IA-84-912, Responds to FOIA Request for Four Specified Documents.Item 1 Not in NRC Possession.Item 3 of App a Encl & Available in Pdr.Items 2 & 4 of App B in Pdr.Portions of All Three Documents Withheld (Ref FOIA Exemption 4)

Responds to FOIA Request for Four Specified Documents.Item 1 Not in NRC Possession.Item 3 of App a Encl & Available in Pdr.Items 2 & 4 of App B in Pdr.Portions of All Three Documents Withheld (Ref FOIA Exemption 4)

ML20127F265
Person / Time
Issue date:	04/24/1985
From:	Felton J NRC OFFICE OF ADMINISTRATION (ADM)
To:	Brierly B TAFT, STETTINIUS & HOLLISTER
References
FOIA-84-912 NUDOCS 8506250095
Download: ML20127F265 (3)

Text

~ ,

s %o,, UNITED STATES T>bR-Cb 9 , NUCLEAR REGULATORY COMMISSION r; a WASHINGTON, D. C. 20555

• \*****} APR <. 4195 Ms. Billie F. Brierly Taft, Stettinius & Hollister First National Bank Center IN RESPONSE REFER Fountain Square TO F01A-84-912 Cincinnati, OH 45202

Dear Ms. Brierly:

This is in response to your letter dated December 6, 1984, in which you requested, pursuant to the Freedom of Information Act (FOIA), copies of four specified documents.

We have been informed by the staff that a search of pertinent files indicates that the NRC is not in possession of item number one of your request.

A nonproprietary version of the document identified at number three of your request and on enclosed Appendix A is being made available for your inspection and copying at the NRC Public Document Room (PDR), 1717 H Street, NW, Washington, DC 20555. The document will be filed in PDR folder F0IA-84-912 under your name. Nonproprietary versions of the documents identified at numbers two and four of your request and on enclosed Appendix B have already been made publicly available at the PDR, and you informed Mr. Richard Lavins of my staff that you have obtained copies of these two records. Portions of these three records contain information which is confidential business (proprietary)information. This information is being withheld from public disclosure pursuant to Exemption (4) of the F0IA (5 U.S.C. 552(b)(4)) and 10 CFR 9.5(a)(4) of the Commission's regulations.

Pursuant to 10 CFR 9.9 of the Commission's regulations, it has been determined that the information withheld is exempt from production or disclosure, and that its production or disclosure is contrary to the public interest. The persons responsible for this denial are the undersigned and Mr. Harold R.

l Denton, Director, Office of Nuclear Reactor Regulation.

l This denial may be appealed to the NRC within 30 days from the receipt of this

! letter. As provided in 10 CFR 9.11, any such appeal must be in writing, l addressed to the Executive Director for Operations, U. S. Nuclear Regulatory Commission, Washington, DC 20555, and should clearly state on the envelope and in the letter that it is an " Appeal from an Initial FOIA Decision."

l Sincerely, J. M. Felton, Director Division of Rules and Records l Office of Administration

Enclosures:

As stated i

g 62 g 5 850424 BRIERLY84-912 PDR

Re: F0!A-84-912 APPENDIX A Item No. 3 10/74 NEDM-13377 Mark III Confirmatory Test Program Phase 1 - Large Scale Demonstration Tests -

Test Series 5701 Through 5708 (86 pages) ll 1

(

e Re: F01A-84-912 APPENDIX B Item No. 2 10/75 NEDE-21078-P Test Results Employed by GE for BWR Containment and Vertical Vent Loads -

Park I (172 pages) - Part 2 (212 pages)

Item No. 4 05/76 NEDE-13442-P-01 Mark II Pressure Suppression Test Program - Phase I Tests - (113 pages) i I

I i

t .

.r*"* '4

D ((h[ *

• < ...w

. s i. .

f

+.;'ag.,). y.,,."..

m w .:r: c ..iusm., / - . - z-:,)

f.U ., ;

h.TdN,.%;n],%g.%r.]

.. c.

.t I

• . ., f t ' *. .y '*.

' ' . 4" j%

a r m* *m . fg,.e .c 7 :l%

.p. f.2h,, A.,.;.c= *$LkSS II5 .*.j.*-

$h.;N;gstM. ,. fyl. L,g.s . . . Q 4NkMM

a. .a t.s $NihhNph...c....h.,:MMI!Mk,kMhk.I .; . r

" o . a. m. m .x , m -.

6;.o.%.x:'.;.3.s s' :;'m .1.m.d .a.x ygwo,Ge  %.%

. ~ . . . . . . .

t*y f,A: .,1- y (l ,- . .;

, c < .g . p:' r,:.(i c.'

r~r

• i . ' . f d .:

",'.~

^ .

L. ,s ,, _T- s - . ., c 3. ; p \ , ; f. w, .y, ., W-

• . . .. :, 6. .*Q  ; \ ; ' ; . . .' *. . .:. .. . . . . . . , r. ., +. .v  :

x . ,7 .. - :s *

~

z*.

e .*...

s

> . .6- -( .4 t.

, ,_ . . . .: , o,,.:;.; ~ . . . . ., -l.p.. . . ,,3 . .. ,, ' .

'-c ' n;,:h. h I'[ .,' '.- , .'..,; uO,I

.. n

.,.  ;...* ,. i .$*.[:.1,1 ' *., v.3 (.[ i f)M,V-$ . !$$. '. l-' *' ';, ' .' .' ).'.

+ ,1 ; '

2

. ,..9.a. .

q.. . . . : .m ,a . '  : .. . M sw .w .. '

W.:. &iW's.u,.;.M,

.c',s,*'.y:n.

-J.#

.av.N h ') ,

[ . ['.f- -

w.,. .;A.. Y

.e uk.:t q_s;ge+n: nh.h*.g~gQ-e.

1 '.';m,y  %. .w.

f(,.  :;..-> w .. : Q ~t.p .- q.Aggyq_ . .. p 4 p.s

.g  :,. .

'.g...s. .- -

l IFD2tec%encWET9mwA1agwmag . g;y __

=

w ,

i G ks W

e. 2.. :ver., y m . g:

w w swM .

M.,c,m @,m.:@a A%m#@v..m. p n%:  %.m Q.- .X.c.=.LW s.~

t. . . _m.M. m.s  ::. 8. .i.. e,m. . . . ._., . . m e w w. M . .. m,..s ., , A . .. ._J .R. mM . .K.u. l.l. .c.

. . . c.yi=.m.1 CONBRMATO ... y,.w;5 m y.m.,RY..w a 4Em.e.. J R.O v G. . A._M,Mnp

.@i F.

g g gp g.~.gg.gg . .c .

i

.p;. p. yy. 9,g,g.:pgg ggg p[q

.s .-

.a p

2. ".; del %ONSTRATiON TESTSi F M

.s/,

~ ~ .. . . '

La t* -

' TEST; SERIES U701 THROUGH 5703 :

~

. ..+ . .;. ,

.-... .~ ..

s.

_ .  ;,. 25

. . . " ' .

• e \:L. L. MYERS  % .

f_

.b y, , .,"

s.

,- - g '. , ;. ... . I ;.*',,./ .);';5.T.R.MCINTYRE . , e,,,  ;.

,- .g,4. g m. . ,

~3- .

'T**;

. .. f. ..

.. ?,

r7.K.'.*,?*.*'cc*a?f"

,;a s .**;' .T'*s*jia -

. '7-S:'- WT'*q .

. . . . .- i.,

k m :.

s.

r * *.:. :.-13.5.ar,?w.W.. ;'w'T*W.y;;.zM .r :i:D -&.,1 "FyogM

. ~

L.h.6[,,,.3.- .;.:; ",.kG;.;y.n V4- gk.',',',,.'s.N*'w'

. ,:a s, g ,

. .; 1

..' g.-

. S. >- t

• - '? ..s. ..

c-. ., .

/ ,

c . 1.;* . . . . . .; .m. ,. . ., . , ,;.

.,g .v.,e es. .- . .. . . ,

.. , .e... -- ., . .. . .

..g 35 ,.. .> g..

~

s,s

.s.... .c, i.r., .# , . .,s

.- ., g, .

m. - ..

m , w. ,~. . ;.,~......,,....,f .

u, m.v.

.t

,. , the Gwa: Livine Co.:.my are nist tw he. sed w th c, e .

.3 s

<.. . . . ,s..

.n n. w . ,c. . . u. .,., . ~

..u:. .

,e u m . ,. . %. ~. , - ., m ,....v. uce m ,s

lj . g..

....x. .. v,.. .. .s<.w.

..~.a .

. -  :. s.: w. . .

' 3

.. 1 P ~

.- . .. ,q

.e

.. . .* ,.,. .q..

. .. JETULH lb s.. .uu w w. co k, n . .

. . u. n c.

%: 5 4. . .. . . ,. .< s .cu ,.

. a. . w a .

- ~......,<... . . .

.. s ..

.n . . .. . ..~. ; ...s.. y . 3, .;.e.,:. 7. . . ., .. .V., ai.. O.G.E N E E..A L , x?. E. L E'C. n T. R ig p'

i' **,Ig c" ,

, *., ' ' *_-k* J e t s ..

. 5...

• .Y . , ' . .

L

%.M'~ .C2i%. . :.Ku .O':. D.".: .El.W:'.2An:.:a .:.2 D.O 3.'c ... .

.I t

' A

/

j ,,h

~

LEGAL NOTICE *

, This document contains proprietary information of the General Electric Company and it is not to be reproduced or furnished to third parties nor the fuformation contained therein utilized, in whole or in past, without the prior express written permission of tiw General Electric Company.

Neither the Get,eral Electric Company nor any of the contributors te this document makes any warranty or repre-sentation (exptessed or implied) with respect to the r.ccuracy, completeness, or usefulness of the information contained in this document. General Electric Conpany assumes no respon-sibility for liability or damage which may result from the use of any of the infotwtion contained in this document.

. I

. l 11

u l NEDN-133U i , .

_ COMPANY PRIVATE CLASS III MARK 111 CONFIRMATORY TEST PROCIWt FNASE 1 - LARGE-SCAI.E DDIONSTRATION TESTS TEST SEMIES 5701 THROUCH 5703 TARLE OF CONTENTS ASSTRACT Pag bli

1. INTRODUCTION 1,

11. SL?9tARY 2

Ill. TEST FACILITY DESCRIPTION 3 A .' Fressure Suppression Test Facility 3 B. D'ata Acquisition Capability

. 2 2 3

C. Test Instrumentation and Data Handling 9

IV. IEST RESULTS 13 A. Steam Generator Response 16

1. F. essure Decay 18

2. Level Swell and Carryover

• 23

3. Blowdown Flow Rate 27 R. Drywell Response 31

1. Pressure Response 31

2. Temperature Ec.conse 39 C. Vent System Response 39

1. Vent Clearing Transient 41

2. Vent Flow Coefficient 47

3. Vent Closing and Ven't Chugging 50 D. Pool Response 54 i

1. Fool Swall 54

2. Japact Loads 61 E. Containment Response 64 R, ITER.ENCES APPENDIX Venturi Flow and vessel .w. ass Calculations DISTRIBUTION D-1 111/1v a

i e NEDM-13377 l , COMPANY PRIVATE CLASS III LIST OF ILLUSTRATIONS Figure Page,

1. Pressure Suppressinn Test Facility Schematic 4

2. Data Acquisition Systen Block Diagram 5

3. Conductivity Level Probes 8

4. Test Series $701 Inst rumentation 10

5. Test Series 5702 Instrumentation 11

6. Test Series'5703 Instrumentation -

12 -

7. Steam Generator and Blowdown Line 17

8. Steam Generator Transient Comparison - 5702/12 & 13 39

9. Steam Generitor Depressurization Transient - 5?91/3,8 & 13 20

~~

10 Steam Generator Depressurization Transient - 5703/2 21

11. Steam Generator Nodal Densities - 5703/2 22

12. Steam Cenerator Level Swell 25

13. Venturi Flow Rate 28

14. Critical Mass Flux 29

15. Total Vessel Mass - 5703/2 30

16. Blowdown F1sw Rate - 5703/3 32

17. Drywell Pressure F.esponse - 5701/10 (short ter=) 33

18. Drywell Pressure Respense as a Funct.on of Vent Submergence 15

19. Feak Drywell Pressure 36

20. Drywell Pressure Response - 5701/10 (long tarm) 37

21. Drywell Pressure Response - 5702/12 (long term) 38

22. Drywell Ter.perature Response - 5702/12 (long term) 40

23. Vent clearing Time - 5701 43

24. Top Vent clearing Time - 5702 (leaking seal) 44

25. Top Vent clearing Tiec - 5702 (non-leaking seal) 45

26. Vent clearing Transient - 5703/2 46

27. Level Probe Trip Times - 5703/2 48

28. Annulus and Vent conductivity lennor Data - 5702/12 52

29. Vent Duct Flow Rate - 5702/12 53

30. Conductivity Sensor Data - 5701/5 (Raw) 56

31. Conductivity Sensor Data - 5701/5 (KcJu:cd) 57

32. Maximum Fool Surface and Bubble Velocities 58

33. C.sntainnent Pressure Rise comparison $9 v

m. __.-.--,- - . _ _ _ . - . _ _ . , _ , - - - - - - . _ _ , , _ _ . . . . - , .

NEDH-13377 COMPAhY PRIVATE CLASS III LIST OF TABLES Table h -

1.a Test Initial Conditione and Data Summary (English Unita) .14 1.b Test Initial Conditions and Data Summary (SI Units) 15

2. Steam Generator Level Swell 26 3.a Impact Pressure Data (Engli.n Units) 62 3.b Impact Pressure Data (S1 Units) 63 4.s . Containment Initial Conditions and Data Summary (English Units) 66 4.b Containment Initial Conditions and Data Summary ($1 Units) 67 G

l e .

g ,

4 e

I

} .

• 1 NEDM-13377 l CLASS 111 l' ,

s  : C090'ANY PRIVATE l l

AB$ TRACT Three series of tests vs rs run in support of the .'tARK I!! pressure suppression containment concept. The tests were performed at the Pressure Suppression Test Facility consisting of ae integrated system of stess generator, dry.vil, vent system, and supt ession pool. The volumetric scale factor used for faellity design was nominally 1:130 based on the BWR 6/251-series MARK 111 containrent design. The pool and vent system both represented full-scale mockups of an 8 sector of the MARK 111 containment including a vertical row of three 27-1/2 inch (697 mm) diar.eter horizontal vents. Tests with one, two, and three vents open were run for various values of simulated break size and top vent centerline

. subr.e rgence. The transient response of the stvae generator, t rywell, ,

vent system, pool, and wetwell air space were seasured and compared to analytical models used for predictions of loss-of-coolant accident translents.

l l

a i

i I

v11/v111

,e,.< - - - - - - -- - - - , - - - - . , - - - - , , - - - - - - - , - - - - - ~ . - - - - - - , , . , . - -

g NEDM-13377 l CLASS 111 Page 1 of 69 COP:PANY PRIVATE I. IN1RODUCTICN Three series of Large-scale tenonstration Tests were run in' nuppert of the horizontal vent pre 9ure suppression system used in the MAkK III containment design for f.eneral Electric BVR plante. 1his paper reports the results of Test Series 5701 through 5/03 which were run in November 1973 through March 1974.

The MARK III Confirmatory Test Program was undertaken to verify the design basis .and performance of the MARK 111 horizontal vent system.3 This test program is an exter.slon of the early pressure suppression testing done by General Electric in support of previous containment designs and consists of both small scale and large scale test programa.

The Small Scale Test Prograe. was initiat d8 l 73 *"d ha' provided data for analytic model develope.cnt.5 3*2 ' Investigation of vent clearing phenoecna was emphasized in this program since the major dif ference of MARK 111 from past pressure suppression containments is the vent configuration and the way vents are cleared.

The Large Scale MARK 111 Pressure Suppressien Test Pregram, which, was initiated in the fall of 1973 as a final confirmation of the analytic modela used in the ZARK I!! degign, is a continuing multi-r.hase program being performed in t? e Pressure Suppression Test Facility (PSTF) et the San Jo9e site of the Boiling Water Reartor $ystems Pepartment.

Phase 1 of the program includes Large Scale Demonstration tests to demonstrate short negeents of blevJovn perfermance using full-aire  ;

l 27-1/2 inch (0.7 m) diameter vents with an 6-degree set tor cf 9uppression i

l pool. Supplemental phases of the program will provide arametric data for rertnenents of analytic design models and optimization of vent and poo?. geometries.

l

? -

4 WEDM-13377 CLASS 111 Page 2 of 69 g TANY PRIVATE 4

v*

I e

\

I

\

! \

r i

i l

n I

I, t

t t

l i

e s I

1

• e l

I h

1 l .

.i . e i

k I

/

t /

l 1

1 j

1 r

/

• e

[

i

- - . , , e

s

l

\ .

l, I

l' afEDN-13377

- l CLASS 111 4

Fage 3 of 69 t COMP.4Y PRIVATE

_III.

TEST FACILITY DESCRIPT!,0'f I i A. Pre,ss;Jre Suppression Tent Facility The large scale pressure suppression tests reported here were performed at the Pressure Suppression Test Tacility (PSTF) which is located at the furtner Avenue site of the Nuclear Energy Division.

j. .

In brief, the PSTF as configured for these tests consists of a

.l 160 2365fft3t.3(66.98 (4.53mm3g) electrically drywell by an 8heated inch (194 pressute mm 1.d.)vessel blowdewn connected line to a

which includes a critical flow venturi, rupture disc assembly, and an

,8 inch'(194 mm) gate valve. The blowdown line is connected to a *

, 10.374 inch (263 mm 1.d.) dip tubo inside the pressure vessel which

, schedult allows steam 40 riser blowdowns which allows and steam inside the to be drywell injected to a at10 theinch top (254 of themm 1.d.)'

'drywell space. The dryvell is a cylindrical vessel having a 10 foot

' (3.05 m) diameter and 26 f oot (7.92 m) height with a 6 f t (1.83 m) diameter discharge duct which enters the pool building and is connected to a test section which is a full size mockup of a single row of three 27-1/2 inch (698.$ mm) vents on 4-1/2 foot (1371.6 mm) centers.

The suppression pool bu!! ding models both the pool and vetwell air space and has a total volume of approximately 20,000 ft3 (566 mJ ). The fraction of this volume containing water will, of course, change as ghe vent submergence is changed. Fool baf fles are used to siculate sa 6 sector of a HARK !!! suppreselon pool.

1he nomins! volumetric scale f actor for these tests is 1/130 although the vent system and pool cre apptoximately full scale.

Figure I shows a schemat' view of the PSTF. and a complete description may be found in P *erence 1.

..- , . . e, s "'

B. Data ArQUlsition CapAblI!!Y A coeprehensive measurement and data acquisition system is provided at the Pressure Sappression Test Facility. This da:a acquisition system is shared with the two test stands of the Nuclear Safety Test Facility and the total date acquisition nystre is shown in block diagram form in Figure 2.

The data. acquisition system is designed to sample data and record it on magnetic tape at a maximum rate of 8000 total measurements per seconJ

, . 1 e

4 e

_ _ . . _. _ _ _ _ _ _ _ _ _ __-_.__._-____._r_. _ _ _ _ _ _ . _ _ _ _ _ -

._, e SPOOL SFJCT10ltS 8!

p RISF2 r yi 10.000 FT ( 283 ) UETWELL AIR STACE .

e ir n Tusz gi 10 rstc (68.9 Ku/ ) max.

,e i-rtou cowTaot I.

e 380ZZLE e .

ee 8 p-- , ,

' [

u _- -)o '

]

s ge s._ - - ; im auen,xt D:sc I SUFFRESSION F001. SATTLF. >

> AsstMBLY

( f'y OFNABLE) 24 FT (7.32 M) li!CH IN JLOC0kN LINE - . _ . f

• TE" 5 i

glCRADE pg,y

- t -y ---

-_:-:- '^: _ :: . -

A 47"(1.19M)1D I 14.5'(4.42M) 10'(3.05M)03 1 26'(7.92M) I 1  : 19 F'  :

i j l l (5.79 m) 160 FT 3 (4.53.f) 2365 FT3 (66.9 M ) g 2

1160 PST: (8.000 KN/M ) DESIGN 75 PSIG (517 KN/M ) g I SUFFRESSICN POOL (DESICN) g {

l '

T nou Zo tar. vtxT

, e- 4 N

TEST SECTION - ,.s FT ct.37 n) ,

• l ~

(RDIOTABLE) g ~{--_~ <L

• y I 4.5 FT (1.37 M) s, yng i

r- -

4 3 5g

- t 3 e- " .!.

% 5" Figure 1. Preesere Suppreestorf Test Facility Schematic. g

' ItEDM-13377 .

1 CLASS III l

', Page 5 of 69 COMPA.W FR!VATE

,d E! h i! 91

!;!! !I.

3/jlin i [q!ili il!

!.ls(l.!11,;

llhii2Iljj I1 f l l_ J _ _.

I, il

- ' ]

sI ,j

..! li r!~ !!I f l.dj h ,

!!!!.Ik! !j['d j!!hli!

' ll[,ti,li:.!,'!l 2 n. .

N .

! no, .

i}

3 .. . n u a ,. .ll ; ).1 j

~

i,

..m___3 , lit

!II 'j,'T 'Ieft J Il4' fI ,

tis llel f _ _

f

e !I l .

I*I

' m

._. i I E ll.

II  !! I

. -- i T

M lL li!

3 5

~

l

11 .I .I E L l 7;dlr,diil.1 a

11 lpi' i,i

' ~

O l' l ll 1

{

3 i  !! rlig ni n .

til,. '1 g i. i

l. I:l .

r A i .11 :1n] trti l s '! i te 4

, _ . - > w ii

l. .

'*)

{ .

d e

9 e

9

NEDH-13377 Class III October 1974 COMPANY PRIVATE HARK III y CONTIRMATORt TEST PROCRAM PHASE I - LARCE SCld.E DEMONSTRATION TESTS ,

TEST SERIES 5701 THROUGH 5703 L. L. Myere

,- T. R. McIntyre R. J. Ernst Raviewedt g Approved

. E. Townsend, Maneser R. T. Lahey, Me ger Containment Systers Core and safet Safety Development Devc.'opment This doeurent contains 82 pages which in6)ude forematter pages i through vii, I through 69, and Appendix A.

TUiDiCTITDntrATTCN17.Tra. wa r. rvi s i e ce ss n 46 a u YT4.TtutiFAW san Jost,cautoama tun GENER AL h ELECTRIC 42p/56c.tWRSD .

JE.10/74 4,

l l

l NgDM-13377 '

CLASS 111 Fage 6 of 69

_ COMPANY PRIVATE with inputs from up to 123 channels which are dedicated exclusively to the ,

PSTF. Data is recorded in blocks of 16 channels with each block requiring l approximately 2 milliseconds for a eneplete scan. Since 7 blocks of data l were recorded during the testa rep 9rtcd here, a complete scan of all data channels was made every 16 to 18 ril11 seconds.

The principal components of the dlaital equipeent for data recording and output ares (a) High speed multiplexer and analog-to-digital converter (30 high level channels and 48 low level channels currently dedicated to PSTF), 13 bit, including sign, with programable gain. The full scale input voltage is + 10 V with a gain of 1, with a masimum gain of a for high level inputs and 4096 for low level inputs.  ;

All low level inputs have an input illter (RC-pasolve) of 10 Ha cutoff frequency. -

(b) Computer, 32K core, with two-channel direct secory access option, time base generator, and floating point hardware.

(c) Dise storage subsystem, 2.4M 16 bit words with one f ase J disc and oae removable cartridge.

(a) Two magnetic tape units, 9 track, 803 kPl. 37.5 inches per second.

(e) Teletype.

(f) 1.ine printer, 500 lines per minute,13J columns.

(g) Photoreader for punched paper tape input.

In addition to the digital equipment, analog output may be taken on two 8 rFannel oscillograph-type reccrders. These recorders base high cutoff f requencies (on the order of 2.0 KHz), and thus are well suited to monitoring transients into the milliseconj range. For the tests reported here, the/ were used prirarily for visual monitoring of experimantal performance.

Pressure measurements were rade using strain-gage type pressure transducers. Differential pressure transducer

• were used for all measure-ments outside of the pool building, with one leg open to the atmosphere for sage pressure measurements. Inside the pool building, waterproofed trans-ducers were usedi differential type for flow an1 level seasurements and absolute pressure types for pressures. References S and 6 contain discussions of the instrumentation ccamonly in use at the NCTF.

a e

NEDM-13377 CLASS III

_ COMPANY PRIVATE Pressure transducer signal conditioning includes 36 strain gage excitation, balance, and calibrate moJules. Low level applifiers are also supplied for each channel to anplify the transducer cutputs to + 10V f ull as.ile. The arplifiers are dual output with 100 mA unfiltered output for the analog recordern and a SmA filtercil (3 pole) ot tput for use by the digital data acquisition myntes. 1he (11tec cutof f f r<quency is adjustable f rom the front panel with a 9 pos'tlon switch, from 3 Hz to 100 kHz. With the exception of impact pressure data, where 10 kHz was used, the filter cutof f

, frequency was set at 30 Hz.

Temperatures are seasured with 1/8 inch (3.175 mm) 0.D. stataless n

steel sheathed, ground ed tip, iron-co'stantan thermocoupjes. Siggal conditioning for thermJcouple Channels consists of a 150 F (65.56 C),

100 channel reference junction. Thermocouple data is recorded using low level input channels on the multipiczer and, hence, the 10 Hz input filter frequency is applied to this data. 48 channels are available for

. the rmocouples.

Liquid iny 4, face movemvits are monitored by the use of conductivity

, fype sensors.T -

4 1

% /

Sketches of all three level probe types may be found in Figure 3.

In addition to the meamurement channels listed above.12 high level data channels are dedicated to auxilliary inputs. For these tests, the channels were unto for a Thunder Scientific Corp. model BR101R Brady array semiconductor humidity sensor which was used to s.easure the vapor content of the wetwell air space during eac's run.

. 5 s

I l 13Etti-13377

, . CLASS III CnwPANY FRivAJE E

I, i

I l

i r

l l

l i

t l

/'

l

' i l

A f

--n-...n, , - , , , , , - _ _ - - - - _n-._.n_._,-.-.m, -_ __,

~

NEDH-13377 CLASS III Page 9 of 69 COMPANY PRIVATE C. Test Instrumentation and_D.sta llandling Tent inarrutentation used during test serien 5701 thrnunh $703 are shewn in Figuren 4 through 6. respectively, process instrumentation and controls are nur ahown.

Most instrur.entation shown in the if Rures is standard and self esplanatory. The level probes and humidity senscra were described in the previous acetion. Static and dynamic pressures in the vent pipes were measured with 1/4 inch (6.35 mm) 0.D. pitut-static probes reinforced to 3/8 inch (9.525 mu) on the shank and inserted approximately 12 inches (305 mm) into the side of the weitt pipe. Static and dynamic pressures were measured and later added together to get stagnation pressures.

Not shown on the P&lD's, but a valuable pource of qualitative data are motion pictures and video tapes of the suppression pool surface during blowJuw1 runs.

• Initial evaluation of test data was performed shortly after each blowdown run f rom an on-line printout of raw data in engineering units.

This data, stored on magnetic tape, was then taken to the Hoceywell 6070 computer for final processing. A listing of the program used for data reduction may be obtained from the authors on request.

4 6

4 I

e 4

N GM-13377 CLASS 111 COMPANY PRIVATE _ Fase 30 of 69 h' N' khI O e es e 1 -

1 G

r ,

3 0 0 0 h@ O 3 'e@ e s <

E O O Oje00000O 3

t  :

O O O 8

• k E

, @ht* L i

. iii i l O~Y h ~

O

. 1. . .~ s

%Q4T

~~

, , g s

V ..

g I

I 0._

r i -

.m s '

e 'd e

, X_ .

y O- E 4=1 e y l - .:

o. . r e

':4) 7 _.

l55t?""

,4

' ,5 g3 s

& ( .. ~ i.~. ' 7 - --

25ee33 4 ......

~O &>& Q>&-kqu_& k n H sn oboe y

did

.i l MEDH-13377

- CI. ASS III

. COMPANY PRI_VA1E Page 11 of 69 h' N ' hhI

b e es e

' - l e 4 E

3 O O !3 O O O

% 1 g t i g: 1 O

'@O '

h3 O O O

%. O O b O O O

!l i s e Q O  ; O O O 9 4 U

- T %L d i YY I

~ _4 __________..h _ . . .,

\

h i

~~~'

G i @ @O -@

r kVTA I. S e

,w j . . __ . __

t

\

r --

e e. e ,6

$ <x, j

yk

& ' . ~_,[~_:t_".n .Y b-nsae

• UUU"

--e + + ow% -e- EU

,;;,,0$;f

, . & ao=a.

e 8

7

!.ED.'l-1337 7 4 C1A55 III l COMPANY PRIVATE Page 12 of 69 h hhI b

i g,

f

-l g

4 .

o /

O

e <

l

• C. i l

E l l

- 1 p. '

1 E .

)

R <

3 <

t l et  ;; r l l @ @

f 9g ggef O p -

3 . . c'i i c

i lp+p, V'_MMMOY W.fi r i_ '

a , .

E ....._2 t

N &W C

, Y -- n,

_g t __

e- q.___ _ _ ___._.

____3 e e 6

e >l i e- I t

6 s 4=-

• _ ~. d ,

c-

NEDM-13377 ,,

CLASS III Fage 23 of 69

_COMPAhY PRIVATE IV. TEST RESUI.TS 1hree series of tests including a total of 41 individual steam blowdown runs were performed at the Pressure Suppression Test racility during the period f rom November 7.1973 throuah Nurch 13, 1974. These

• testa provided transient data for the large-scale demonstration phase of the HARK 111 Confirmatory Test Program. ,

Test series 5701 consisted of 21 steam blowdown runs with the top two of the three horizontal vents in the full-scale test section plugged.

With this configuration, the vent area open to flow represented the scaled area of the HARK 111 dealan with the nominal scale factor of 1:130. Test series 5702 consisted of 17 steam blowdown runs with the top vent plugged. -

Test series 5703 included three steam blowdown runs with all three full-size vents open to flow. The scope of this latter series was reduced considerably and instrumentation was minimised in order to espedite work on pool swell testing and still demonstrate vent clearing in a three vent system.

The principal' variables for these runs were pressure vessel blowdown rate (achieved by varying blowdown line flow restrictor diameter) and horitental vent initial submerg nce. The initial renditions ar.J a summary of test data for each run are summarized in table 1 (Run 15 in Test Series 5701 was not reduced due to probleme with the data tepc). Th? internediste-size 2-1/2 inch (63.5 mm) flow restrictor represents the scated area of a steam line , Loss-of-Coolant AcciJent (l.oCA) in a 251-site plant and the

• range of submergences bounds the anticipated HAAX 111 design submergence.

I The first two runs in .est Series 5701 were initiated by action of a quick-opening (250 maee.) valve; all other runs were initiated by rupture disc action because of operational problems with the valve. .

Evaluation of reduced data from the tests indicated that the riser tube in the steam generator had leaked during a significant number of the runs permitting an undetermined mass of 11guld water to be discharged to the drywell during the first part of the blowdown transient. The subsequent flashing of this liquid mans resulted in a' higher effective charging rate to the drywell than is experienced during vapor-only blowdowns. The effects of this 11guld mass carryover are discussed in the sections concerning steam generator response and drywell response.

The runs affected by this riser tube leakage are noted in Table 1.

O e

n 0

l r i . i

- NEDM-1337;

]  ;*

l class aar Co*An ra Ai r. 4 et te i .

ll I

O l 1 l

I

'I is Ia l n u

I l

E l

l

/

/

\- I i

i / , '

9 g

U

,.M= .-

v

, 1 1

l L

I .

I NEDH-13377 8

Ct. Ass III I I ' Pese 15 of 69 4

i a

1 s

I 1,i .I s

E i

l l

l l

I

/'

l d.

-l I

6

hEDM-13377 C1. ASS III

• Page 16 of 69

. CO:#AN*,' PW lVATI.

A. St eam,0c,nerat or Response

'the PSTF ste in generatrar t re.icnt is ut intercat beraune it in the stean f low int o ti:e drywell whid, en=ent ially not h the responM uf every-other icer pone: r in a pressure nuppre-wlon mystem. Blowdown flow rate sets the dryvell prcanurization rate, which in turn in reflected in the vent c1 caring transient, drywell niming, and pool dynacles. In addition to being the driving f orte behind a preneure nuppresnion test, the blowdism flow rate and vessel depressurization rate are of keen interest in reactor system LOCA analynis and ECCS equipactit design, and the PSTF, being a large critical flow carcriment, is a valuable nource of data in these arean.

Figurc 7 is a. detailed scher.atic of the PSTF steam generator, blowdown line, and annocia'ed instri:mentat fon. The Anstrumentation was stellar for all three test serl,5, the changes being the use of a venturi entrance to throat AP measurement in test series 5701 and the deletion of the vessel lower plenus: pressure bring test serten 5703. The two venturi pressure measurements were uwed en generate flow rate data and the seven pressure vessel nodal AP's were used for water level, density, and total mans data.

A suseury of the ve:nturf flow anj nodal density caitulations may be found

• In Apg endix A.

Since the de*1gn basis accide-nt for the MARK lit containment is defined as the instantaneous guillotine rupture of a main steam line and the first three Leut series ve re confirstatory in nature, the PSTF steam generator was set up for steam-only blowdownn. As may be seen from Figure 7 thts required installation of a dip tube within the stea'r: generator to take steam f ron the upper end af the vennel and still utiltre the vennel penetration level with the blowdom !!ne, which was below the water surface.

. All blowdowns were initiated from a prennure of 1050 psia (1240 kN/m2 ) and a vessel water level of 4.0 icet (1.22 m),

f Standard tert procedure was to vent the dip' tube a few minutes prior to the start of a test to terwve any water that might have collected.

Analynis of the data han shown, hnweier, that in a majority of the blow-dovne perforried .i slug of saturated water was present in the dip tube when the blowdown was initiated. lance, in these cases, a slug of saturated water preceeded the steam flow. Suhnequent investigat'lon revealed the seal between the dip tube and vessel nonle to be Iraking, which allowed the dip tube and blewdown line to the rupture dise to rapidly refill with water af ter the venting. Although the size of the water slug is in general unknown,1.s nine mSy he bounjed by the total volurre in the blowdown line to the rupture dine anner,bly and in the dip tube to the steam gererator water level. Assu .ing the voltme to be f f!!ed with saturated liquid at 10$0 psia (7240 kN/m 2 ), the euxleus rass of the water slug in 125 lba (56.8 kC).

From critcal flow considerations, however, the true water pass is estimated at about 1/3 of this value.

1 4 9 i

V i .I f

I Wan swF I

• CLASS III I 's Page 17 of 99 COMPANY FRTVATE Till PRESSURE VESSEL i

i i, . ,

gDIF TURE '

[

f .

~ '

i .

g 4 FT WATER LEVEL

. . . . , )

l SEAL YENTURI SF00L RUP1URE DISC AESIMBLT y .

o g ,, ,

VENTU I i

F101 - VESSEL DOME PRESSURE

!f Of F102 - VENTURI INLF.T FRES$URE N2 F115 - VENTURI THRDAT FRESSURE

. D103 - TOP MODE DIFFDENTIAL PRESSURE D104 - 2ND MODE DIFFEPENTIAL FRES$URE D105 - 3RD NDDE DIFFERiNTIAL PRESSURE

, D106 - 4TH NODE DIFFERENTIAL PRESSURE 9107 - STM NODE DIFFERENTIAL PRESSURE D108 - 6TH N0DE DIFFERDTIAL FRES$t'RE

! D109 - BOTTOM NODE DIFFIRENTIAL FRESSURE D110 - VESSEL OVERALL DIFFERENTIAL PRESSURE Till - VESSEL DOME TEMFIAATURE

. T112 - VESSEL thWrt PLENUM TEMPERATURE d

Figure 7. Steam Generator and Blowdown Line.

0

NEDM-13377 j CLASS 111 j Page 18 of 69 Co.TA*.T PRIVATE Ironically, althout.h the saturated v.ater o. lux precceding the stearn blowdovu has a r.ajor effect en the dr>wll transient, the effect on the steam generator is s'r.all; there is a delar in the initiation of tiie stcan blowdown due to the t i e reqaised t o a 1. :.r the writer " ro n the ble down line.

This is demonstrated by Figurc 8 s-ht.s. J vs th. vessel .lorte ,.rctsure and venturi throat pressure lor lest 5702 runs 12 and 13, eso with a 0.18 second time delay applied to rua 12. From a steart renerator transient viewocint, these two runs are identical; i.e., steam blowdowna f rota 105t1 psia (7240 kN/s,2j and 4 foot (1.22 m) vessel water levels, In.itiated by '.uptune discs. The only difference is that daring run 12 the seat leaked and in run 13 it did .

b

1. Pressure Decay Figure 9 glves the vessel doce pressure history taken during test 5701 runs 3, 8, and 13, which utilized 2-1/8 inch (',4 cm), 2-1/2 inch (63.5 en),

and 3-5/8 inch (92 rn) venturls, respectively. For a given venturi size, the pressure vessel depressurization transient was repeatable (adjusting starting tices for water plug removal) an! thus the curves shown are typical.

ad mumm.

The transients are shown in detail in Figures 10 and 11, whith are the vessel dome and vent art pressures and the nodal d(nnitie8 trem Test Series 5702 run 2. This has a siie&pi hiovdown through a 2-1/2 ti,ch (6 3.2 =m) santuri. Figure 10 shod she sa6 prer-sure behavior .as e een in the previces tests. Note fro:2 Figure 11 that st the beginning of the blowdovn% des g. f of 'j I through 4 are filled with satorated vapor, codes 6 and 7 ate filltd with f saturated liquid, and node 5, havio; a density slightly less than that~ 'g ,

g of saturated liquid, f 4 partial!y (flied with water and the ren..ining volunc is iflied with c:eam. /

tuum i L

s i -

flT3'r *o thic point the water evaporation at the **8 Iree7urf ace was TeTP%an the venturi flow, resulting in a'ner degessuri-zation of the system. Once the vatic- filled nodes be,;ia to flash and errand, ,

the vapor formation rate cuss be greater than the venturi flow rate for repressurization to occur. , ,

t l

l -

NEDH-13377 CLASS III Page 19 of 69

-J o

e O

M l

l l

/

/

l I

l

. - J .

L .

1 l.._--..__-..- - - ,-m.

7 1

i i

l t

m r.

R */

BJ C=

re O **

Tom - sec  ;

L -  %

e 9

0 I

i I

p i

d 6 ee A

l M

S T*

j m I e.

/

C.

O is o

e O e

s

+

t e '

I 0

. < f

_ l 1' \

t I

i 1

f en i

6 I

N I

L M

ft t.o o

a g . ..

P l

I W

r i

l e

a l

l I

l

U 1 NEDN-13377 I CIASS III Fage 23 of 69 CO$ N4Y PRIVATE In sumury, the steam gener.stor pressure transient is regular and predictable, and the behavior in the first second of the trenstent may be explained as a departure f ro:n thernedynamic equilibriu:n that is necessary to lower the saturation temperature suf ficiently below vessel water tec.perature so that flashing of the water to steam cay occur, followed by a delay time required for acceleration of the f Juid in the lower plenum.

2. Level Swell and Carryover When the water in the lower plenum of the steam generator pressure vessel flashes, the vapor bubbles which are generated below the liquid surface cause an increase in the specific volu=e of the bulk fluid and the liquid level is forced to rise. The phenomenon has been investigated previously (Ref. 7) and is generally referred to as vessel level swell, f

0

NEDM-13377

' CLASS Ill COMPANY PRIVATE The level swell data is not as repeatable as the steam generator depressurir.ation, due prl::arily to the inherently nalsy pressure signal generated in the flashing two-phase fluid and to variations in the initial water level. Table 2 gives the tires that the two phase mixture reaches the indicated positions in the vessel. Only those blowdowns which yicided reasonable data are included; those runs which had very noisy density measurements or slow scan speed are not included. The error band on this data is about + 0.05 second.

f i

/ (!

W$

1 1

t

M

/

m I

E DM-133M s

CLASS IIg Y*Re 25 of 69 O

9 i

l l

l'

/

e

.I .

y . - .- _

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

g __ _ _ . _ _ . _ _. _

l .

i I  %

}

NEDM-Il)?7

' l Ct. ASS !!!

. Fase 20 of 69 l l

l l

l i

I a

l l l

~

1 _

NEDH-13377 C1. ASS III 88' 27 of 69

_CO?fr gT PRIVATE 9

3. Blowdown Flow Rate Figure 13 give - typical blowdown flow rette histories f or the t hree venturi stres used during the testa being reported. The peak blowdown flew raten vere on the order of 50 lbm/sec (22.1 kG/sec), 70 lbm/ wee (31.8 kc/.ec), and 150 ! bra /nec (68.2 kc/sec) f or the 2-1/8 inch (5t. rm),

2-1/2 Inch (63.5 cc) and 3-5/8 inch (92 ra:i) venturi 8, respectively. In increasing sire, the venturis correspond to 72;, 100%, and 210% of the scaled area of a single Bk'R-6 anatn steam line.

By dividing the blowdown flow rate values by the throat area of the venturi usef, critical mass flux data stay be generated. Figure 14 si.ows critical mass flux as a function of vest.el do::ne (stagnation) pressure for all of the blowdowns perf ormed during test series 5702. As may be seen, i

the data is very nearly linear and data frcm the other two test series fall on the same line. The scatter in the data at high pressures is f probably due to clearing of the water slug during.the first second of l the transfeat.

t

/

s M

t I

i

~ - - - _ . - _ " - - __, __

y l

4 W enemmes Figure 13.

I t

i i

fi

i I

I w

t 4

5 NEDM IJ3U CIASS III Page 29 of 69 9

l

l l

l

/

i l

1 i

l I

?

./ *

/..__,

W

)

l

(

l l

e ey%9 I

l

\

l 1

I. n~. u.

f l

I i

I l l

j .

,v, - - - -,,- ,,e ---- - - m~n-v--- .mn,ne- ---,c-,-m-r--,

\

I

'

• MEDM-1337F

) class 111

/ Fase 31 of 69 CojfANY PRIVATE

~

I'""

f up8dl

' #F ammma B. Drywell. Response The drywell in a pressure suppression containment systen serves as a passive surge volume for the blo.'Josn flow being charged into the drywell before it is discharged through the went system into the suppression pool. The renponse of the drywell is therefore ef fected by the parameters associated with both the blowdown and the vent system as well as by the par seters directly associated with the drywell.

The drywell configuration anJ initial conditions remained unthan ed during the runs of Test Series 5701, 5702, and 5703. The drywe!! heater was,F 200 (93 C) before each run to minicize condensation ef fects duringused go pre-heat the ves the transient. This is necessary since it was not possible to design the facility with a scaled volune and attl1 maintain the proper surf ace area-to-volume ratio (lanoring finors and interint walls, the surf ace-to-volume ratto for MAKK !!! is about 0.16 wh!!e the ratio for the PSTF is about 0.44). The dryvell vessel was vented untti immediately prior to a blowdown run to assure that atrcopherir pressure ent<t. .! In tue drywell and vent duct. The 10-inch (J54 an) blowdewn !!ns extension discharged through a pipe tee located in the drywell dome. This was done in an atterpt to purge the air fece the drywell early in the transientI however, tests indicated that the inconirt flow creates adequate turbulence to provide essentially a homogeneous air / steam misture at the vent duct inlet by the time flow through the vent system begins.

1. Pressure kesponse A typical prespute response for the drywell during the initial pressure suppression transient is shown in Figure 17. The pressure begins 9

9 9 9 4

_ a. A ..aa4 w _a_a a mm, __ __ ____w _ _ _ __ . _ _. -.a_.aa _-u. Aa , m. - ----- m_., amA_s m&.a_,_A_ _ _ __- . _ -

-4A+.-e O

1 f

I Tigure 16.

1 1 l i

l l

I V

I i i t

i b

k i

t L J

• gmummed t

4 0

I

l r

I#

m Figure 17.

4 I

I e

e i.

D I

I l

l l s

( pr4,r<d2 r1/ ,

1 4

\

6 I

e s

9 l

MDbl3177 Ct. ASS lli Page 34 of 681 C(Wi MY PRIVATI.

to increase as flew through the blesdowT. line is initsted and p. inn **

t!'rcugh a re15 af t.rr tl e first supa'esal. n r*1 scet 1% t icire.1 of water.

• rc Ire of vi ct clearinr. tire of seal. Fre* orv. t i a nrext ite 'ite s1 air habSle forcaithrench in the r mpressicn pool are shoen for refercucc.

U.c Mal drywe;l r' ensures aul masi n vent luct r ie vs (an<uminn IDP .elr) ,

art .sl.osn in Table 1.

The ef fect of ver.t >ubrer;tence on the .!rvvell tres=ure responac is indt ated in Fl.ture IR. The figurs also shows the s!fert of the water slug in the blosdawn line which was discussed in the ssct f en on steam  !

generator respeso. 11ie runs representsd bv the flaare were for fiv'c ,

difforent vent s e crgences. All cf the rmit used the sa e blowdosu .

line flow restrictor and two of the runs were p:aJe with a " leaky" steam pe:.orater riser tube. In general, increasing vent cubnerr,cnee increases the duration hf the initial drvvell presourc rise resulting in s. Sir,her peak valuu. The rate of the initial predsure rise wa9

. significantly inervased b;. the flan'iing of the vat.r slus in those runs

i. . lag a " leal.y" riser an that the neat pressure la almut it)7. h!Fher -

(

! and occurs about f1.1 second r.coner than wa ald be ersected. Note that the

pressure responsen.of all five ruca forn a family o' trices paranetarlacd ca vent auSecrgence af ter the pea *

pres *ure is rea-%!. l'igure Ic indicates the dep .dence of ped drywell presrure on both b) vJosn and vent pystem

.arametsts. In all caven, th. pcak Jrywell pressure increases wit . vent s b crgence and with blowdovn flow restrictor dlaaetur. T1.c stral ht b linen on the fi/ure ate hasa t on ti e stat a point e f rt . "ni n-!c*.l.y" l Icwdown r.ms an l. in ill et, en but .ve. Lt c "ledy" d1ta N!ats f al* abov the lines t .e two exe "f lons are the first tws ri:14 of T.*t S. rice 1201 whici.

were inf t tatcJ by valve action rather thsn by runte r. df e e,s. # .

F!gures 2'l and 21 show the len ter i alrwell -ressure rest on.e for tysical ene-vent a,J two-vent risa.. reew ettvelv. *-tse flyaren are t inluded to illustcate a characteri tic dif tvrence 6ctween singic- anil Nltt-vet.t runA f or that portion of tc tranalent (M!n. Inc the puak try. ell pressure. The pressure dur!9c t%* s angle-ve at ru,* deet aanes f ren l

the peak and 10velg out at a value verv near the p*< esure corrte.ponding to steady state ficw thr.euph tie v nt, i.e., alcof ric pressut.: plus vert sub ergence he plus the bael-rosure i!ue to -res<urlastien of the

,...w.I' ilt 4g i 411e l

I l,

D

--- , - . - - . . . _ _ . __.,.__,.__.--y ._ __ _ . - _ . ._

.,,y _ , , _ _ , . _ _ _ _ _ , , , , _ - _ . _ _ ___ r__

_ .. - - -__.__.- . -_ -_- - _ _- .. _ - ___=__ -

r 4

i j

• 4 i

1 i

1 i

I 4

I J

i, r

4 i

i I Figure 18.

u I

4 t

i t

1 I

l 4

l i

i 1

O I

l .

i 1

l

?

I i

j r I

• k I

1.

l l

"3 i r

i i

- .sg a

I ## I

1. I

$0 [

• • l i

/ .

4 L /

v 4

i .

1 I

I 1

1

...n___,,,_. _ _ v,_, .,

.. . ( - .

1 1

I l

" t

) 48 Figure 19

.1 1

i i

i I

l i

)

f 4

I I

i

(

I k

5

. T i M

/ t-t

/

1

• L I

l l

I

l.

e. .

1 I

l W  %

i i

l Figure 20 ,

/

I I i e

e

- i i  : .

j i e

b

.4 I ain I *I b

"N l

1 ) i

• l r

I i

'. l I

] i 1 1 4

l i N ;

I i

i l b l i

1 i <

1 i

i i i i

1 i

.l i

<e t

J 6

(

Figure 21, j

l \

l 1

i i

I en I

' ed

'1 H

i I

-9 N

i

\

t e

6 4 1

Y 6

9 W '

,--_.---_.-__n . _ - . . _ - - _ , . -_ _ _ _ _ , . , _ . . _ , , _ _ . . _ _ _ . _ . . _ . _ . _ , . , _ _ . _ _ , , _ , . _ , __ . ___..,n_ . _ _ _ _ _ _ _ _

l 1 Nr.DM-11177 Cl. ASS !!!

,, Page 34 of 69 ceit v.y darsir P 7

- .w

2. Terrerature Resp.*nne The tenperaturo response of the drywell is shown in Figure 22 for a typical tso-vent run. Some initial ter.perature s trat ification entsts in the dryvell with tes.peratures sp. cad over a ran e of 7') to 90 grenheit deg/ces (44 to $4 Centrirrade degre a) e a

V' lie rurs of the drywell vessel form a s erv 1.irge t cat e. Ink which 15 pre-heatcJ ever a period of hourn priar to a run and the fliw threv;h tl.e drwell JJrlnd 1 Fun 19 not suf flFlPnt in Chance t':e "Ptal tchperature ta any signifiennt entent. Tbc te wr.atures would th referc be expected t o equilibrate with approxtratelv t' e nne st rat ificat !.'s range as exhibited initially. A tb rnal l>.un. tars laver efft:t rav contribute te t'io observed strat titcation ef fret during the perint cf equilibration.

f. Vent %veten Resec%c The ntncle basic difference betvren the P Rk lit nressure wuppres=Un es e t cei and earlier presourc surpresplen Ji.irne lg tle e.nfleurattou of t!.c pret anJ wn s v s t er. . 15e coners of nressure serresstien la vull understood and the prir,try purpmc of the PSTr la to Inustigato pool Jvn nits anj vent perforrance in ti c pK 111 design. est per f orr.inc e .

6 !!! be discussed in this section wt c. :il respon=c in t'e neit .

l l

l L

I 1

6

? rn't-13177 CLtSS III Page I.9 of 69 P

B W

l l'

/

1 I

/ ,

e p 4

j e 9 l

l

[

l l

l t

. _~ .

NEnM-13377 CLASS Ill Page 41 sf 69 rn!icA=;y ersyter j

Vent systen reminse r iv t.e lorir. illy i'ivi lci! Into t hree par ta - ver t clearing vent flow, and vent recover ice an f ch'.eging. Vcnt clearing in of inte:r t minec na flev 1% po**lble irm tte drvvell to th" pool  ;

until'tl.e vent his cleared and, therrferv. Is of critisat leportance in 1 netting t'.c pe d prt>. orc. In a siellar manner, vent flev runt

be known so that drvvell preneurisation rate af ter top vent elearing and depressurization rate at later times nav be calculated, rinally,

! the vent flow area is redi.ced and the trannicat en. led as the vents recover.

l.caJn occurring during the vent recovering pro:ces hnown as chugging also perd to 1,e knevn. .

One note of cautlen is necemparv regarding P.cfr data in the area of the vent systems recall that the scale f acint for the steam generator and dryvell is neelnally tilVI while that fer the sente anJ pool Is 1:43. I'ntti the top vent is cleared, the dryvelt and pool are: casentially decoupleJ since the vent fline is near rero. nnce the too vont clears, 1

however, the vent flow area is three times t>e scaled MARK !!! value j and cawv.the ef fect is the same an if all three vents vs.1 etrated in the MARK 111 1 This leaJs to a P. ore rantil strvvell Jerressurtrat ten in the PSTr than in MARK lit and, in fact , the data ahnva the peak drvwell pre =sure to be set hv tcp vent clearing rather than niJJte teet ricaring as predicted I for PtAtX 111. '

1. Vent Cleario.: Tran 8 cut Vent clearing dati was githercJ eturina all three te st atrica reported here. In test nertes $791, thlm elata was limited to vent clearing tiecs ,

in nost cases. Additionil level proten were aJJed in the vent annulus 1 i

during test meries 5702. the two vent tests. Finsliv, durtra the three vent tests, 574), the artulus and vente were heavily instruesntcJ with a tota,1 of 26 level proben.

The lack of data for test scrire 5711 61< Juc to failuren of the i or!Pinal level probe dentrn. As notcJ in 16e f acilit=' .fcarrirt ion rect ion, i drag forces in the vent system scre sufficient to break the level probe '

witca. In canv> who te the level pr3be istleJ. t$ vent clearing tire could be infer'c4 from the peak dynamic pressurv researr.! with a pet.*t tu%c a frv inches back it.m the vent exit. Corptrisen of vent 'clearini times reasur<d in thls r.anner with the lev.1 probe rs suits f or the two and three vent tests indicatta an error hand en tle pitot tu'e /ati of 14.A) eccend. All other data in accurate to within ca.c scan of the dita acquisition avstem, i.e., generally 0.016 second during test series 5701 and 5701 and 0.18

! second durina test serten 57n2.

j Vent clearing tires for all teste are given to Table 1. In all cases the vent clearing ti es hart found fren a level prebe located A inches s (153 mm) f rom the psol end of the vent anl asprostratelv 4 inc$es (lio en)

I e

l i

i  !

I i

i

Nr. % l3377 CLASS III pn e 42 of 69 t'Ovi'A'.1 PR I VAT F f ro-. the top of the vent . Ths ton v. nt m i'.rs re nces n. te t irr .nral us s!Je sobr trentes to the centerlin. of tle vent. Durt., te u serien S M I there was considerable uncertainte in the -innutsa wate r levol. iftner ,

suberg. nres for this tent met we are no .Inal salues. .tll etScr utri rgences are calculated init ial cor.dit icr...

It has been previously stiewn that for a niven size venturi the vent clearing times may be correlated as a linear function of tof. vent sub-eergence. (Ref. 2, 3, and 4.). Fevcvt r, when a correlation of thlu type was atter.pted for the PSTF data, it was unsuccessful. This prublen was overco,e with tFe realitat clearing is a strent functien of

. dryvell pressure history e{on l that cente9eIO.and, as noted in the drysell section, the drywell pressure history was markedly dif ferent for the so-e venturi eine depending on whether or ru.t the seca generator had a 1 caking riser seal. Hence, twa correlations are retuircJ - eno for blavdsvn runs. with a

leaking seals anJ one for those without. The correlatice.s for the ene and two vent tests are given in Fir.ures 21. 24,* and 25. There is no correlation for t$e three vent tvate since enly three blowdrwns were run and enough

. data has not yet been generated to allow correlation.

Note fron Tinutes 23, 24, and 25 that the same general trends occur in the larr.e scale P5TT data as was seen durine the sr.all scale tests.

For larger breaks vent cleiring is innter but tle tf-e receirc1 increases with suhrersence. Conrarinun ci e tc.arev 21 an t Ja als.. s!..u vent clearing to be f aster in a two vent case tbsn in a single vent case, u vauld be exptets J.

Durina test series $7n3. the it.rce vent tents, sufficient Itvel prcSe instrumentatien vsis installe.1 no th is a vent annulus esa r I.,el histerv ro 1d be rencratcJ. Fleure 26 sbc. the result for Te t $ W run 2. a 2-l/2 inch M3.5 m) blevdown with a to; vent su' .reenre cf 11 feet (1.15 r).

The annulus water Icvel shows the s es hehsvict 39 seen in earlier horistntal vent pressure surpecessen te=tst an initial accelerat!< cf tav vater morface followed 1.y a tenjency en enn tant velucity lat. $

In t.e traniier.*.

this can be esplained bv frict13nal le==c= in the vent synts,a'proac?.ing the r.agnituJe of the fsreing prcosure as the water velrett) increasen.

Figure 24 almt. sho.,e vs nt clearin,t t! en ani c.-? 1ris en af the flearra with the values piven in Table I e c.ie so . dinrepane.. In the ene anJ two vent test scrics, the only ve-t c lear t e..t level tirete was lecated at the enJ cf the vent and .sprrosta.ita le & inches fl'12 *rl f r** the top of the v.nt. Inc pres 6 vais insta!!cJ in tl.at posit ion sin.e it var thourt.t that bouyancy in the gas flow veutd enund the l'ubble to te ejectcJ f ra-!

the vent near the top. l'uring It e three vent tents, a '!!!!c9s) level PreSes

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

Nr.nM*11377 CIASS ll!

Page 4: of 69 row?A?.i rnivAir

.r.

f r.,r the top of the vent . Ths t e . vent hin'.rs rc. nces svqr ! annulu ,

side sabr.ergentes to the centerline of the vent . I)ur t u teu series 5M1 there was considerable uncerta!nte in the innut a wate r level. I!cnee, j su1** err.ences for this tent scer:en are no inal salues. All ether s ub a rgences are calculated initial condit ier.s.

It has been previously shown that for a niven size venturi the vent clearing times may be correlated as a linear function of tof. vent sub-siergence. (Ref. 2, 3, and 4.). Pevever, when a correlation of this type was atter.pted for the PSTr data, it was unsuccessful. This problen was overro,e with tre realtration that vent clearing is a strcnr function of

. dryvell pressure history l.4e9el0 anJ. as noteil in the drysell section. I the drywell pressure history was markedly different for the sa.e venturl else dependinat on whether or n.it the stea* generator had a ! caking riser f seal. IIence, two correlations are rueuireJ = ene for blavd.svn runs. With l' leaking seals anJ one for those without. The correlattens for the one and two vent tests are given in Ftr.ures 21, 24.* and 25. There is no enrrelation for the three went testa since en1v three blevdowns were run and enough data has not yet been generatcrt to allow correlation.

Note f ron Timures 21, 24 and 25 that the saze general trends occur in the large scale PSTT data as was seen durine the sr.all scale sc=ts. l For larger breaba. vent cleirina le fanter but the ti c receirc1 increasv4 i with pub crgence. Corf arikun of e 6 cures 21 an t .ta also of..v. vent clearing to te f aster in a two vent case thin in a sint.le vent case, s4 vauld be exiwettJ.

, l l During test series $7n1. the it. rec vent tests, sufficient Itvel prebs

! Inst rumentat ten via inst.alle.1 so th is a vent ann. stun <ia r lovl bl*terv co.14 be generatcJ. Fleuro 28. st es= the renult for Test SN run 2, a 2-l/2 l

inch M).5 rn) blovi'ovn with a top vent ss.* .reenee of !! feet (1.15 r).

l The . annulus w1ter Icvel shove the seat behivier is sse in earlier herittntal vent pressure surpeemsten trotst an initial acceleratite. cf the water surface followed 1v n tenjeney to ron tant velusitv Ints in t' c t ransler.* .

II.is can be esplained bv f rict l3nal le.ac. In the vent svets aproaching the rat.nttuJe of the f6rcing pressure as the water velrctty increa en.

Figure 21. also sliowie vant c!,aring t! es an1 ce tirisen if the fic.res I I with the values plven in Ta!le 1 **.e. ve . dihrepano. In the "nia anJ l two vent test scrics, the only ve-t clearit..; level tirete ses !vr.ited at the eni cf tbo vent and appront .its le a anches fli2

• .*1 f r** the top of  !

the v.nt. The preb. van instal!cJ in tl.at posit ion sin.e it var thouri.t I that bouyancy in the ga. flies veu:J cmne the bubble to te ejectcJ f ra-the vent near the top. !)urinr the thrva vent trots, a ?ittiesal Ivvel rt ches 1

i r

l l

l 1

I 1

i j

i l i l

4

l.

1 i

i

) 1 6

.1ED'f- 13 37 7 I CIASS III i rage 43 of 69 i

I i

i '

l l i

i i

i l

N I i

h t

i

/

/ ,

t

/

.n i

l

_a_ -____..__a. m ._m__.__ 4 _ ma-A_m a 6-.. w-4 _h_m _. a.- _--.w- .--._h., _m.-_, u - - . - , - _ a-- am- ,a -n__. _h-am b

NED!!-13 377 CLASS 111 Fage 44 of 69 i

a 4

I e

, t I

i e

i l

i

,I 1 l 1

1

)

I, l.

4 i

i I ,I 6

/  ;

/ i t

i i I ,

I I

d 1

I

k. /,

-/ - - .

.- -== '

I .e# j an J. --*** t i ,.

o ,.

l I

d i l

I I

i t

--.we, _ .- ,

. r-- e-eve-tv-*-w. ,c--.----,-,._,. - _ __ _ - , _ ,,,_ 3 ,,-.y _

! t i

I s.

4-(.  %

a

.l

. E

~

NEDH-13377

[ ^ CLAS3 III I

Page 45 of 69 r 4

L N

4

• f ik 0 '\

• 6

\  %

4 T

I

.s 4

i 4

4 i

.I i

I

\

s

\,

\

, n 4

l

, 4

/

1 6

ie

/

l ,

~

g I e 1 _

e -- --

I.

0

+

= d

• N l *,

- - - - , - - - - , - - , - . . - - ,_. .._.s.-_.._ ,,_s_ . _ . , . , _ _ _ . , - - _ _ _-- _ _ _ . _ _ - _ ___ _

i 1

o W I

\

• NEDM-13377

" CIASS III '

Page 46 of 69 l i

l l

i

/

l l

% ./

c l

9

NEIN-133U Cl. ASS III Panc 47 of 69 CfWPA'N l'RIVATF were lustalled at t h. vent exit *, one data given in Table I is, con-i-tent in th.st all specificJ vent clearin et imes were 7.c.isure.1 at the top of the ver.t pipe. The dat s rIntted in Firure 26 shses tht first ind icat ion of vent c1 caring w!.ich occurred at the center of the vent pipe.

Figure 27 gives the level prohc trip tines for all of tlic level

, probes during test $70) run 2. the s.ine data that is plottc6 in Figure 26.

The location of the trip times on the sketch is appro:tinately the same as the location of the level probes. The data f rom all three runs in the $703 Test Series display the sane trends as sho.m in Figure 27.

A

2. Vent Flow coeffielent

.To accurately rrcJict steam flow into the si:ps rens lon pool for the MARK III contain ent d.. sign, it is necessary to knew the flow coef ficient for the vent system. rhe flow cecificient. C . Is defined as the rat io of the actual r. ass !!c. rate to the ideal ise5 tropic f!cv rate. I t w.v.

not possible to exper! entally measure a flov coef fielent for the individual components in the vent system (i.e.. annulus entrance, annulum, annulas to vents, etc.) because the length tn dianeter ratios were too srall to allow fully develeped flev within er downstrean of the cc ponents.

• Vent exit level probes are located 6 inches (151 c.~.) eack f ro., the actual end of the vent. Due te the interface velocit) at vent clearing being very hiti, the ef fect on Nasured vent clear!ng 'imes can be shown to be small.

- - - ~ . -.- .-e- -, ,, ,

s

.I i

I

-W lied:!-13377 CLASS III COMPANY PRIVATE w

i i 1

1 0

l

/

[

/

,/ .

p .

w l r w r. 27.

• a w-w'-- -

+ve7v-e-rqu-e sm --yv- -

e. .---e , .

4 .

1 1

'2DM-13377 ,

CLASS III l

.. Page 49 of 69 j j

COP ANY PRIVATE i

Thus, accurate neasurements of velocity ar.d' static pressure could not be obtained within the system. However, an attenpt was made to determine an overall flew coef ficient for the total vent systen. To determine an overall system flow coef ficient, the pressure difference between the drywell and the suppression pool, at the elevation of the vent exit, was considered. The vent system flow rate was obtained from an Annubar. flow meter Jocated in the entrance of the vent annulus. The ideal flow rate i was calculated assuming 1sentropic flow of a perfect gas.

• Cencrally a flow coefficient'is only a function of the system geometry and thus would be a constant for the vent system throughout the blowdown i p*,ransient. 'tevever, for the single vent test series (5701) the flow coefficient was not conste.nt but varied from small values to near unity as the vent fully o,nened. ,

This variation is a result of a geometry change in the vont syste as the dryw=11 air starts to flow through the vent, i.e., the actual air flow agga of the vent is' changing as the water drains from the ventq, /

_a1 Tat the flow coefficient change is a direct resu n j m .... . ... nua cm.mge i s furtner substantiated by comparison of the dr>vell to suppression pool pressure difference and the vent flow rate. Durin. this time span, the pressure difference (the driving force for vs.it flow) is

-decreasing yet the flow rate is increasing rapidly. This can be explained if flow area is increasing. The system pressure drop is decreasing (due to increasing fl u area) faster than the driving iressure difference, therefore the flow increases.

Af ter the vent is coepletely opened (i.e., the vent flow area equals

. the vent cross-sectional area) the flew coef ficient becomes constant and approaches a value of 1. It is not possible to obtain an exact value of r vaf ter the vent is coepictely open because the drywell to suppression cool pres n re difference becomes very small (on the order of 0.5 psi) and the accuracy of the measuierent deteriorates. This deterioration is due l

to three factors - the accuracy of the pressure transducers, variatiens The drywell inatmosphericpressure,andsuppressionpoolwaveactiog).

transducer had a full scale range of 100 psi (689.5 kN/m plus 0.1 psi (0.689 kN/m2) ninus 0.54 psi (3.72 kN/m2), while the suppression pool transducer had a range of 50 psi (344.75 kN/n2) plus 6.0 minus 0.09 psi I* (0.55 kN/m2). Vhen these two me1surenents are subtracted to gut pressure dif ferential across the vent system the error could be as large as 0.46 psi (3.17 kN/m2). The pool transducer was an absolute pressure transducer while the drywell pressure was measured with a differential pressure transducer referenced to the atmosphere. Henee, variations in atr'ospheric pressure introd.sced uncertainties in a pressure difference obtained i

f rom these two r.easurements. Later tests indicated atmospheric pressure l

I n

. NEDH-13377 $

CLASS III l Page 50 of 69

• COMPANY PRIVATE i

variations f rom 14.52 psi to 14.91 psi (100.12 kN/m to 102.80 kN/m ).

In addition, no cntum ef fects in the pool or even surf::ce wave amplitudes of I foot (0.9 r) would appear to be large enough to rmk the vent system pressure drop. Therefore at is not possible to calculate an exact value for the flew coefficient of the vent system because the dryvc11 to suppression pool pressure difference tw too small. llowever, the main reason for wat g'aC v value is to be able to predict the vent systen m nreuure drop.a

• Comparisons of the flow <oef ficient for the single' and double vent J

test series also indicated some interesting phenonena. During the opening transient of the first vent, in the two vent configuration, the flow coef ficient was always higher than the corresponding single vent test.

This indicates different vent opening characteristics in that the actual

, flow area was increasing faster for the two vent conf f puratioM

3. Vent closing and vent Chueging The three rows of horizontal-discharge vents in the MARK III design dif fer significantly in operating characteristics f ron the vertical discht.rge vents of the earlier designs. For the vertical vents, since the discharge end of 411 the vents is located in the same horizontal plane, the total vent area is, open for flow as long as the dryvell pressure exceeds the pressure necessary to overcone the vent subcergence head and wetwell air space back pressure. With horizontal vents, a row of vents closes when the hydrostatic pressure in the vent duct falls below the value necessary te ove rcome subnergence head and back pressure. rhis type o coeration -

y s been previously reported for snall-scale tests (Ref. Q 4

The sane types of behavior were noted for the current series of large scale tests althcugh the difference in scaling between the vent system and dryvell made analysis less c1carly defined. In general, vent closing occurred much earlier in a run and vert chugging had a much note enounced effect on the overall system. F

- -)

i.

j NEDM-13377 CLASS III Page 51 of 69 COMPATY PRIVATE A

2FI'gure 28 presents the con uet vity sensor data for a two-vent run, tiid data provlJes the inforr.ation concerning vent clearing tincs, vent closing t iees, and water icvel ir.

the annulus during chugging. In sore runs, the bottom vent appeared to chug once after it closed. The effect was very snall and is thought to be a momentum effect caused by the vent pressure swinging below the hydrostatic head as the vent closed and then returning to near the hydrostatic head. The three-vent runs were very similar to the two-vent rurs with regard to the tcp two vents and chuccinc. ~~~ ~

um es Figure 29 presents the measured vent duct flow rate for e two-vent run. Figures 21, 22, 28, and 29 are all data from the same typical -

two-vent run and can be used to illustrate the following comncots regarding the relationships And inter.-dependeneles between the drywell pressure and tenperature, the vent flew, and chugging. The cycle that an individual chug follows is that the vent reelears when the drywell pressuto is at a maximum and the vent flow is at a rinimum. The flow the

• ricas sharply and the pressi.re decreases until the vent .ecovers. At this point, tl.e flow begins to decrease and the drywell pre sure centJnues to decrease apparently due to condensation collapsing the secas r. ass in t!.e vent duct. During this time, the drywell pressure r.sv actually drop below atmospheric before it begins to increase again. The cycle is t'.en repeated as the increasing pressure forces the water f rom the vent duct and recisars the vent.

• A very interesting observation was made concerntre the overall pattern of the chugging during the first 20 recends,_T' g TEis same pattern is apparent for both two and three" vent runs, rur ati roree flow restrictor stres, and for all vent sub ergences.

Since this behavior was not noted in the small-seale te.ts, the lar ge condensation area presented by the large scale anr.ulus region and ves.ts is presumed to be responsible.

e e

• m

l l

M i

l e~

, p I ih P4

• n

'4 a m W

I n-Tt i 4-r f*

0,

\, v to

• o; I  %

/ e 4

C.

Figure 28

/ *

-.---.-,,,_--n- .---,--,-,--,,.a , .., - . . . ,-. - --.-- ___,, .. __ ,__ .--. , . , . - _. ...-- _ , _ _ _ ,. , , -. _ . . - . - . , _- - ., . , . . _ , , . - - - , , . -

\

l I

e

'l w}

t W

i l

l Figure 29.

• 1 e

I i

l r

0

3 k

i i

1 l<

(>

h i

l

(

\ '

't se Ge l

y r

/

g C

/ e, l

e tC l

l l

I l -

I i.

e l

IsED:t-15377 CIE S 111 Page $4 of 69 COMPANY PklVATE 13 . Poal Nespon,sy One of the stated purposes of the single- and two-vent testm in Tcut serie:. 5701 and $10: was to provide a map of pout well to be used for verification of analytical models. This informtion extends tte results of the swall-scale pool swell tests reported in Ref. 3. After the test plan had been formulated and instrumentation requirearnts bad -

been deterrined, the scope of interest in the rool swc11 phenomenon was expanded to include the ef fects of

• the impact loads which are anociated with the rising cans of Inquid or two-phase misture. 1his interest developed because of the location of structures in the HARK lit contnin-ment design. A pool swell action plan was for:malated to deal specifically with these concerns through isproved instru=en?.stion in conjunction with a series of full-scale air tests and a series of one-third scale tents.

These tests will be reported later; the present report involves only that data obtained during Test Series 5701 and 5702 although interpretatinn of the data is influenced by the more comprehensive data obtained during the full-scale air tests of Test herie. 5705 ar.d 5706. Test scries 5703 was not instrumented for pool swell or lepact loads.

1. Pool Swell Fool swell refers to the respunse of the ;;,11 ms4 during the initial portions of the pressure suppression transient. The pool sur f ace exhibits a gradual ac:eleration until the first vent cleaLe as theyter in t'r e mauent and annulus region is forced into the poq,LF Althougn the cuw.e Yers the pool at a pressure near th.it of . drywell at the ti".e of vent clearina, the bubble rapidly expands until it reaches a pressure closer to the achient hydrostatic head. The leaJing f ront of the bubbla expands upward until it approaches the rising pool surf ace at the psint of bubble k reakthrough. Follo int, bubble breakthrough, the tws9h ..e spra -

follows a ballistic trajectory path.

The positions of the pool surfacc and buhle front were tra:ked during the runs by an array of conductivity seesars which were used to provide a signal of " wet" indicating the sersor to be covered with water or

" dry" indicating the sensor to be uncovered or the passing of a bubble.

These sensers were positioned en the pool centerline in four vertical strings at no:inr1 horiza-tal distanced of 1.0, 6.5, 11.8, and 15.8 feet (0.376,1.95, 3.60, and 4.82 m) f rom the vent exit and were spaced on either 2.5 or 5 foot (0.76 or 1.52 m) vertical centers.

i

III2)M-13377 CLASS 111 Page SS of 69 C05tPAhi 4%1VATE Figure 30 presents a typtial met of t races preJured f ric. the conductivity acnsor data. Trm en I thronh 4 reptv4cnt nenmorn in c.ch of the vertic I st rings at a n.minal elevation of 5 fest (1.5 s.) f rom the pool botturn. Initial water Icvel for this particular run was 14 feet A.4 ed and the tiee of vent clearing was 1.00 second.F g Tn general, the

--.- Yo1 surf ace was found to remain relatively flat untti the bubble neared the surface. In all cases, the bubble resa in the central region of the pool and the two vertical strings of seru. ors statsst the c=nter were the first to detect Lt.c bubble at a gives elevation as the bubble approached the surface.

Figure 31. represents a typical net of data from the conductivity sensors in a single vertical string. A putnt reproentina the tize of vent clearing and the corresponJir.g cair.ulated value cf initial swell is also included. Curves thrcush tha. data points repres cnting the poul surface and the bubble leading edge .tre used to calculate anim.sn indicated I

velocities as shown en the figure. In sene cases, the indicated bi.bble velocitiew are thought to bu al normally high due to the near vertical e 'ge of the liubble tpnr. ti e men. orsi rather ti.an the ocarly horizontal _

..drant of the bubbir Table 1 suresarises the pool swell and bedible breaktifrour.h data for Test Series .%701 and $702. As expected, ther masir.s= poc! surf ace velocities increase as the size of the blew 4own nosale and therefore the charging rate increase and the velocit te* decrease as the initial vent subrergente increasen. The,e trendu are inJIcat=J in Figure J2. In all but two run*, the tsa Aic'ut Aurfd(* Velocity onurreJ before the second vent cleared. h 1:pict t.irtets discussed in the f. ext section af f ected the results in sooe runs.

(

/

_. -___..__e - a* . . m m_u_ , M._._____~_.___.. _ _ _ . _-_m o_4 _*_. ___am---_.__,, _m , . _ _ __ ___,.__u_ _ _ _ __ _ - - -w-_ m  % e.-

1 4

I I

h

, . .' ,**g

%h W '

v s

• taNCE 1

I g.M t ?,

O' E e.=

9 W

.-es l>

8 e e=-

1. 'p

==

L M

.e.

• 4 l

J 8 i I

4 1

4

/

-g d 1 NEDN.1))77 N Ct. Ass III f

lr P848 37 of 69 1

1 2

4 i

I e

l l

l \.

l'

/

i l

l

,/ ,. .

l

• i l

l

/

Mh13377

% CLASS Ii1 Fase SS of 69 1

1 l

d

! f e

i i

t l

/

i s s 8

. Ak

I

! t ,

' NEDN-13317 ca. Ass II

/ *E'

• N

1. CO.M*4Y I"tlVATL t y considering the two processes which define the tclationahtp l'etween the bubb!c pressure and the airflev rate into the bubble. Fir st , the air flow rate into the but ble increves as the burble pressure decrcases.

The magnitude can b.i defined fron ec pressibic flow considerations anJ the trend is shan Are the sketch as curve 1.

4 ,

BUBBl.E gM L PRESSURE, Fg PRassvR s

%s N

' ~

(2)

AIR FLott

+

Second, the bubble pressure defines the motion of the water in the ' pool.

Considering the thermodynar.!cs of the bubble, the air flow rate required to maintain the bubble per.6sure as the pool eoves increases with bubble pressure as shown in the sketch as curve 2.

t>

/

/2)

/

311BBLE

/

PRESSURE #

IIYDROSTATIC /

PRESSUREg 2

AIR FLOW 6

9 6

4 9 9 4

4

1 i

i

4:f -

l $r.D';[1d7[

CLASS !!!

l Page 60 of 69 ro.' W .Y Pk!VAfr.

t l

j ro.tbining ther tva relation bips dr fis, the bubble tre rsure an.1 the air it.sv rate .it a point in t!-e.

httstLg s,(2)

PRESSL'Its SotJIT1oM '

~  % , %% . 3

,',~% - V, ~ s

.f

%s ~ ' " - (1)

. ,-' so ton .,,, g g3 i

'f  :

e s e m eu l

l  %

I l i j . 2. Irpset Lo,a,J,4 Impact 1: ads are ter. posed on nur' aces in the patie of the rising pool i s6 ell.

The need for test data on tep.*.rt loads during raol swell develeped af ter test planning anj instru entat ter. require-ente for the large-scale der.onstration twsts luj been co plet*J. Data of a prellrinary nature was obtalacJ during test Series $701 anj SM2 by usin.1 sp to five pressure transducers of varitus types reunted un plates shic? were supparted at different distances alave the initial ecol surf ace.

Impact lo. ids are characterleed 1y relatively larre reaction forces acting for a very arall interval of tire. A el plistir approach using the relational.ip that ir.rulse is eq.elvalsat t.* re ente, et.ange can be uscJ to indicate that on impact pressure is a functier. of the density, thicknehe, and velocity of the irpacting fluid and of the time interval over whiO l the pressure acts. The dependence en tir.e interval requires that careful e

4

l. ,

f -

i

[ - N .-.

[ . ..-.

~ f HEDM 13177 CLASS Ii1 Page 62 of 69 i

]

i i

f l

1 i

S e

J t

i l

l I

/ ,

O

• * * * . I N

EEE IIE m

a

Nt!N-13377 CIA 55 III foge 63 of 69 4

• e a

e a .

1

/

l'

't k / ,

,a f

L N

i

\

l i

l j NEDM-11377 rLAS5 III page 44 of 69 C0"PVN PRIVATE o

. coatlJe rat ten be p,1ven to the rate of data ec11ertian and to the r.spanne ,

of the senstra used. The other variables are related to charging rate. l vent subresp n:c, and distance to it.c surafcc be!ng impicted.

11.c lepact pressure data is eunnarleed in table 3. The lepset  ;

- target vag an la inch (457 n) square plate for *est Series $701 Rurs 13 l through 21 and for Test Se les 57n2 Runs I thtougt 11. The plate was  !

located on the longitudinal centerline of the pool and the pressure >

transdorers were .pproximate', 9 feet (2.7 m) f ron the sent wall. For test Sert.. $702 1:uns la through 17. two impact targets were used consisting  !

cf 12 inch (305 m) oguare plates on the longitudinal centeiline of the pool. The upper plate was located a. a nontnal distance of 4 feet (2.4 m)

. above the pool surface and i feet (2.4 m) from the vent wallt the lower plate wie nominally 4 feet (1.* e) above the surface and 10 feet (3.0 m) f roi the vent well.

The pressure transducers Itst.*d in Table 1 were all strain gauge {

. type =4rept the Eistlers which were quarts crystal type. The frequency  !

.. respense of the transducers was on the order of Soo Pa for the cavity- l i

l j

type cenisto to practee the flush-*o9nt provide Senacr.

rise tiresSensetec.

in the millisecog(g range.

and Kullt ,with 50tires rise kHaoffor  ;

tens el' Microsecondt, and $0) kHz for the flush-nounted Xistlers with  !

rise ti es in the nieremeconJ ranao .

The let'act pressure Jata showed considereble .catter and could not

, g,ggafeq.atelycorrelated. :ertain observations can bc eade, howegegummme, , ,

A

.m.s I"f55 trevidad valuable infornation for design of th!'fu. ... ..,ll-scale air tests  !

of Test series Sio$ and $706.

E. C;$tain ent Respan.e A major advantare of the pressure suppression type reactor centainment t to that it'etnintres the containrest design pressure. Co plete condensation cf it.e stest vented to the real is a baste assurptien for pressure scppression ,

l containtents such as the ti\R2. 111 design. The F$TF pool building is i a scaleJ pressare building anJ can be used for determining contain-ent  ;

i i

response Juring the pressure suppression transient.

I  !

t 1

l

D i -

I NLDM-13377 Cl. ASS III tage 65 of 69 fo'it'ANY PRIVAT,5 The preneure and treperature respon.e of the pool building was r.rcrded J4rin;. the one Snd two vent rune of Test beries 5;nt ani 5702 to s'eterrine the eatent of pream condensat ion in the soppression pont.

Witl' covplete rendensat ten of all the stcan which was dischargeJ into the pool, the pool building pressure would increase because of the aJJitional air mass (f roe the drywell) which was added to the pool building and because of heating of the air. If there was a trussure instease greater than could be acteunted f or by these twe phenomena - it was considered to be a result of incur.plete steam condensatten (i.e., staan bypassing the pool).

• To calculate the reak building pressure it was assumed that honogeneous conditions entsted in the boliding air space throughout the transient.

Table 4 sunmartres the initial containment test conditions for the 5701 is not and $702 Test Series. Test serics 570), the three tent test series, included because the pool building was cpen to the atmosphere during these tests. The initial mass of air in the drywell war calculated based on a naus weighted average of the nine measured temperatures and an assunto sero humidity. The assu.ption of aero humidity was reascn.ble because inceJiately before the blewJovn the dryvell was rurged with ambient a;r.

When the ambient air. no .inallv 700r (21*C) and 503 humiotty, is heated to approximately 225"r (1070C1 the hunidity drops to aroand 1%.

Since the drywell was vented prior to the run, the initial pressure was equal to atmospheric pressure. For talculational purposes, it t'as further assened that all the air in the dryvell and vent systen was transferred to the pool building air space.

There were two therececuples mounted in the pool bullains air space at elevations of approxicat=17 25 ft (7.62 r1 and 35 f t (li .67 m) f rom the pool bot tom to evasure the air temperature. These thernecoutles showed very good agreement throudh nut the transient, giving no indication of thermal stratification. The pool building pressure rise was rocorded by the building pressure transdacer as wc!! no by the various pool ewell impact pressure sensors, which provided a gooit cross check on the accuracy ef 'his erasurerent. The pool building vse also vented prior to the run, hence. the init tel pressure was equal to .itrcopheric pressure. The building hus.idity vas measured by a humidity probe however, the transient response of this instrument 6as doubtful and the final building huildity was assured to be 100*.. This assumption was supported by rotion pictutes taken during the tests which stoved that the air space was usually thcroughly strayed during pool swell / bubble breakthrough. In addition, fogg193 (inficating 1002 hunidity) was of ten observed.

The drywell ard pool building air volumes (in cubic feet) were determined from the following equations.

V = 2615 - (ll.09)L Drywell V = 19,378 - (450)L Pool Building where L equals the pool sater level in feet.

i l

1

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

I I

(

l e

l l

l l

f i

I

[

/ COMPN 4 PstvAYg t A

i f

jW i t 1

i r l

l 4

I s

I l

4 l

%..= 3 I

I ei i IM I '.}.

N 1 enum m.smeis e

{

k

• t #*

L ==

t hsvh

.m.

@ Sue j M i

8 P

e d

/

l l'

• 4 A i

J ww w e w v * ^ 1-epe-e w meeww*wei- m- et-

- - - . - -- A - -4 -. _ _ _ 1 a -A- luaAJ A um -- . - - -- .A-----a-----------+-----------A- - ------- --- ---- - -A- - > - - - - = - - - - -

\

~

I i

l .

l i

,TdIN kh 5 m

estas I

i*

i f

1 43 I

l 1 4

1

/.

P L

Tv

/

4 g

EP MN P4 OM M

o 4

/

I C

i i '

l

- . - - . - - . - - - - _ _ - - - . ---- , . . . _ _ _ ., - __,,--,__---__..._________-_._,_-_._.-.-____..,,,,,,--a.._ ,- _ . ----- -,_, ..

,t 1

I uttet-11377

.CU SS III f

Pete 60 of 49 Co*fPAW PalvAfg The hosoger.ccus model wee used to calculate the manime pressure rise in the pool building and Figure 34 le a roegartsen of t

. teasured pressure time for tiie one and two vent to.ite. g* ralculated --l versus l

.I I

l 6

I e

6 i

l l

i 4

l .

l i

l l

e e

L ..

4 l

i i

I 3C H-13377 r i

l CIAss 133 Fase 69 of 49 l rmravy raivur  !

l ' \ ) .mmmmmmm==m p r

b W i

t l l

l l

i i

i ,

l I

l I

J

. l, r ,y \ '-

4

. I ',  !

1

/

i i

i I' Figure 33. Contairwent Pressure Rise Comparison.

t l

. - . - - - - - , - , _ _ _ , _ , , ,-_y.._._y,-__.. --,,,,r-, -

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

i h

i i letlPf-l 3377 i

Class 111 ry tt'A W PrivA1Jr, i

kr6ist"eps l 1. TownsenJ. it. T., **tW' Ill a ctiflinal erv Tent fre r r,em--Progres*

I

\

Report." N Wt In844 (April 1973) (rl. ass Ill).

! 2. Myers, L. l.. " Pressure Suppresolun les Pro 6tas:**5 mall itcale i

Norisontal Vent Te sts," Ntl*l.13136 3 (October 1972) (Class ll).

j 1. Melatyre. T. h., "Preneure furpression Tsat Progran--Ssall Scale c j

Poc) Swell leets," NEDM 131)l (August 1971) (Class !!).

i

. (

l 4. McIntyre, T. N., " Pre 64ure Suppression Test Fregram- Seall Scale '

Tests," NURI.13365 (January 1914) (Class II).

5.

Danielsea. D. W., "NSTF Hessurer'.nt and Data Acquialties l i

• Techniques," Mif1 11159 (Jimo 1971) (Class 11).

6. burnette, C. W., Danielson. D. W. , 4tleson, R. A., "AWR Slowdown l

Nest Transfer Program T.tsk C-4 Report-Preltninary Systen Deslan Description of Tw Locp Test Apparatus -Revisten 1 "  :

i CEAP-11276-1 (Nover.ber 1973)

! 7. Miller, C. W., turnette, C. W., Saari, C. L., " lower Plenus

! Flashina/LevelFwellActlenDuringSlowdown,"NEDM-13192([t[ '

('tay 1971) (Class lit). ,

I -

8. Moody F. J., '?l.ninus Two l'hsee Vessel Blevdown f rom Fires "

1 ArrD-4827 (April 146%). f:

i

,r I

9. McIntyre, 7. R., "An Analvtical Investlwation of tiie Vent  !

Clearirs; Transient for Pressure Surpression Systers Havleis 1 i

1 Multiple Horitenta' '.'ents," NFDE-1129) (July 1972) (CI.tse !!!)a a

[

l 10. Bilsnin. t.'. J., "rourth Quartertv Progress 'erort i RV.X 111 II t

Confirr. story Test recuran." NE W 201&% (Class I) anJ i

3 Supplerent I (Class til) (April 1974).

U i

i j i i

^

e 4

I 4

4 I R1 l I

f

,6 9

i NEIN-131F F C4. Ass III CfMt'AYt_PalvAff, APPrf:fe!X A.

Vr.N1t h,lltfA' N fi VF4N'f, M\%4 CAf ft'lAtitM Al. Ventu,,ri Flc, rel_citatq,c3 For staan hinwJowns at the Pstr, tha polnary flow meneuring element le a critical flow ventuel. Preneur ' tape are provided at the venturi entrance end throat ansi d.ata taken at flase toestions is converted f rom pressure to flow rate. The purpose of t .it section in to document this procedure.

Cemelder a evntrol volume within the ec9 verging section of the venturi.

as shown in the ehetch.

G M: l )

i I .- @

Ver t ebl_e s, Subeericto X = quality 1 = at entrance A = ares 2 e at throat y = pressure p = density U = velocity a = entropy h = enthalpy A1 8*

l

! i WEDN.13377 ct. Ass !!!  ;

COMPANY PRIVATF.

Assee the flow to be

1) ene dimentional i 2) quaal-steady 1

l 3) sesgressible ,

4) hosegeneous -

S) edialestic l

6) isen repic l

The integral feree of the eenaervation equations for anssa and j

energy ares ,

, ,] dV = = ,,,,, I' d (A-1)

I'#I d U

• c.s.I" * *8)pU (A-2) 4 = 9. .v. s Applying the e aations of conservation for mass and energy to the cor. trol volume with the given assuriptions yielda  ;

AU p222*#AU111=0 (A-D i 11 2 2 ~

U2 g

=0 (A-4) h2-hg+ 2 c

Note the shaf t work .W. has been set to aero and the nossle has been  !

aesmed to be horisontal as in the actual installation.

In addition to the conservation .tquations for mass and energy, the assertien of isentropic flow yields a term of the second law of ,

i therriedynmates ,

(

I

• (A-5) s2 'l f .

1 A-2

i ~/

- r >

Ah ( .

1 -  ;. : .

\

s s . .\ x l '\

i a e ,

+ 6 i ,

'i

\ NEDM-13377 CIASS III 3 1 CO*4PA N PRI% ATE t s The thermodynamic properties of a saturated steam / water mixture may be expressed as d = (1 - Q ih +rh8 t (A-6a) 3 \ 1 s = (1 - X) og 4'1 e , ,

(A-6b)

. i :', ;

(w

, s' .

-r s bl (1 - X) v +Iv d (A-6c)

.k,

'E - /

. i Since the pressure at the thtoct and entrance are known, the saturation properties:may .he found from si, tam croperty tables and if the quality is known the thermodynamic pro : ties of the flow will be known. In goeral, an assumption may be made concerning the entrance q>ality '

(i.e., X = 1.0 for a steas Mewdown). For the throat quality, solve eq. (A-6b) '

8 y= 2 ~ 'f2 X , (A-7) 8 '

32 ~ 'f2 Substituting eq. (A-5) into eq. (A-7) s 1 u o(2 \' 't y .

(A-8) 2 s h, g2 ~ *f2 Y '

\

N

. i and row the therr.odyna-ik poperties at the entrance and throst may be )

found irom tquations (A-6).

i' .

'i.

Chening from density to specific voltune and solving for the throm:: velocity, eq. (A-3) may'be w-ittet as

j 3.

v.

v A U2 =3d v A U

1 (A-9) y 2 x r k #

4 t

i 1

t

\ s A.3

( '* e

, ~

4 g

3 4 (

'g (', g.

s \

\_ l ...\ .- __.

0

,1 e

i s

. .1" ". 6 NEOM-13377 CLASS III COTANY FRIVATE Su!'stituting eq. (A-9) into eq. (A-4) and solving for the entrance velccity 1/2 2Jge (h3 -h) 2 l

• U = '

1 vy (A-10)

A) 2 -1 l

, 'I ^2 Once the velocity is known, the flow rate may be found from W=7UA 3g3 (A-11)

A2. Vessel Mass calculations For water blowdowns and as a secondary nessurement fer steam blowdowis, the flow rate is calculated by numerical differentiation of the verssl mans. The pressure. vessel is divided into seven nodes, with a difftrential pressure measured for each node.

F v

FRESSURE Il VESSE1 WODE l

.J... . . . . . . . . . . . . . .t.

I c1 lo hi y . . .

PRESS 1'RE TRANSDUCr.A 4

A-4

NEDM-13377 CLAS* III COMPANY PRIVATE The pressure on the high side of the transducer is the pressure at the top of 11.e node plus the static head in the cold Icg P

y=P,+[e act "cL IA-I2) i Similarly, the pressure on t' 3 low side is the pressure at the top of the node plun the static heat a tos: the node plus the static head of the low side cold les P g= P, + ({H) + p g (H - H) (A-13)

The difference in these two pressures is, of course, the differential pressure see, by the trans luree AP = Pg-P = P, + [e pd et H

-P,-[e (3,H) -

c p

d (iid' l) (A "14) l

• i Simplifying and solving for the nodal density, p , '

7, = pcL -

(A-15)

Note that A, is the average density in the pode, an'd for nodes containing a steam steam / water interface it will be a value inteilnediate t- the water and density.

The total mass may then be found from 7

M, = I (7V) 3 g A-16) i=1 A-5

l I

1 NETtf-13377 CIASS III l COMPA?N PRIVATE wi ert siand v 3 ar. the nuaai censit ies .,nd volees. The flow rate is then calculat'ed f ron, dM y._

(A-17) .

e

'**e- 9

• e I A-6 I

l t

.NEDM-13377 COMPANY PRIVATE CIASS III DISTRIBL' TION M/C Name M/C Name R. P. Barr 377 f.L.Myers 584 687 H. D. Powell 386 J. L. Denson 762 .J. W. Power 682 W. J. Bilanin ,

657 F. Reuter 665 A. P. Bray 670 D. A. Rockwell 762 T. O. Brcr.m C. H. Romer 145 C. L. Davis 6,82 767 R.'A. Roof (6) 126 D. C. Ditmore.

$84 C. C. Ross 682 R. J. Ernst Bethesda A. Rubio 716 L. S. Gifford W. D. Gilbert 685 W. E Smith 373 591 L. J. Sobon 682 R. G. Hamilton 741 682 J. M. healzer 1. F. Stuart C. *. Hilf 685 R. L. Tedese s (3) USAEC J. E. Hench 762 J. E. Torteck 583 A. M. Hubbard ,

37i h. E. Townser.d 584 D. H. Imhoff 533 C. E. Wade 665 A. J. James 767 C. D. Wilkinson 670 L. E. Koke 672 L. E. Wood 399 R. T. Lahey 583 NED Library (5) 328 L. N. Larsos 682 VNC Library (2) VNC-102 161 A. J. Levine 682 ISS Storsge (3)

T. R. McIntyre SR4

• .s

I i

a , 0.-

• ~

. =

k,.

i A .1 Lhb@,.

h% ,

E t,

- \

.s .

..? Y.,3 I m

h'ER0 PHOTOGRAPHER _w- 1~------_

_ f_-'_'-_'__'_--

I. .

I f.a%.'s-s '. ' I i 1 J- fl:

i As . ...o#

MICROFILM SECTION NAVY PUBLICATIONS AND PRINTING SERVICE OFFICE SUILDING 157 2. WASHipeGTON NAVY *tARCi WASHINGTON.D C. 20374

• 9 .