S. Dougall, " Local Heat Transf er Downstream of an Abrupt Circular Channel Expansion," Journal of Heat Transfer, February 1970, pp. 53-60.

l l

0 14

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

i- ;.?  :

Appendix A EECHT CORRELATION BENCHMARKING At the December 15,1986 ' meeting between YAEC and the NRC staff, the NRC stated i that additional benchmarking of the YAEC version of the EECHT correlation is expected. . The.YAEC EECHT oorrelation(l) was developed before much low flooding -

rate data and data with skewed profiles became available. The available data-have been reviewed, and data from the EECHT low flooding rate cosine (5) and skewed tests have been selected for this benchmarking.

I Table A-1 presents typical reflood parameters for the Maine Yankee reactor and for the EECHT tests chosen for use in the benchmarking. . The six tests which were selected are felt to provide an adequate indication of the correlation performance in the range of parameters expected for reflood in Maine Yankee.

The YAEC version of the T00DEE2 code was used to model the heater temperature response of the simulated fuel rods in the EECHT tests. . Some minor oode ,

modifications were required to eliminate the metal water reaction and gap conductivity calculations in T00DEE2.

The results of the bendunarking analyses were evaluated by comparing the calculated t peak cladding' temperatures as a function of elevation to the test results presented in References 5 and 6. These results are shown in Figures A-1 through A-6.

The T00DEE2 results can be seen to be clearly conservative with respect to the test data. This is especially true for the tests with top skewed profiles, i .

.i 4

a 0

4 e

1 15

l

\

Table A-1 l

YAEC EECHT CORREATION BENCHMARKING MATRIX Test /Platt Flood Rate Tgg Presstre Q, DI Coments l sb l

Maim Yankee 2.0--0.8 1500. 30. - 40. 0.64 148.

14331 1.55 (100 me) 1204. 32. 0.7 103. Sewed Prdue 1.0 (Omard) 14647 1.0 1610. 21. 0.47 142. Sewed Prdue 16110 0.8 1617. 20. 0.7 132. Sewed Prdue 5917 3.0 (4.6 see) 1600. 40. 0.95 141. 00sim Pr d u e 0.77 (Omard) 4831 1.5 1600. 40. 0.95 142. Oosim Prdue 4641 1.0 1601. 20. 0 95 139 (bsim Pr&He

=

16

8 g i i i i i g - _

o 8

O bo @ O w? -

e S -

5" s

O O N O6 w 0 So .e O@

!" g -

C, x" O I O O

8 O O o - _

o 8- 0 0 -

CALCULATED VS MEASURED 8

co 1 I I I i 0 2 4 6 8 10 12 ELEVATION (FT)

, FIGURE A-1. TOODEE FLECHT COMPARISON, FLECHT SKEWED TEST 14331 17

8 i i I i i g

o _

g _

o _

g _

~

'808 G

0 h w e s e 48 _ 8 $ _

5S e e 6 U

e 8 y O

• O a i_

g _

)

6 5

8 O O

l o _

l o _

1, O CALCULATED VS MEASURED o O I i 8

I I I j

D 2 4 6 8 10 12 ELEVATION (F)

. FIGURE A-2. TOODEE FLECHT COMPARISON, FLECHT SKEWED TEST 14647 o

18

~

• o-

. 8 i g 1 1 1 i

• o g _ _

N o

g _ _

N n

b -

E

?o

<o _

O h3 $

s

$g

=

s 4

=

8 S o

~

O Oee -

O O O @

i O o O g _ _

CALCULATED VS MEASURED g O 3 @ l I i I i

0 2 4 6 8 10 12 ELEVATION (FT) i FIGURE A-3. TOODEE FLECHT COMPARISON FLECHT SKEWED TEST 16110 19

'T

. . . ;. o .

.I I I l 1 l e

6 o O g _

N O

O n

6 O b N O O Oo

. sg -

e -

O \

! \e O

8 -

9 -

)

> 6 CALCULATED VS MEASURED O g

e o

a l l I I l O 2 4 6 8 10 12 ELEVATION (FT) 3 FIGURE A-4. TOODEE FLECHT COMPARISON, FLECHT COSINE TEST 5917 1

o

. . .po . .

8 g i 1 e I i O

O '

O O 6 0 o

g _ _

O b

w 6

>a

~

&n - -

W~

5 e s ,

D x

0 o

4) e i

g O CALCULATED VS MEASURED o

? I I I I i 0 2 4 6 8 10 12 ELEVATION (FT)

FIGURE A-5. TOODEE FLECHT COMPARISON, FLECHT COSINE TEST 4831 21

. l 1

8 y I i i I .I o

g _ _

N 8e a 8e e

l

~ ~

e e e O C

vo

~

g td 8 ~

,E ~

e e

=

0 58 HN

_ e _

e o

8 ~ / g

~

O g O 8

o g _ _

O CALCULATE 0 VS MEASURED o I

• I I I O 2 4 6 8 10 12 ELEVATION (FD FIGURE A-6. TOODEE FLECHT COMPARISON, FLECHT COSINE TEST 4641

.. . . . - - _ _ _ . . _ - - - _