ML20023E065

Forwards Section of Idaho Nuclear Corp Rept Ny-123-69 Re Statistical Verification of Macbeth Correlation

ML20023E065
Person / Time
Site:	Calvert Cliffs
Issue date:	08/01/1973
From:	Kling C ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY
To:	Salah S US ATOMIC ENERGY COMMISSION (AEC)
Shared Package
ML20023E061	List: ML20023E060 ML20023E065
References
NUDOCS 8306070159
Download: ML20023E065 (15)

Text

e COV!USTfCN Olvf SION COV!UST10N ENOINE EPING.e.C

,.)

WIND 50" CCNN C6C55 203 6eB 19n CA8tt CoveENG "l}

mm 5J COMBUSTION DIVISION August 1, 1973 Dr. S. Salah United States Atomic Energy Cor: mission Rocm 511 7920 Norfolk Avenue Bethesda, MD 20014

Dear Dr. Salah:

~

Enclosed is a Xerox r.opy of the section of the Idaho Nuclear Corporation Report,Ny-123-69, concerning the statistical verification of the Macbeth correlatien. The pages of specific interest to you are II-3 and 11-4.

If I can be of any further assistance, please don't hesitate to contact me.

Sincerely yours:

g.I w

Dr. Charles L. a ing ~

Supervisor, Nuclear Safety Dept.

CLK:vp1 Enclosure I

8306070159 830525 PDR ADOCK 05000317 P

PDR

lif y,C C 1

(-

p u

~ l yy-d f 7.. ',.J. '..

IDAHO NUCLEAR C 05R P O R A T.I O N p

g

^

X '. *

~

- - - -- O

'3801 9

.. hg,4

• P.O.

831 ISAS 10860 FALLS. IDaro

'i;

~'. '~ ^'

L0.$lllll:j'i

ovester 10, 1969

aclear Safety Drc rs=

enth*.~.

? er er:. C-::?er

..y 123-c,y To:

Addressees s

Gentlemen:

Inclosed please find your copies of the October Monthly Report of the ::uelcar Safety Prcgram Divisien of Idaho ::uclear Corporation.

Very truly yours, WE:::be

a. :.. i.y er,..r..ager

uclear Esfety Pr:gran Division

Enclosure:

Nuclear Safety Program Report - October 1969 Addressees:

C. ::. Kelber - A1L H. A. Sindt. nAPD L. Earer - A::L - L:72RPO D. F. Ju dd - 3 & '4 J. A. Redfield - BAPL R. E. 'das cher - 3 & W D. L. *torrison. 3:'.I M. F. Valerino - CE T. A. "aker - IITRI L. E. *!innick - CYAPC T. W. Trout - KAPL A. P. Bray - GE

'4. 3. Cottrell - OR: L (5)

D. O. Fischer - GE L. C. Oakes - OR:!L G. E. Wade - GE G. J. Rogers - P:lL D. Free an - GE. Sunnyvale G. M. Brown - S:!E T. P.ockwell. III - lGR

.a-er -

a v

-.a..

a u.

...s.

, R. G. Bock - GE - San Jose R. A. Wicsemann - WCAP W. K. Dodson - KE J. F. Palmer - AECL J. D. O'Toole - U::C

.'. 3:ur6ecis - CFA F. R. Fc.rner - U:'AEA

?. Tagani - HCRL H. G. Ru rf - TUV

'i. I:hik:/.a - JAIRI A. Jahns." IRS e

9 8

b U

v 4

it, t

~

"povenber 10, 1969 To: Addressees Uy-123-69 Page 2

_ Assistant Director for Nuclear Safety, DRDT cc:

Assistant Dire:ter fer Project :.!anagement, DRDT Assistant Dire:ter for Plant Engineering, DRDT Assistant Dire:::r fer Reicter Engineering, DROT r.._,

u..

a...

-.a LCFT ?regra: an ger, DRDT

~

Chief, Research ar.d Develc; ment Ernnch, DRDT Chief, Engineerir.g and Test 3 ranch, DRDT Chief Environmental and Sanitary Engineering 3 ranch, DRDT Chief, Fuel Engine ring 3 ranch, DRDT Chief, Contr:1 l*echanic: 3 ranch, DRDT Chief, k*ater Systens 3 ranch, DRDT Division of Naval Reactors, D::R (R. S. 3rodsky)

Director, Division of Reacter Standards Deputy Director, REG Assistant Director for Reacters, REG Assistant Directer for Special Projects, REG Director, Division of Reactcr Licensing, REG Chief, nuclear and Systems Tc:hnolcgy 3 ranch, DRL (2) '

Direct:r, ::u: lear Technolegy Divisien, ID (3)

Director, 'Euclear Safety Division, ID

. Directer, LCFT Project Divisien, ID (3)

Directer, Hu:1 car Engineering and Cenetruction Division, ID Chief, Technical Services 3 ranch, CSTS Division, ID Chief, Budget Branch, ID g*

9 d

4 s

1 6

e a

e i

J

~.

s o

c o

-p p

IDAHO 1*UCLIAR CORPORATION Nuclear Safety Prcgrer. Division t

m

,,0..-.. Y a PO.u

c..a nu October 1969 1

e l

i l

l l

l I

i 1

The inferna_ior. cer.';cir.d in this rep:rt ic' prelini:...ry a r.i : ':j e:t to further eV31MatiOh.

i I

"s 1

l

.t 1

l r

l 1

s 1

O.

U e

r ng-TA3LE OF CO :TE;;TS Page I.

Sr o..a I-1 II.

KUCLEAR S/JETY DEVELOPM2HT.

........ II-1 II-1 A.

Analytical Developent II-10 B.

Separate Effects Tests II-11 C.

Power Reactor Safety Analysis.

II-12 D.

Fission Prcduct Tests.

E.

In-Plant Testing of Reactor Engineered Safety Syste=s II-1k F.

Reliability Monitoring Prc6 ras II-15 G.

E=erGency Core Coeling Heat Transfer Tests II-16 H.

Acoustic Flav Detection and !{onitoring System II-25 Developent.....................

49

\\'

4 s

8 ii 9

U U

~

.Detober 19c9 y

II. 1:UCLEAR SATE T DEVELGII'E"T BRA"CH A.

Analytical Development 1.

Highlights None 2.

Technical Activities Core Heat Transfer Anelyris: A users manual has been ' prepared and distributed within Idanc ::uclear Corporation for the single rod, single channel ecde TI: ITAL.

Ther cphysical properties for UO2 and Zircaloy-h have been inccrpcrated in the code.

Werk is continuing tcvard development of a three-dimensicnal core heatup code for the FEUST 8x3 rod bundle. A subroutine to calculate the fluid energy balance at each axial level is currently being developed.

With the addition of the subroutine the code vill be operational.

Dhe CEF study continued. A total of 729 lea pressure (150 to 725 psia) red bundle CHF data points has been collected and correlated, and a report of this study started. Also, a report describing the ec=parison of selected CHF correlaticns to data previously ecliected was started.

A portion of this report follows.

One major parameter in 1 css-of-coolant accident (LOCA) analysis is the time-to-critical heat flux (a) (CHF), that is, the ti=e frc: initiation of the break to the time critical heat flux occurs in the cere. I: ediately fellouing the occurrence of a break, the pressure differential across the core results in forced convection which begins to transfer the sensible energy stored in the fuel rods and the decay energy scnerated by nuclear fission frc: the rods to the coolant flowing thrcush the core. Eventually the critical heat flux conditien, which is a functicn cf the red anM core geenetry, axial heat flux distributice, and the fl:v and ther:cdyr.anic state of the coolant is reached. At JEF e film of vtp r exists adjacent to the red surface and results in severe reduction in the a cunt of enerEY that can be removed from the core. Tclieving the attainment of the CHF condition, the rod surface ter;crature 'cegins to increase as the energy rc:aining in the rods and the generated decay energy 'cecoces redistributed.

Alth:ngh predic'.icn of th: Oritical heat flux is neces:try fer reactor design purposes, the phenc:enen of :ritical heat flux maintains a unique status in engineering in that no theoretical analysis has been atte=pted

() Critic;; ". cat.1x ic uced in thic reper: *o decerite a large decrease in the lceal hea: tran:fer ccefficient which results in a large increase of the surface temperature.

e e

y x

.,4

n 6ctober 1969

~

and consequently experimental data are the only cource cf information Therefcre, in crder to obtain design values regarding it s. cccurrence. flux, ex.ccrirents are ucually conducted at steady-of the critical heat

spacing, state with electrically heated rods of design diameter, pitch, and length and with fluid conditiens at er near the design operating If sufficient data with a systenatic variatien of the core inlet reint.

cendition and c- ' *+ --se flux are taken, a ccrrela ion of the data may be attempted and this correlaticn in turn say be uced as a design equation.

In contrast to the problen of' cbtaining informatien required for denign purpcses, the probler of predicting when CEF vill be reached during a,less-of-ccolant accident requires data over vide ranges of pressure, = ass flux, and inlet conditions and, in addition, the effects cf rapid ch2nges in pressure and cass flux must be taken into account.

An investigation cf all CEF data available in the open literature for axial flux prcfile, red bundles and of several CEF steady-state, unifer:

correlatiens has been undertaken with the cbjective of determining the applicability of the data and ccrrelations to =cdels being developed to predict the core thertal behavior during a loss-of-coolant accident.

The correlaticns censidered in this repcrt are prcbably the costThey have been ci at the tresent tir

, 3,eger(3, Gencral Electric (e). and '..~estinghcuce (Shippinspcrt)(h).

.videlv2[2)vnanduce n-3:rne':t Eenuits obtained by using these ccrrelatiens were ec pared trith 1C96 CEF data points for rod bundles which cover the fc11cving ranges of parameters:

156 to lk00 psia Pressure 6 lbe/hr-ft2 0.02 x 106 to 4,o x 10 Mass flux Fluid inlet ecndition 0to3733tu/lb subecoling 0.250 to 0.625 in.

Roi diameter 30.0 to 178.0 in.

Rod length 0.022 to 0 30T in.

Spacing betwee'n rods Axial heat flu); distribution Unifonn tra'".-i. F r ~* Ey..ir aticn cf Published R. V. :*a cb e th. 5 "r -u'. R :p

\\Lyu~).

(1)

V:r.d ?c a for ?.ci ;z.ur..: u, n za-A Ccrrele-i-- of Furneut Dsta fer Uniferrlr Eeeted (2)

P. G. Barnett, ci rcrri.

...s.e: hed Eunizes, An-uli and Its Us e _for !r,lic:inr ?.. :. is

'i. W-h -03 (19cc).

'<. 2 + e r (3)

K. M. Becker, ' ? : rn c u. O '- "

' - * ' - T ' -"

in Vertical.c4 Sundic e,.2-eTO (1jui).

9 (h)

L. S. Tcng, T.cilinc l'ect Tranrfer an-1 P.re-Fhare Fiev, New York:

Jchn Wiley and Scns, Inc.,1907 II-2 a

U V

.-,j; October 1969

.y For purpeces cf eczpariscn, the percent error between the calculated and experimental values is defined as

~ Q ur o

c"c(100)

(1) cer'a ERROR =

S'CEF,e where qcgy,, = experimental criticc1 ?. eat. flux, cleulate riti al heat flux.

q EF,c

=

C The results of the calcula:1cns are presented in Figures II-1 through II-5 as

• graphs of the frequency that an error is encountered versus the percent error of E u2tien 1.

Whereas these figures do not reveal the cetails of the error distribution with respect to pressure, length, mass flux, inlet subcooling, and bundle internal ccnstructicn, they do give an overall view t'

of the accuracy of the correlation.

Figure II-1 gives the error distrit.'ticn of the Macbeth ec: relation ( ).

The results are goci considering the fact that the correlation is based cn a very lizi ci a: cunt of data (172 points) given in Reference 1.

'Jhen cc pared with CEF values determined fcr ranges of parameters within, as gerrelation predicts well as cutside, the ranges en which it is based, tha 97%cfthedatavithintheerrcrbounds-207,toF255(8

/.

Figure II-2 61ves the error distribution for the Barnett correlation (2).

Again, the data were obtained fren 'cundles uhose sec ctric and operating characteristics were within and outside the range fer which the correlation was based. All of the data are predicted within -32 5% and +27 5% and 97% is predicted within -22 5% ani +17 5%. These results and an examination of Figures II-l thrcugh II-5 indicate that the Barnet correlation is v.

superier to the c hers for predicting steady-state CEF data frca red I

bundles.

Figure II-3 gives the errer distributien for the 3ecker.ccrrelation(3)

Although the ccrrelation is based on data over the pressure ranse 31 < P 5 lh00 psia, and all ef the rod bundle data are within this range, the predictions are not as accurate as these of the Macbeth and 3arnett correla-i tiens. The calculated results for 97% of the data agree with experimental 1.

re.c., a-. a n. a...a-u,,'3.

(a) The 975 lev;l of cerrclatien is videly used in the open literature.

f Ecuever, the :hcico cf -he errcr ecunde within whi:h 97,' cf the data is f

pre dicted s quite a #"---

y.

Fcr this repcrt, values were chcsen ~:etween which the errer distributien is sy :ctrical.

I I

l II-3

(/

i ep 140 U

.c 3

~

g-12 0 C

100 80

~

60 40

-l 1

=

20 n.o l

~

H t

t i

f i

' ' I t

g

-40

-30

-20

-10 0

to 20 30 40

!O IQ

~GCMF c CHFe C M F. e tNC-A-14 240.

I l

Fig. II-l Error Distribution for the Macbeth CHF Correlation II-b e

l

y U

9

..,p..

zg 180 u

c t,

3a 160 w

=

'14 0

~-

12 0 10 0 BC

^

60

- 40 f

l

- 20 l

1

\\

I t

g i

-40

-30

-20

-to O

to 20 30 (g

-o 1800/o l

CMF.e C MF.4 CMF.e r

Fig. II-2 Error Distributien for the Barnett CIIF Correlation l

l II-5 I

l

%.)

U u0 r.y

- 10 0 3

.EF C

90 80,

1 70 60 50

- 40

- 30 i

- 20

- iO I

C t

_fo

-4o o

to 20 30 40 l

,, o 3o

( o.,,, - a.... ) ' 0 0 / 8 c

c c.r.

Fig. II-3 Error Distributien for the Seeker CEF Correlation II-6 t

a

~

. e

.+

I 0 s

6 i

0 4

0 I

t 2

no i

t I

a l

l e

0 5

0 8

5 r

3 2

2 g

0 0

r o

C

$:Tr:w 1

F I

I C

t 0

r 1

2 o

p g

I r..

n 4

i c

p I

p I

0 I

i

/

0 h

4 0

S g

e I

i e

F h

I r

t ac ro I 0 -

f 6

r n

e i

o c

i I

t l

u b

I 0

i 8

r ts i

s I

D ro 00 r

1 r

9 E

i 0

f 2 9

I i

0 I 4 t

l I

0 6

9 gH,4 i '

]1i' 1

C s

0 F.. I 7

w i

0 i

6 F_

i 0

t i

5 e

vru i

C ng 1

i 0

s l

4 eD T

i IC y

0 n

i a

3 p

mo C

i c

ir 0

t I

2 c

e l

9 5

E i

F H

I l

C 9

I ar e

0 I

1 g

n i

e F

G I

4 I

c, F

e H

h C

t N

o 0

0 0

0

~

r 0

s 1

3 2

1 0

o 6

e.

f

. jl-F H

n o

C

=0?.'

9 i

l 1

i tu f_

b 0

1 i

I r

ts D

i i

ro r

0 r

I 2

E i

0 i

g 3

1 i

0 1 4 0

mo s-lt i

lt l

v V

~

• ?

h October 1969 ny correlation (IT-h presents thaOnly 33% of the data vere predicted within -160% to Figure arrer distributica fer the Shippingpert k).

45% error. The rc aining 62% of ti.a predictions produced errer larger than -16C% and, therefore, are not tabulated in Figure II-5 As defined thus by Equatica 1, negative errers indicate an over-predicticn of e any correlation that yields negative errers is undesirable fer"gr,bbth design n

and LOCA analyses. The results given in Figure II i clecrly indicate the ne c c a a- * *."so.'.-r.----*-.

"-.-*'.'..*-..'..da.a=.*.>..-a.. u

.4..s-

'.'..a. c-.e'a-S

-: -. - ~

.s tien. Iigure II-l aise in:icates ;ne inaccuracies tha are pcccibic when a correlaticn based on data frc: a given sc etry is applied to c her gecretries.

The General Electric Cer;nny design curve correlation was derived by censtructing curves under CEF data ccisidered c c:ver the operating range cf SWR *s (that is, a Iceer envel:pe was determined). That the correlation is ecnservative (pcsitive errcrs) as intenaed is indicated by the results shcun in 71gure I~'-5

.-2nevar, the success of using this cethed of correlation reqaires data fcr the entire range of parameters over which the ccrrelation is to be applied. An e.xamination of the predictions shcvs that the ncncenserva-ive errcrs (negative er*crs) in Figure II-5 were obtained fer lev mass flux data (G < l.0 x lob lbm/hr-ft).

2 The results presented in.this section have shcvn that the method of 7

correlation used by Macbe:hs1 a->

• --att(cl produce accurate CHF correla-

/

tiens that =cy be extrapolated with reasenable certainty cutside the range.

of pcraneters en which they are based.

The develeprent of(nov heat transfer.

Therral-Hvdraulic Anelysis:

The Macbe:h 1) and Earnett(2) todels fer use in RELA?3 has ccntinued.

critical heat. flux correlaticns are being used for calculating the critical heat flux. Currently, the Earnett correlatien is being used for pressures above 1000 psia and the Macbeth correlation for pressures belov 725 psia.

For intermediate pressures an interpclation between -he two correlations is used. The correlations have been ceded and are currently teing checked s.

cut.

An extensive analysis Of the semiscale Test 813-2 (30% double-ended break)isbeingcade". The tes; data are being ec pared with RI'A?3 predictions made prior to running the test. Preliminary cc parisons shev scod agreement between predicted and ceasured prescures.

Cciputer Sciences Support: An explicit technique for solution of th=..""---~....".c.e"-..'..

-. - "....* *.... *.. *. - '.. ~. ~'.~.~. n..= '.s"...~='.'.*~.~~s'.n of a L1.s.-Wendriff technique. O:ic cchnique ;rcvide: a =cre stable and accurate solution cf abrupt chan;23 in bcuniary ecnditiens (that is, the channel inlet enthalpy change that occurs during flev reversals) than the precent explicit er implici: rethods currently in THETA 1.

This new technique vill replace the pretent enplicit technique.

II-9

~

(.,/

(/

au V0ctober 1969 -

B.

Separate Effects Tests l.

Highlights i

A preliminary systen design description document has been prepared for the two-loop Semiscale Blcudewn and ECC System.

A series cf channel bicekage experiments in support of F1ECET has been cenpleted 'en a 9-pin heater asseibly.

t 2.

Technical Ac-i"ities Semiset.1* ?2ev.detn and ECO (Sin le loop): The se:.iscale blevdown assedbly is presen.ly underacing ::ifications in preparation for Test Series II Grcup 2.

ciifica:icns ecncist of the installation of an electrically heated cere, the addi:icn cf three ganza ray densitometers to cbtain measurements cf aterage fluid density in the enre, a change in the locp ccnfiguraticn frec two rup:ure devices to one rupture device, and the installation of additional instrumentation.

Specifications have been issued to the Procurement Branch for 3-foot heaters for bicudevn heat transfer studies. Request for cuotations

- were issued October 27; tha bid cpening date vill te UcVedber 17, 1969

'A delivery date cf February 27, 1970, hac been c;ecified.

Semiscale Ele.dern ced E00 6.o lect): The t:c-lccp blevdcrn system is in the ccnceptual desi6n siege. A preliminary syste: desi6n descriptien document has been prepared for review and cctments. Work has begun on specificaticns for long lead time components.

t I?ine-pin Ch:nnel Ele:ka e Experic nt: A series of experiments has been ecepleted vi:h a 9,in ecre to s.udy the effect of flev channel blockage on eter ency ccre cooling hea; trancier. The experiments vers condue ei by ficciing the ccre at tenper :ures of 1600 and lg;;c? vith restricted ficv area at. he center of the cere. Restriction in flev l

crea was varied during the test series fre no restriction to 90 percent restriction. Results are being evaluated.

FHUST: The THUST core for Test Series II testing has been assedbled and installed in the FZUST vessel. Tecting is scheduled to begin the seccnd week cf ::cvember after the instruuentation has been connected and the cytten has been checked cut.

II-lO

-