ML20081C069

Analyses of Steam Generator for Tube Ruptures in Arkansas Nuclear One Unit 2, Apr 1983

ML20081C069
Person / Time
Site:	Arkansas Nuclear, 05000000
Issue date:	04/30/1983
From:	Blakeley J, Chon Davis ENERGY, DEPT. OF, IDAHO OPERATIONS OFFICE
To:
Shared Package
ML20079E633	List: ML20079J564 ML20081C062 ML20081C069
References
CON-FIN-A-6354, FOIA-83-168 EGG-NTAP-6226, NUDOCS 8310310084
Download: ML20081C069 (66)

Text

._

EGG-NTAP-6226 April 1983 ANALYSES OF STEAM GENERATOR TUBE RUPTURES IN ARKANSAS NUCLEAR ONE UNIT 2 7 /\\

. p e, C. B. Davis J. E. Blakeley Idaho National Engineering Laboratory i

Operated by the U.S. Department of Energy t

w i

1

.L Y5

{y

~

~'

hh}E%51 f' dh A

2C f'ti+i

.{

2 *.-

h 4[+%M%%Eqs,2h,

a. a.---..-

-- -Q :t I

r.d,=

7_-

R% -m: +mv ~, u +W ^

,-4 yny

---

x

\\

~

_ % p"r'Teiq'd 80E:::MEM;---C. ~

l~~~ TF

.a

-m 4' h l

i

- iU^ :'E N

.nisiffg,;-

=,-

2-----gM Y-

, ~

• i C _M e M6dl f ',~.

c.,

4

_., a,. -

, - c_.

v.6 err _.

3 D

y*

GisEMndi$was5 Man;Ah CDD This is an informal report intended for use as a preliminary or working document 8310310084 830720 PDR FOIA i

BELL 83-168 PDR Prepared for the W U.S. IlVCLEAR REGULATORY CO.*ISSITl

[]

q" Under DOE Contract No. DE-AC07-761001570 E 6 E b idaho FIN No. A6354

EGG-NTAP-6226 ANALYSES OF STEAM GENERATOR TUBE RUPTURES IN ARKANSAS NUCLEAR ONE UNIT 2 C. B. Davis J. E. Blakeley Published April 1983 EG&G Idaho, Inc.

Idaho Falls, Idaho 83415 Prepared for the U.S. Nuclear Regulatory Commission

. Washington, D.C.

20555 Under DOE Contract No. DE-AC07-761D01570 FIN No. A6354

(~~.

ABSTRACT Analyses of hypothetical steam generator tuce ruptures (SGTRs) in Arkansas Nuclear One Unit 2, a Combustion Engineering pressurized water teactor, were performed using the RELAPS computer code. Hypothetical accidents initiated by single and dual SGTRs were analyzed. Calculations were performed to compare the effectiveness of the auxiliary pressurizer spray versus a pressurizer power-operated relief valve in depressurizing the reactor coolant system. The analyses were performed to determine if the reactor operators could control the thermal-hydraulic response of the plant during a SGTR.

i 11

C

SUMMARY

Analyses of hypothetical steam generator tube ruptures (SGTRs) in the Arkansas Nuclear One Unit 2 (ANO-2) pressurized water reactor (PWR) were performed using RELAP5. AND-2 is a Combustion Engineering (CE) PWR with a rated core thermal power of 2815 MW and is owned and operated by Arkansas Power and Light.

RELAPS is an advanced computer code designed for one-dimensional, thermal-hydraulic analysis of nuclear reactor and related experimental systems. Hypothetical accidents initiated by single and dual SGTRs were analyzed. The single SGTR was initiated by a double-ended rupture of a single steam generator tube in one steam generator. The dual SGTR was initiated by the simultaneous double-ended rupture of two tubes, one in each steam generator.

The RELAPS calculations were performed to determine if the reactor operators could control the pressure, temperature, and level of the reactor coolant system (RCS) during SGTRs using the steam generator atmospheric dump valves (ADVs), the high pressure safety injection (HPSI) system, and I

either auxiliary pressurizer spray (APS) or a pressurizer power-operated relief valve (PORV). The effectiven'ess of the APS versus the PORV in depressurizing the RCS was determined. Calculations were performed with the PORV to gain insight on how a CE plant with a PORV would respond during a SGTR even through AND-2 does not have a pressurizer PORV. The calculations also provided insight as to whether backfitting a PORV on ANO-2 would be beneficial for mitigating the system response to SGTRs.

The analyses of the SGTRs were performed at the Idaho Na'ional t

Engineering Laboratory by EG&G Idaho, Inc., as part of the Severe Accident Sequence Analysis Program. The analyses were requested by the Nuclear Regulatory Commission (NRC) in order to audit similar calculations performed by CE. The CE calculations were performed to provide generic input to the response of the CE Owner's Group to NRC questions regarding long term decay heat removal and depressurization in CE plants with and without PORVs. A summary of the results of the RELAP5 analyses is presented below.

iii

The analyses indicated that the reactor operators could successfully mitigate the effects of single and dual SGTRs in ANO-2.

The RCS level, temperature, and pressure were adequately controlled without seriously challenging the safety functions of the plant.

In the RELAPS calculations, the RCS level was controlled with HPSI. RCS temperature was primarily controlled by adjusting the area of the ADVs. RCS temperature was also significantly affected by the manner in which HPSI was controlled. Reactor pressure was controlled by using either the APS or the PORV.

The tnermal-hydraulic response of the plant was similar whether the APS or the PORV was used for depressurization.

Since the APS system was adequate to depressurize the plant, adding a PORV to ANO-2 would provide little benefit for mitigating SGTRs when feedwater is available.

(

\\

D 4

(

iv l

i

r'~

i CONTENTS ABSTRACT............................................................

11

SUMMARY

iii 1.

INTRODUCTION.....................................................

1 2.

MODEL DESCRIPTION................................................

4 2.1 Nodalization........

4 2.2 Initial Conditions.........................................

11

?

2.3 B o u n d a ry C o n d i t i o n s........................................

13 3.

RESULTS..........................................................

15 3.1 Single Steam Generator Tube Rupture........................

15 3.2 Dual Steam Generator Tube Rupture..........................

37 4.

Conclusions......................................................

53 5.

REFERENCES.......................................................

54 i

APPENDIX A--CODE UPDATES..............................................

55 FIGURES 1.

RELAP5 nodalization of ANO-2.....................................

5 2.

RCS and steam generator pressures for the single SGTR with APS..............................................................

19 3.

RCS fluid temperatures for the single SGTR with APS..............

19 4.

Average RCS fluid temperature for the single SGTR with APS.......

22 5.

Pressurizer liquid level for the single SGTR with APS.....$.....

22 6.

Hot leg mass flow rates for the single SGTR with APS.............

24 7.

Break mass flow rate for the single SGTR with APS................

24 E.

Total HPSI mass flow rate for the single SGTR with APS...........

26 9.

APS mass flow rate for the single SGTR...........................

26 10.

Hot leg subcooling for the single SGTR with APS..................

28 i

V

7 11.

Steam generator liquid levels for the single SGTR with APS.......

28 12.

RCS and steam generator pressures for the single SGTR with PORV.

32

'13 RCS fluid temperatures for the single SGTR with PORV.............

32 14.

Average RCS fluid temperature for the single SGTR with PORV......

33

(

15.

Pressurizer liquid level for the single SGTR with PORV...........

33 16.

Hot leg mass flow rates for the single-SGTR with PORV............

34 17.

Break mass flow rate for the single SGTR with PORV..............

34 18.

Total HPSI mass flow rate for the single SGTR with PORV.........

35 19.

PORV mass flow rate for the single SGTR.........................

35 20.

Hot leg subcooling for the single SGTR with PORV.................

36

21. Steam generator liquid levels for the single SGTR with PORV......

36 22.

P.CS and steam generator pressures for the dual SGTR with APS 40 23.

RCS fluid temperatures for the dual SGTR with APS................

40

(

24.

Average RCS fluid temperature for the dual SGTR with APS.........

42 25.

Pressurizer liquid level for the dual SGTR with APS..............

42 26.

Hot leg mass flow rates for the dual SGTR with APS...............

43

27. Break mass flow rates for the dual SGTR with APS.................

43

28. Total HPSI mass flow rate for the dual SGTR with APS.............

44 29.

APS mass flow rate for the dual SGTR.............................

44 30.

Hot l e,g subcooli ng for the dual SGTR wi th APS....................

45 31.

Steam generator liquid levels for the dual SGTR with APS.........

45 l

l 32.

RCS and steam generator pressures for the dual SGTR with PORV.............................................................

48 t

33.

RCS fluid temperatures for the dual.SGTR with PORV...............

48 34.

Average RCS fluid temperature for the dual SGTR with PORV........

49 35.

Pressurizer liquid level for the dual SGTR with PORV.............

49 k

vi

,.,v.

F,

(

36.

Hot leg mass flow rates for the dual SGTR with PORV.............

50 37.

Break ma ss flow rates for the dual SGTR with PORV...............

50 38.

Total HPSI mass flow rate for the dual SGTR with PORV........

51 39.

PORV mass flow rate for the dual SGTR 51 40.

Hot leg subcooling for the dual SGTR with PORV...................

52 41.

Steam generator liquid levels for the dual SGTR with PORV......

52 TABLES 1.

Initial conditions...............................................

12 2.

Sequence of events for the single SGTR with APS..................

16 3.

Sequence of events for the single SGTR with PORV................

31 4

Sequence of events for the dual SGTR with APS....................

38 5.

Sequence of events for the dual SGTR with PORV...................

47 vii i

4 i

I 1.

INTRODUCTION i

j Analyses of hypothetical steam generator tube ruptures (SGTRs) in the l

Arkansas Nuclear One Unit 2 ( ANO-2) pressurized water reactor (PWR) were l

performed using the RELAPS computer code. Hypothetical accidents initiated by either single or dual SGTRs were analyzed. The analyses were performed at the Idaho National Engineering Laboratory by EG&G Idaho, Inc., as part of the Severe Accident Sequence Analysis Program.

The analyses were requested by the Nuclear Regulatory Commission (NRC) to allow the NRC to audit similar calculations performed by Combustion Engineering (CE). The CE calculations were performed to provide generic input for the CE Owner's J

j Group to NRC questions regarding long term heat removal and depressuriza-4 tion capability in CE plants. The NRC questions addressed the reliability l

and safety of CE plants without a power-operated relief valve (PORV) on the pressurizer. The NRC also wanted to determine whether or not plant reliability and safety could be improved by backfitting PORVs on existing CE plants.

i i

{-

The purpose of the analyses was to determine if the reactor operators could control ne pressure, temperature, and liquid inventory in the reactor coolant system (RCS) during a SGTR by using the steam generator atmospheric dump valves ( ADVs), the high pressure safety injection (HPSI) j system, and either the auxiliary pressurizer spray (ApS) or the PORV. The RELAP5 calculations represented the response of the RCS during a SGTR and the effect of operator actions on the RCS response and hence provided i

insight on the operators' ability to control the plant, i

A SGTR presents a challenging scenario to the operators in order to control the plant subject to several key, and sometimes conflicting, considerations.

First, the operators want to minimize off-site radiological release, which requires determining, and then isolating, the affected steam generator. The affected steam generator is the one in which the SGTR occurred.

Second, the-operators want to maintain adequate RCS inventory so that the core remains' cool. HPSI will be required to maintain l

inventory for sufficiently large SGTRs, but the operators want to avoid

{

filling the pressurizer which could impair their ability to depressurize j

l

[

1 l

t_.

=_

the RCS using either APS or a PORV.

Third, the operators want to depressurize the RCS so that the shutdown cooling system can be emoloyed.

Fourth, the coerators want to cool down the plant in order to maintain loop subcooling so that core voiding can be prevented. Technical Specifications impose a limit on the cooldown rate.

In addition, the subcooling reauirement ruy limit the RCS depressurization rate.

Fifth, the operators want to minimize the differential pressure between the RCS and the affected j

steam generator in order to limit the loss of RCS inventory and to prevent filling the affected steam generator with liquid, which could challenge the i

steam generator safety relief valves and result in additional release of radioactivity to the environment. The results presented in this report

)

provide an indication of what the operators would see during a SGTR and how I

well they could control the plant relative to the above considerations.

t ANO-2 is a CE PWR located in Popa County, Arkansas.

It is owned and operated by Arkansas Power and Light Company. The plant is rated at 2815 MWt (912 MWe) and has 177 fuel assemblies in 16 x 16 arrays with an active length of 3.81 m (12.5 ft). There are two hot legs connecting the

{

vessel to two U-tube steam generators.

Each steam generator is connected N

f to two cold legs which, in turn, are connected to the vessel.

Each cold leg contains a reactor coolant pump.

For purposes of comparison with other CE plants, ANO-2 has a small pressurizer [34 m3 (1200 ft )], a small 3

3 upper head [14 m (500 ft )] and low head HPSI pumps [9.6 MPa (1400 psta)].

a

(

RELAP5/M001.5 is an advanced computer code designed for best i

estimate thermal-hydraulic analysis of postulated light water reactor transients.

It is a one-dimensional analysis code based on a nonhomogenous, nonequilibrium hydrodynamic model that utilized two continuity, two momentum, and one energy equations. The code has the l

~

The SGTR calculations were performed with an updated version of a.

RELAP5/M001.5 cycle 25.

RELAP5/M001.5 cycle 25 has been stored under Configuration Control Number F01138 at the Computer Science Laboratory at the Idaho National Engineering Laboratory.

The updates used in the SGTR calculations are presented in Appendix A.

These updates were incorporated in versions of RELAP5/ MOD 1.5 subsequent to cycle 25.

2

= - - _.

- = -.

i 1

capability for modeling all systems in a nuclear steam supply system necessary for simulating SGTRs. Features unique to RELAp5 include j

subcooled and two phase nonequilibrium and nonhomogenous choked flow models, horizontal stratified flow and stratified choked flow models, noncondensibles in the vapor phase, and a two phase mechanistic abrupt area change model.

In addition, RELAPS has control systems to solve simultaneous algebraic and differential equations for simulating and controlling the behavior of components such as valves and pumps during transients.I The remainder of this report describes the SGTR analyses. A description of the RELAP5 model of ANO-2 is presented in Section 2.

Results of the single and dual SGTR calculations are presented in Section 3.

Conclusions derived from the SGTR analyses are presented in Section 4.

References are presented in Section 5.

(

4 f

3

I 2.

MODEL DESCRIPTION The nodalization of the RELAP5 model of ANO-2 used in the SGTR analyses is described in Section 2.1.

The initial and boundary conditions are described in Sections 2.2 and 2.3, respectively.

l 2.1 Nodalization The RELAP5 model used for the SGTR calculations consisted of 207 volumes, 216 junctions, and 171 heat structures.

Nodalization diagrams for this model are shown in Figures la through lf.

The hycrodynamic l

components were modeled to represent the actual flow areas, lengths, i

volumes, and elevation changes that exist in the plant.

The heat structures were modeled to represent the actual storea energy and heat transfer area of the metal components of the plant. The nodalization of the coolant loops is shown in Figure la.

Each coolant loop was modeled separately so that two hot legs, two steam generators, four cold legs, and i

four reactor coolant pumps were explicitly represented. The loop containing the pressurizer will hereafter be called the "A" loop. The loop without the pressurizer will hereafter be' called the "B" loop. The pressurizer model included representations of the heaters, surge line, main and auxiliary sprays, and a PORV. ANO-2 does not have a pressurizer PORV, but one was added to the model to compare the effectiveness of 1

depressurization with the APS versus the PORV.

Connections from the HPSI system to each cold leg were represented. Connections to the charging and letdown systems were also modeled.

The nodalization of the reactor vessel is shown in Figure lb.

The reactor vessel was modeled to include the downcomer, lower plenum, core, upper plenum, upper head, core bypass flow paths, internals, and fuel j

rods. The modeled core bypass flow paths included the outlet nozzle clearances, the alignment key-ways, the support cylinder holes, and the core shroud clearances. The outlet nozzle clearance and alignment key-way paths were lumped together and connected the inlet annulus to the upper i

(1 I

m A l oop B luoj.

Steam l itu?

'I8 g,y

-.l g3g l

~

,56'

' ~E ~~~ I Z' - ~ ~ ~ "'

~ ~ 'l-.. -_, b_ T l ILO ,

,

• 110 ILS II8 g

417 AP5

_f44; Sn Y i

line Preuurlier

'5I d5 g,9 f ee*fw.s t er 4%. ---

e-g 3

g g,,

l ee ilw.s t er GenesJior 5tvam

?g:

5yy 5I5 4 II--

y 5

Generatos s<.

. /

- - ~

(4 15 8 ppsg ispsl

[I -

l}[E

) 1 '-

r_b)I---

I p

y

}g- ' -

I---j - ikst lee) l11 182 7 10 I til lil 1

181, 280 1 T _.. r -'

_1 -~_.. yto I IE-Cohl I egs 12 213 750 I"E lil* 262 292 linup d g g,, g,, ' 2:~ ;

d--'r_

' yh _. _ r r rc 6 " ~ ~ ~ I trz --- ~ 5I I al i e.. - 575 l u-- j- '-- 6,] 962]. f -,, - lips t g p,,, thargiruj Figure la. RELAPS nodalization of ANO-2.

( U ee Head g 724 rs -x !29 -N N N-a3 722 N -N % fr* N N; N N N N. ^

220 21:.

^ 252 262 I 293 243 N] D l Upper Plenum k 720 Q D 700 Q y 700 h j l I l s s S N N N N. 7C3 71E N 'N 700 N N N N N = a l -N x s N N N N N N N N. N N N N N 'N N N N N N. N. N N Q Downcomer \\ Q _ -Q k N N N N' N N N N N '\\ N N N N N, N N_ __N N N N 700 ' ' Og 716 7 14 N s N N N N N N N N N N . _ _' ope q_ _ _Q N N N N N q N Q N N N s q p_~ s N N N N N N N N N N N N N N N N N N k k N N N N N N, N N N N --N N N N i N N i l N N N N N N N. N N N N 712 N 'N N N_ N N N N ,i 1 .I N,NNNNNNNNNN\\\\NNNNNNNNNNNNNsNLNNNNNNNNN W (, 70S ie N L:wer :lenum 710 N m m# ( 'N l l Figure Ib. PELAPS nodalization of Af;0-2 -- reactor vessel. 6

To A Steam Line J A ;; i } N / ,pt! / /'t l/ sr, i i -.-- - -- - - - - -- - -.-- ~ \\ .\\ j-n 416 n\\'j ) 4 '/ + 415 Separa tor A i 1 405 '/ I Main 1 413 / '00 Feedwater i l 407 l P Emergency l:02 j -, - ' ' \\,() I y / Feedwater 413 6410 \\ 410 [ r 4 + ' 411 f Riser 41' I / 413 / / / l l*/ro // / / / f o 1 7 f I I i / s' g / l/ / /. Y- - p' h / / / 302 T / 7p A h t' h h // / / U Tubes /- Y j Downcomer / 9 S 9l 9 y / /. ) /\\ / / 7 . - ) /, - ,t-y / 5 y 3 /) ,/ / / / c s' / / 7 s. / h(/ '5 9 'd I W p /' / V / / / / / 'A/ / / / / / A / 6 / / 301 / J / l+ LJJ/'7 Tim ~~ 4m/ '412 u 7-412 j 1 Pi

• /

FA g[/' r l300 it-303 h From A Hot leg To A Cold Legs # L' i { Figure Ic. RELAP5 nodalization of-ANO-2 -- A steam generator. 7

~. Io 5 Steam Line i 467 l /} / / j, / 466 @'\\ /- ) 1 i a / 465 i Separator 3 A P k,C, 455 h) Main jj 463 k,3,/ Feedwa ter .0 j,s ,,, ; i '3' .,1 i t c.ergency [: p~i Feedwater y(, ). 460..i il / f/ ,' 4 60 k' 463 v ~ T A l t N 61 [/ M 461 / 463 Riser .A /. t ,/\\/ ./ \\' iiaisisisi,,ii / V b' Y j T 352. s U Tubes /y. y f i / v h/ _ Ol _.(&u.- ~t'/.- _ $ _p p/ ,/ '/! ( 6 G A v b l h Downcomer / /' / / 6L A G_.-.'. L i / 9 ,p v/ > - -_r - / l,/ / / / / / ,d_ I/ .d' _. / - _ d _ d__ h / / 2 f 0' / // / 7 / / l / / / / / / / / 4 / G / / &l 't i + / ~ ] g t 1, 362 1 V g ~ / so2 361 351 l,,,, , 360 y

h..

/ s-353 4 350 To B Cold Legs 4 From B Hot le9 '(. q./ ( Figure Id. RELAP5 nadalization of ANO-2 -- 8 steam generator. l

Gg,... t f. I t = C. T <= x*e OCZ p =u w s 41,' A m z .O- + \\,;. - .J ~ E 12 / E. eN ~ g ma _N / = v Y /A .o a ,T M f \\ ~ I i N O 9 = 3M 'N Z

n

= w v 0 s -= C Y p o M ~ CJ od ,y H q 4 .N y 3r 9 O 3 2 4 m 4 m O T C t~ 2 E M4 a

• • 2 C-J

%/_ C w M y ,/ ,= N C; -= o L E 3 1 Z r-.. L /e Z_ * ~ M4 ?

• ta

/ \\ 2> / ~ i W Y c. a

• W iP I

I 9

e e o O l 1h 2 Ow $d Y <x s.: t. C52 g =., w 6 O *.' T 4 z r C -= 0 \\ '/ g. ',4 ea

• E iO N b e.

/ g ~ T CQ [ /

• J

-e-s m . /u; = v ow O W b /

• / ~

s* 0 7 T \\ N I O=< I w E Q .Z. g b \\ .-0 = a 1 W Y m e = .N N 6 w r::: Y is c = o b 6 C e La 6 LA y C F= 7. c =. y< J t. LAJ e m~ c. [s\\ ~ b. T b& C "3 W T \\, [-

e.

• Z ad>

y C 's 2> / \\ W H I e C l~v t 1 I I i l i 10 A l

. _ = _ _ _ pienum. The support cylinder holes and core shroud clearance paths were lumped together as a flow path parallel to but separate from the core. The guice tube bypass flows were not modelec explicitly. Portions of the secondary coolant system, including the steam generator, main feedwater, emergency feeawater, main steam line, one safety relief valve on each steam line, one atmospheric dump valve on each steam line, and steam supply lines to the emergency feedwater ana main feedwater turbines, were modeled. The nodalization of the secondary coolant system is illustrated in Figures Ic through lf. The STGR was assumed to be initiated by a 200% double-ended guillotine break of a single tube. The break was modeled on the inlet side of the steam generator at the top of the tube sheet. Both broken ends of the ruptured tube were connected to the lows.it riser volume of the steam generator. A discharge coefficient of 1.0 was applied on each end of the break. For the single SGTRs, the break was assumed to occur in the "B" steam generator as illustrated by Figure Id. For the dual SGTRs, a break I was assumed to occur in Doth steam generators. Although not shown in Figure Ic, for the dual SGTRs a break similar to that shown in Figure ld was added to the "A" steam generator. 2.2 Initial Conditions The initial conditions for the SGTR calculations were representative of ANO-2 operating conditions at 100*, rated core power. Table 1 compares initial values of selected parameters from the RELAP5 model with desired initial conditions which were obtained from the ANO-2 Final Safety Analysis Report.2 The RELAPS initial conditions were octained from a steady-state run that used. control systems to represent the behavior of the reactor coolant pumps, the chemical and volume control system, and the main feecwater and steam control valves. The table shows that the actual initial conditions were generally in excellent agreement with the desired ir.it-ial conditions. 11 \\

5 TABLE 1. INITIAL CONDITIONS Paramete-RELAP5 Desired Core tnermal oower, MW 2815. 2815. Dressuri:er pressure. MPa (psia) 15.51 (2250.) 15.51 (2250.) iressurizer liquid volume.,3 gft3) 17.2 (608.) 17.0 (600.) Het 'eg temperature. K ( ! 595.7 (612.6) 595.7 (612.5) Cold leg temperature, K ( F) 563.1 (553.9) 562.9 (553.5) Total loop flow, kg/s (ibm /s) 15170. (33445.) 15170. (33444.) Reactor coolant pump speed, rad /s (rpm) 93.4 (892.) 94.2 (900.) Charging flow, 1/s (gpm) 2.8 (44.) 2.8 (44.) Letdown flow, 1/s (gpm) 3.2 (50.) 2.5 (40.) Pressurizer spray,1/s (gpm) 0.088 (1.4) 0.10 (1.6) Steam generator pressure, MPa (psia) 6.32 (916.) 6.21 (900.) ( Stean generator level,a % 80. 75. Stean generator feed / steam flow (each), 793. (1748.) 797. (1756.) kg/s (lbm/s) Feedwater temperature, K ( F) 506. (452.) 506. (452.) a. The parameter corresponds to the steam generator liquid level as detected cy the narrow range level instrumentation and converted to a percentage of the elevation between pressure taps. l I, - l 12 4

2.3 BoundarL onditions C i i Portions of the reactor protective and tne engineered safety feature j j actuction systems were modeled with trips. A reactor trip was initiated by 1 nigh or low pressurizer pressure, nigh or low steam generator level, or low steam generator pressure. Tne safety injection actuation signal, which started the HPSI pumps, was generated by low pressurizer pressure. The emercency feedwater actuation signal was generated by low steam generator level. l A loss of off-site power, which was assumed to occur coincident with i the reactor trip, caused the main steam isolation valves and feedwater valves to close and also tripped the reactor coolant pumps. The power table used in the model was obtained from a separate effects calculation which utilized RELAP5's point kinetics model and accounted for standard and actinide decay heat. '[ The H,'51 system was represented with time-dependent junctions which specified flow as a function of pressure when the HPSI system was operating. The total flow represented the output of two HPSI pumps. The HD51 fluid temperature was 311 K (100 F). s The chemical and volume control system, which includes the charging, letdown, pressurizer heaters, and pressurizer spray systems, was modeled mechanistically and was controlled to respond during the calculations as the actual system would in the plant. The modeled PORV was based on the PORV in Calvert Cliffs # and was sized to pass 19.3 kg/s (42.5 lbm/s) of dry, saturated steam at 16.5 MPa (Iz00 psia). The APS flow was 8.31/s (132 gpm) of 322 K (120 F) water and represented the output of'all three charging _ pumps. Emergency feedwater (EFW) was injected at a rate of 201/s (320 gpm) per steam generator, corresponding to the normal flow expected in the rlant, when the EFW system was operating. Tne normal EFW flow was 13

i i 1 l l only about half of tne maximum injecticr. capacity. Tne EFW temperature was l 304 K (55 F). The safety vahes on each steam generator opened at 7.54 MPa (1092 osia) and were sized to pass 180 kg/s (40C lbeis) cf cry, saturated steam. The ADV on each stearr generator was sited to pass 210 kg/s (46C len/s) cf cry. saturated steam at 6.21 MDa (900 psia) when the valve was full open. 1 i l i I ] 4 f I i ( 14 I l

~ _~ 4 i 3. RESULTS The results of the single SGTR and dual SGTR analyses are described in i Section 3.1 and 3.2. respectively. 3.1 Single Steam Generator Tube Rupture The results of analyses of hypothetical accidents initiated by the couble-ended, guillotine rupture of a single steam generator tube in the "B" steam generator are cescribed below. The "B" steam gene.rator, located in the loop which does not contain the pressurizer, will be called the i affected steam generator throughout the rest of this section. The "A" l steam generator, located in the loop which contains the pressurizer, will be called the unaffected steam generator throughout the rest of this section. Two RELAPS calculations of a single SGTR were performed with the l model described in Section 2. The calculations were identical except that j in one calculation depressurization of the RCS was accomplished with the APS while in the other calculation the PORV was used for depressurization. The analysis with APS-initiated depressurization will be described first followed by a description of the an'alysis with PORV-initiated depressurization. The sequence of significant events occurring in the single SGTR calculation with APS-initiated depressurization is shown in Table 2. The SGTR was assumed to occur at 0 s. An automatic response of the plant to the SGTR, with no operator intervention, was initially modeled. A reactor trip occurred at 990 s due to low pressurizer pressure. A loss of off-site l power (LOSP) and turbine trip were assumed to occur simultaneously with the 4 reactor trip. Two HPSI pumps started delivering water to the RCS 30 s after the reactor trip. Emergency feedwater (EFW) was initiated at 1040 s due to low steam generator liquid. level. The plant responded automatically with no operator intervention, until 1590 s (10 minutes after reactor trip). At 1590 s the operators were assumed to open an atmospheric dump valve (ADV) on each steam generator in order to reduce the hot leg fluid temperature below the saturation temperature corresponding to the lowest set point pressure of a steam generator safety valve. This action was taken to insure that'the affected steam generator could.be effectively. 15 m

( TABLE 2. SEQUENCE OF EVENTS FOR THE SINGLE SGTR WITH APS Time (s) Event 0.0 SGTR occurred 990 Reactor trip: LOSP: turbine trip 1000 Secondary safety valves opened 1020 HPSI started 1030 EFW started 1590 ADVs opened 2110 Affected steam generator isolated; APS initiated 2340 Upper head voiding began 2410 HPSI termination criterion met 2590 HPSI initially terminated ( 3250 Calculations terminated 16

1 isolated, with no staam discharge through its safety valves. Both dump valves remained open until 2110 s when the hot leg fluid temperature had i been reduced to 561 K (550 F) which was bElow the saturation temperature i corresponding to the lowest setpoint pressure of a safety valve. Three major operator actions were represented at 2110 s. First, the affected steam generator was isolated by securing the EFW. the EFW turbine, and the ADV in tne affected generator. Second, the ADV in the unaffected generator was throttled and a control system placed on the valve area in order to comply with the maximum cooldown rate allowed by AND-2 Technical l Specifications. Third, manual depressurization of the RCS was begun using the APS. A control system was used to maintain hot leg subcooling between 14 K (25'F) and 11.4 K (20.5'F). The control system simulated an ) operator depressurizing the RCS when the hot leg subcooling reached 14 K (25 F) and stopping the depressurization when the subcooling dropped to 11.4 K (20.5 F). Flow from the HPSI system was initially terminated at j 2670 s when a liquid level was reestablished in the pressurizer. HPSI was t cycled on and off four times after 2670 s. A control system was used to turn HPSI on when the hot leg subcooling dropped below (11 K) 20 F and turn I HPSI off when the subcooling exceeded 11 K (20*F). The calculated pressures of the RCS and steam generators are shown in Figure 2. The leakage of fluid from the RCS to the affected steam generator through the SGTR was larger than could be compensated for by the chemical and volume control system. Consequently, the pressurizer level and RCS pressure decreased following the SGTR. The rate of RCS depressurization decreased near 80 s due to energizing the pressurizer heaters. The rate of depressurization increased at 350 s when the heaters were tripped of f by low pressurizer level. The rate of depressurization decreased again at 750 s when voids were generated in the core due to f subcooled nucleate coiling. Low pressurizer pressure caused a reactor trip at 990 s. The reactor trip caused a rapid cooldown and decrease in l specific volume of the liquid in the RCS, resulting in an outsurge of flow from the pressurizer and a rapid depressurization. The increase in RCS pressure after 1050 s was due to the LOSP which caused a trip of the reactor coolant pumps at 990 s and an eventual transition from forced convection to natural circulation flow. The hot leg fluid temperature 17 l

I i increased, as the loop mass flow decreased, in order to obtain the dif ferential temperature across the core necessary to remove the core decay ] The increase ir, hot leg RCS fluid temperature caused an expansion oower. of tne RCS fluid and an increase in RCS pressure. Pressure in each of the I steam generators was not significantly affected by the SGTR and remained nearly constant until the reactor trip. The main steam isoiation valves anc main feedwater valves closed following the LOSP, thus momentarily j isciating the steam generators. The pressure in the steam generators increased rapidly following the reactor trip, due to the removal of decay 1 ano sensible heat from the' RCS, until 1000 5 when a single safety relief i j valve opened in each steam generator. The safety valves then opened and I closed as necessary to relieve pressure. The operators were assumed to l. correctly identify that a SGTR had occurred and determine which generator was affected, based on radiological considerations, within 10 minutes of the reactor trip. An ADV on each steam generator was opened at 1590 s in order to cool down the RCS. The steam flow through the ADVs depressurized the steam generators. The resulting cocidown of the RCS also resulted in a slight depressurization of the RCS. The hot leg temperatures had been [ reduced sufficiently by 2110 s to allow the affected steam generator-to be isolated. The affected steam generator pressurized slowly after it was isolated, primarily due to leakage from the RCS. A manual depressurization of the RCS was begun at 2110 s by initiating APS. APS was cycled on and off 11 times between 2110 s and the end of the calculation. The injection of cold APS water into the pressurizer condensed steam resulting in relatively large depressurization rates while the APS was on. The APS-induced depressurization allowed the upper head to begin voiding at 2340 s. The response of the RCS pressure was altered when HPSI was initially terminated at 2590 s. The effect of HPSI termination ~on RCS behavior will be described in more detail later. Based on an extrapolation of the pressure traces shown in Figure 2, the pressures of the RCS and the affected steam generator would be expected to equilibrate at 6.9 MPa (1000 psia) near 4100 s. The equilibration of pressures would effectively minimize leakage from the RCS to the affected steam generator. Minimizing l the leakage is a goal the reactor operators try to attain during a SGTR. Calculated fluid temperatures in the hot leg and cold leg of the. ( unaffected loop and in the cold leg of the affected loop are shown in 18

i f 20C00 i i a acs J c usArrEeno sTEAu cENERATCR .2500

Heaters On a ArrEcTEo sTEAw OENERATOR 15cocf/

Heaters Off i m y ubcooled tjucleate Boiling E20DC j j ]- S i g i y Reactor Trip 3 .APS Initiated _. 350e o o .nene. Safeties (NF HPSI Terminated E 4 i g y j Opened j g C C C C C Cf Sc0c I Reactor Trip ADVs Opened ~ 3

500 c

j Affected Steam Generator Isolated l i 'O O ~ C 500

• 0C0 1500 20C0
• 00 3MO 35CO Time (s) rigure 2.

RCs end stoorn generator pressures f or the single sGTR with APS. i L650 620,

i 1

a bot LEG C UNAFFECTED CCLD LEG i O +- Reacto'r Trip 600 1; 'o 60C D 6 2 l lJ e

'80l-(A HPSI Off g

3 1--k C C N H b

• '60 %

\\ -550 m ~ ~ ~ ' ' ~ 5' ADVs Opened k 540 Affected Steam Generator Isolated !.500 2 520 O 500

000 2 00 2C00 2500 3C00 35C0 Time (s) t Figure 3.

RCs rund temperatures 'or the s'rgie sGTR with Ars. 19. j u

1 Figure 3. The hot and cold leg temperatures presented correspond to k locations near the steam generator inlet and cutlet plena, respectively. No significant differences were calculated between the temperatures in the two hot legs or between the temperatures in the two cold legs attached to j the same steam generator. Consequently. only three of the six total loop { temoeratures are snown. The fluid temperatures remained nearly constant until the reactar trip occurred at 990 s. The reduction in core power folio eing the reattor trio caused a rapid reductiun in hot leg temperature. The Screase in hot leg temperature beginning at 1050 s was caused by an increased temperature difference across the core due to tne transition from forced convection to natural circulation following the trip of the reactor coolant pumps. The cold leg fluid temperatures increased slightly after the reactor trip and then stabilized. The cold leg temperatures stabilized because they were in thermal equilibrium with the steam generator temperatures which were equal to the saturation temperature corresponding to the lowest set point pressure of a safety valve. The cold leg temperatures were in thermal equilibrium with the steam generator temperatures because of a relatively long fluid transit time through the tubes curing natural circulation coupled with a large heat transfer area. j Since the cold legs were in thermal equilibrium with the steam generators. the cold leg temperatures began decreasing at 1590 s when the pressure and temperature of the steam generators were reduced by opening the ADVs, The reduction in hot leg temperature due to opening the ADVs was delayed 60 s because of the fluid transit time between the cold and hot legs. The increase in the temperature of the affected cold leg at 2110 s was due to the isolation of the affected steam generator. The cold leg temperature in the affected loop generally exceeded the hot leg temperature after 2480 s as the affected steam generator acted as a heat source rather than a heat sink. Consequently, the RCS fluid began cooling down the secondary liquid adjacent to the tubes at 2480 s. This ccoling resulted in temperature stratification, with three distinct regicns, in the af fected steam generator. The lowest region, which was adjacent to the tubes, contained subccoled liquid. (The liquid.was 3 K (6 F) subcooled at 3250 s when the calculation was terminated.) The middle region contained saturated liquid. The uppermost region contained steam. l ( 1 20 4 a s ~..., _.. ,-.-_r--.

l l Turning HPSI on and off significantly affected the calculated response of the hot leg fluid temperature. HPSI was initially terminated at 2590 s when the pressurizer liquid level had suf ficiently recovered. Wnen HPSI was flowing, the fluid temperature at the injectior. location cecreased because of the mixing of the cold HPSI with the cold leg fluid. The temperature at the injection location increased when HPSI was termiAated and 60 s later this temperature inc* ease arrived at the hot leg. The increase in hot leg temperature caused a decrease in hot leg subcooling. The control system used in the calculation turned HPSI back on because of low hot leg subcooling and about 60_s later the hot leg fluid temperature began to decrease. Thus, the four oscillations in hot leg temperature af ter 2670 s were caused by the manner in which HPSI was terminated and restarted in the calculation. In reality, the reactor operators would tnrottle HPSI and/or would stop one HPSI train at a time. Thus, the plant would not be expected to experience the temperature oscillations tnat occurred in the calculation, although the cooldown would be affected by a reduction in HPSI flow. The average RCS fluid temperature is shown in Figure 4 A weighted 1 average of the two hot and four colo leg temperatures is presented. A l reference temperature based on the maximum cocidown rate allowed by ANO-2 Technical Specifications is also shown.a One ADV on each steam generator was opened 9'o of full open at 1590 s to cool the RCS. The ADVs were opened 9'; in order to produce a cooldown rate about twice that allowed by Technical Specifications based on the assumption that the operators would want to cool down quickly in order to isolate the affected steam generator. The affected steam generator was isolated at 2110 s. The ADV on the unaffected generator was throttled to 3*. full open'and a control system placed on the valve area at 2110 s in order to comply with the cooldown rate allowed by Technical specifications. The termination of HPSI at 2590 s caused a decrease in the cooldown rate. The ADV control system responced by opening the valve area to 13*4 full open which was the maximum a. The Technical Specifications allow a maximum cooldown rate of 56 K (100 F) in any one hour period when the average RCS temperature exceeds 330 K (225 F). 21

^ E-590 - -500 l RCS !i r-REFERE!!CE ; , Reactor Trip .j 5s0 - m E', ! *80 m I p x [ADVsOpened S

• 57c -

V

• q.

-sec 2 3 3 C w 3' " 560 - 8 j C. E Affected Steam L540 3 o Generator Isolated " 550 - y \\. HPSI Terminated MD, 5,o -i 540 C 500 'OCD 15C0 2000 2500 3000 35C0 ] Time (s) ri gu r e 4. Averege RCS fluid tenseroture f or the single SOTR witn APS. ( i 6 i 4 ~200 HPSI Terminated I g i 1 m S' .E., 4 -- --15 0 _ -t, l v i t e Q 1, 2 h100 ._s - i t i

• s a

cr 2r ADVs cr 3 l Reactor Trip Opened - i' 3' J j 50 i Elevation / / of durge ,APS Initiated i

Line No2Zle-*-

0 O C 500 1000 15C0 2CCC 2500 3C00 3500 Time (s) Figure 5. Pressurizer !!quiJ level f or the single SGTR with AAS. 'j e i y i 22 ._ _, J_- f Je r e OOf,

I area arbitrarily allowed by the controller. An increase in cooldown rate sufficient to offset the effect of HPSI termination could have been obtained by further opening the ADV on the unaffected generator. The calculated pressurizer liquid level is shown in Figure 5. The leakage from the RCS to the affected steam generator caused a decreasing p-essurizer level until reactor trip occurred at 990 s. The RCS cooled cown rapidly following the reactor trip which caused a large outsurge from tne pressurizer and a rapid decrease in level. The liquid level dropped to the top of the surge line nozzle near 1000 s which allowed steam to vent from the pressurizer into the hot leg of the unaffected loop. Because the surge line nozzle extended above the bottom of the pressurizer, as illustrated in Figure la, a residual amount of liquid remained in the oressurizer below the surge line nozzle and prevented the liquid level from i going to zero. The transition from forced convection to natural circulation following the LOSP caused a heatup and expansion of the RCS fluid and a corresponding rise in pressurizer level. The RCS cooldown and corresponding decrease in liquid specific volume that resulted from opening the ADVs at 1590 s caused the level to again drop to the top of the surge line nozzle. The pressurizer level rapidly increased shortly after 2110 s. The increase in pressurizer level was primarily caused by APS, which depressurized the RCS resulting in increased HPSI flow, and the decrease in cooldown rate that occurred when the ADV in the unaffected steam generator was throttled and the affected steam generator was isolated. HPSI was initially terminated at 2590 s which caused a slight reduction in pressurizer le,el. HPSI was then cycled on and off for the remainder of the calculation causing a gradual increase in pressurizer level. Calculated hot leg mass flow. rates are presented in Figure 6. The mass flow rates remained nc'.rly constant until 990 s when a LOSP was assumed which caused the reactor coolant pumps to trip. The mass flows coasted down until near 1300 s when the flows stabilized, indicating that the transition from forced convection to natural circulation had.been completed. The flow through the affected loop decreased after the affected. . steam generator was isolated and stopped acting as a heat. sink. The 23 pr

~ f. i i SC00, Y - - - _a + Pumps a imArrECTED LOOP f j Tripped o AFFECTED '.0CP !- 150C0 i scoe,- O m< l N E m 6 'scoco $, l I 3 l j acco - i o 3 l tlatural Circulation i U Established I a } h5000 E t + 1 1 ~

• CCC Affected Steam Generator Isolated i

,i l /: v =&:::6 0: 0 C $00 'OCD 1500 2000 2500 3C00 35C0 Time (s) Figure 6. Fot leg mess flow rates f or the single SCTR with APS. (' 25 l so { i 20 APS Isolated 7 m 7 Reactor Trip j # [4 h D

/

/ E t huA l)l' ! -30 i /v 5 E j j. l o j i I., ! tJ ADVs Opened a ~ 10 r ..,3 E l i N 2 - (J 1 o2 5* 10 I I i i C 500 1000 15c0 2000 2500 3000 3500-Time (s) I. TIgure 7. Erock reass ftow rot e f or the sigle SGTR wi*h APS. 1 ( l l 24 ~

__ - _ ~ ( driving potential for natural circulation flow consisted of the differential pressures across the vessel and steam generators due to temperature differences. When the affected steam generator stopped acting as a heat sink, it no longer developed the temperature differences that contributed to the driving potential for natural circulation. Thus, the flow through the affected loop was reauced after the affected steam generator was isolated. The calculated mass flow rate from the RCS through the SGTR to the affected steam generator is shown in Figure 7. The flow shown represents the total break flow through both ends of the SGTR. The break flow generally decreased with the RCS pressure. The break flow decreased rapidly near the time of reactor trip due to the presence of a small (less than 2%) void fraction at the break. The break flow increased once subcooling in the hot legs was recovered and further increased when the ADVs were opened. The isolation and repressurization of the affected generator and initiation of APS to depressurize the RCS at 2110 s decreased the break flow. The total HPSI mass flow rate into the RCS is shown in Figure 8. The safety injection actuation signal was generated at 990 s due to low pressurizer pressure. HPSI flow was initiated 30 s later to account for delays due to starting the diesel generators and the HPSI pumps. The HPSI performance was determined by the RCS pressure until 2590 s when a manual 6 termination of HPSI was simulated. CE guidelines for SGTRs allow the reactor operators to terminate HPSI if the RCS is at least 11 K (20 F) subcooled, the pressurizer level is greater than 2.54 m (100 in.) and constant or increasing, and at least one steam generator i.s available for removing heat from the RCS. HPSI termination may be desirable in order to prevent overfilling of the pressurizer or the affected steam generator. The HPSI termination criteria were first met at 2410 s in the calculation, but HPSI flow was continued until 2950 s when the pressurizer level reached 4.45 m (175 in.). A control system was then used to turn HPSI on or off depending on whether the hot leg subcooling was less or more than 11 K (20 F). This control system was responsible for the four on/off cycles of HPSI after 2590 s. In reality, reactor operators would probably throttle HPSI rather than terminate it, as was assumed in the calculation. 25

I 60 -12 0 l HPSI Initially Terminated s f g i s(('d [ico O 'k!:l!', Q,N J N 40 - ? i C 3 .- i 7e0 e J m l ll hs0 8 1 i, .oo = 20 l i ;i iti I l 6 a o a ll F'O S 1 l i o il 'i l i 1 2 \\ i APS p,! j J ' 20 -i I. l I Initiated I j '. ' i l j : i 1 0' 'o 0 500 'OCO 15C0 2CCC 2500 3C00 3500 Time (s) Figure a. Total HPSI mass flow rote f or the single SGTR with AFS. 10 - 1 HPSI Initially Terminated 20 I e ' t :C "l j.' Y - .'I a-a l j ni i! . l m i 7 l pl; h' l i1 I \\ i ol ' N r15 m i

'I tit '

• t i

l l!

E + i 6,' '... U l '.l :i 'l - J ..Q 3 9' j yl ] ;;, l ji l' q i v J j 'i.' j t. i.. v j 3 i . '!, l l i 3 ii.. ;. i

j c

4 t *.i ' 10 o 4 - h., ' i ll =

ll(

o. -

1 a

u..; n a

o .! m i 3 *.. n I j li l o i o 2 l .I'; :l l, ; o li[ ly l.;ll:'! .l t '5 2 2, n;y,,.; ll

j '- '

i l l MF.-l:i. ,I.:, , U il i Niibl

Oi!! i'u.l.

j i! ii i l 0 0 0 500 !OCO 15C0 2cCC 2500 3000 3500 Time (s) T i gu r e 9. APS ncss ficw rate f or tre sirgle SGTR. ( l l 26

i The calculated APS flow rate is snown in Figure 9. The maximum possible APS flow, representing the output of all three charging pumps, was assumed when APS was operating. APS was started and stopped 11 times curing the calculation to simulate a manual depressurization of the RCS based on hot leg subcooling. The frequency at which APS was initiated decreased af ter 2590 s when the cooldown rate slowed due to the termination of HPSI. i The calculated hot leg subcooling was nearly identical in the affected and unaffected loops,'as shown in Figure 10. The subcooling was determined by the RCS pressure and hot leg temperature curves shown in Figures 2 and 3, respectively. Hot leg subcooling began increasing shortly after the ADVs were opened at 1590 s. The subcooling increased until 2110 s when APS was initiated. The flow of cold APS into the pressurizer depressurized the RCS and consequently reduced hot leg subcooling. APS was turned off when the subcooling dropped to 11.4 K (20.5 F). Continued cooling of the RCS caused tne subcooling to increase to 14 K (25 F), at which point the APS was turned on again. The cycle of turning APS on and off based on hot leg subcooling was repeated eight times between 2110 s, when manual depressurization was begun, and 2590's, when HPSI was initially terminated. The termination of HPSI significantly affected the RCS and, in particular, affected hot leg subcooling. As explained previously, the hot leg temperature increased after HPSI was terminated, thus lowering hot leg subcooling below 11 K (20 F). CE guidelines require the operator to i restart HPSI if the subcooling decreases below 11 K (20 F). Consequently, HPSI was restarted and, in addition, the ADV on the unaffected steam generator was further opened to obtain the required subcooling. HPSI was-terminated for the second time at 2780 s when the subcooling increased to 11 K (20 F). Due to the loop transit time the hot leg subcooling continued to increase until it reached 14 K (25 F) at 2810 s and APS was turned back The hot ieg subcooling was determined by the cycling of APS and HPSI on. for the remainder of the calculation. After isolation of the affected steam generator, the maximum loop fluid temperature and minimum loop subcooling generally occurred in the cold legs of the affected loop, as I 27 j l i

3C, A UNAFFECTE0 HOT LEC j O AFFECTED HOT LEG G L40 m a J b 5 20 k APS Initiated !\\ N ci ? i )h%^fe})(P ADVs Opened _-20 j ,o _ \\ 3 = 3 s i 3\\ 3 c- / HPSI Initially Terminated _o ] I w 6 I I i, -scl C 500

• 0C0 15C0 2CCC 2500 3000 3500 Time (s)

Figure 'O. Hot leg subcooling f or the single SCTR with APS. 14 A UNAFFEC*ED STEAW GENERATOR 1 O AFFECTED STE/.M GENERATOq -500 d 12.- l l 9

2. - Reactor Trip 7

5 10 % f 3 -4C0 - 3 T e> ^ i o "f c - 0[ h SW 1 I 300 I f7 l,5" i (k I l t j -b 6 !- ' f. A3('f.k(, j i i lI j h l i i: ,200 i J d ~ I .t l-- i l 10 0 y* t 0 500 1000 15CO 2000 2500 2000 3500 Time (s) F i gu r e 11. St eam generato*.iquid tevels f or tre single SGTR with APL 1 l I L', i 28

i ~ illustrated by Figure 3. Consequently, after isolation of the affected steam generator, the reactor operators should closely monitor the colc leg where tne minimum loop subcooling may occur. Calculated steam generator liquid levels are shown in Figure 11. The levels were computed by integrating the downcomer liouid fraction from the top of the tube sheet to the top of the downcomer. The liquid levels remained nearly constant until 990 s when the reactor trip and LOSP occurred. The feedwater valve in the affected steam generator was assumed to modulate to compensate for the SGTR which allowed the level to remain nearly constant prior to the LOSP. The rapid decrease in liquid levels after the LOSP was primarily due to a redistribution of fluid mass between j the downtomer and riser regions following the closure of the feedwater and main steam isolation valves. The steam generator levels generally increased after tne initiation of EFW at 1030 s. However, an anomalous decrease in level in the unaffected steam generator was calculated at 2700 s. This decrease in level was due to a redistribution of fluid mass within the steam generator and was not realistic. The level in the affected steam generator was generally higher than in the unaffected generator due to the break flow from" the RCS through the SGTR. Based on an extrapolation, the liquid level in the affected generator would not reach the bottom of the separators before 5000 s. Since the pressures of the RCS I and affected steam generator were expected to equilibrate near 4100 s, which would prevent any f urther increase in steam generator level, the affected steam generator should not overfill during a single SGTR. Consequently, a challenge to the safety relief valves due to overfilling l the affected steam generator could be avoided. The single SGTR calculation with APS was terminated at 3250 s. The reactor operators appeared to have reasonable control over the plant at the end of the calculation and no major problems appeared imminent. The status of the plant at the end of the calculation is summarized below. Core decay power was removed through the ADV in the unaffected steam generator. The affected steam generator was isolated and was repressurizing slowly due to leakage from the RCS. The RCS was depressurizing slowly, primarily due to i l 29 l )

ir.termittent use of APS. The void fraction in the upper head was increasing as the RCS was depressurizing. The affected steam generator was being cooled by the RCS and was thermally stratified. HPSI was being-cycled on and off to maintain hot leg subcooling. The intermittent use of HPSI was causing a slowly increasing pressurizer level. In order to prevent an eventual overfilling of the pressurizer due to intermittent HPSI, the ADV on the unaffected steam generator would have to be further openec to provide tne necessary subcooling in the absence of HPSI. An additional single SGTR calculation was performed in which the RCS was manually depressurized with a PORV rather than with APS. This additional calculation was restarted from the previously described calculation at 2110 s. The model used to make this additional calculation was identical to that used previously except that manual depressurization was accomplished by opening the PORV rather than by starting APS. The results of this calculation are documented in Table 3 and Figures 12 through 21. Since this calculation was restarted from the calculation with APS-induced depressurization at 2110 s, the results orior to 2110 s are identical to those shown in Figures 2 through 11 and hence will not be discussed again. A review of the results after 2110 s shows that the calculated behavior was generally similar to that described'previously. However, some slight differences were noted. For example, the pressurizer level was generally lower with PORV-induced depressurization than with APS-induced depressurization, as shown by a comparison of Figure 5 and Figure 15. The relatively small difference in pressurizer level was caused by the different depressurizing mechanisms associated with the PORV and APS. The PORV depressurizes by venting steam from the pressurizer which decreases the total RCS fluid mass. The APS depressurizes by adding cold water to the pressurizer which increases the total RCS fluid mass. A second difference between calculations was caused by the more rapid RCS depressurization associated with opening the PORV compared to starting the APS. Since less time was used to depressurize the RCS in each individual cycle, more depressurization/repressurization cycles were calculated with the PORV than with APS. A comparison of Figure 9 and Figure 19 shows that APS was started 11 times between 2110 s and 3250 s while the PORV was ooened 14 times during the curresponding time period. 30 l

TABLE 3. SEQUENCE OF EVENTS FOR THE SINGLE SGTR WITH PORV Time (s) Event 0.0 SGTR occurred 990 Reactor trip; LOSP; turbine trip 1000 Secondary safety valves opened 1020 HPSI started 1030 EFW started 1590 ADVs opened 2110 Affected steam generator isolated; PORV cpened 2340 Upper head voiding began 2460 HPSI termination criterion met 2670 HPSI initially terminated 3250 Calculations terminated b l t 31

?\\ 20CCC lI . a RCS 4 ,' O UNAFFECTED STEAW CINEPATOR l { lO AFFECTED STEAW OENERA TO'_ .l2500 '5000 % 4 7 r2000 0 2. 1 d N. I I v 'CCC0 - 1500 o 3 m w - ~ ~ ~~ -~:~= I= = u {000 g r1 0 5000 - I ~ r500 I 0 -0 0 500 'OCD 15CD 2CCC 2500 3000 3500 Tirne (s) rigu r e 12. RCS one stec*n generator pressures f or fl'e single SGTR with PORV. ( 520 1 L l l 650 6 HOT LEG C UNAFFECTED CCLD LEG - C AFTECTED COLD '_EG I S00 r 2. p 4 i i ,-600 ( v [ $80 '- } ~ e _s I' s 3 3 1 ----- Lc } O [560 \\ e t ~ ~ ' ~ ~ ~ 550 i c. = 1 f, % E 'V] 3 l N 4 I- ~~ 540,L l F500 \\m ~ 520 C 500

• 000 1500 2000 2500 3C00 3500 Time (s)

Fi gu r e 13. RCS flute romperoteres f or 'the siagle SGTR with PORV. 32

=.... _ 1 j I i j i $90 - 600 l ! - RC3 i l--- REFERENCE } 580 - l i I ['., 580 n m x t o v [570h \\ e a g I 2 i 6 s'. L560 \\', } s l

• 560 -

w d f =- 8 1 C. E 540 a j .l k e n 1 550 *- l u L520 1 i I I { 540 0 SCO

• 000 15C0 2CCC 2500 3000 35C0 i

Time (s) { F i gu r e 14. Avercge RCS f?uld ter9eerature for the sirgle SOTR whh PCRV. r a i.. ll f 6 a 3 r . goo 1 } N m I 1 e mc 4 0 4h C 1 's t 15 0 7 i 2 l 4 I i = I i 2 plc0 e 2l- = 4 J i cr g-k i a r50 f f / .j 1 3 0 O C 500

• 0 C0 1500 2CCC.

25CC 3CCO 3500 j Time (s) F i gu r e 15. Pressurizer Equid sevel for the single SGTR with PORV. 3 33

8000 0 *'**j l A LNAFFECTED LOCP - 1 i t O AFFECTED LCCP - 150C0 1 I 6000l-I O mm I N N 9 O ,S g -1C000 ,> 4C00 - 3 O t o = i I = a y j g -5000 o ,- 200C - 2 i i i i j i 1 -M.--3 0 0 0 $00 'OCO 15C0 2000 2500 2000 3500 Time (s) r;gure '6. -o t tog rness fi c e rates 'cr the single SGTR with PCRV. 25 l50 I 20 - i ^ ^ 1 xo g E 4 W.l j j l, 3 , / [', g Ii y gi[1)7. j. h' af i 2 .i o i "j l'U.j! 'l.. li j. ' 'ii:" 10 r h = -20 3 I)F i1'1U 5.ne a ^ l' o et U l o s

1. d 2

ii 3 ,10 ' I I i i 0 0 0 500

• 000 1500 2000 2500 2000 3500 Time (s) rigure '7.

Erock rross fi ow rate f or the sir;le SGTR with 3CRV. 34

i 60 I I 2 F t20 k 7100 7 N [h \\ !! I 5 e i i N N i n. 99 l, ; E ) 40 r N i- ! -50 -Q a N v 3l t ,s i 4 i j 3 i....I -60 0 3 J I I i ; i = i $ 20 -- - 40 m p t i 3 / O f l 2 2 .l I ! \\ /" i di i o i 0 O O 500 0C0 15C0 2C00 2500 3 COO 35C0 Time (s) Figure 18. Totcl FPSI mass flow rate for the single CGTR with PCRV. 15 30 r25 Q 1 l g l s m= )- 10 W i i 4 20 2 6 l I i 1 l: D o i ~15 0 } = i 5 l- [ W e r. a LO n I o i i 0 I'.,i,i 2 i l 1 2 4 l i 1, ( j' i !j }; f5 1 i j ? j;. hp i' i { O 0 O 500 '0C0 1500 2000 2500 3000 35C0 Time (s) F i gu r e '9. PCRV mass flow rate f or the single SG7R. 35

j 1 ( l 30 A UNAFFECTED HOT LEG O AFFECTED HOT LEG ~ 6 [40 m m M 2C ~ ~ v v = l c, .O C i a s O NI vb k/ _;o a 10 - \\ o D 1 D I 5 5 g s% is% 2 0 0 I L 6 I 6 i l' -10 4 C 500 10C0 15C0 2000 25CO 3000 3500 Time (s) Figure 20. Hot tog succooling for the single SOTR with PORV. 14 A UNAFFECTED STEAW GENERATOR i 500 O AFFECTED STEAM GENERAT04 12 1 10f 3 ,-400 e s g. c e _ -7,,si 300 - 7 I kbE k t ./ t E(f li 'h](I i h h6 d 1 [ -200 a -I

• I kb

~ b t00 i i 0 500 10C0 1500 2000 2500 2000 3500 1 Time (s) Figur e 21. S t ect, generator liculd levels f or the single SGTR with PORV. (- ~36 +

Comparison of the results shown in Figures 2 through 11 with the results shown in Figures 12 through 21 shows that tne thermal-hydraulic response of the plant was similar regardless of whether manual depressuri:'ation was initiated by the APS (Figures 2 through 11) or the PORV (Figures 12 through 21). The reactor operators appear to have aceauate control of the plant during a SGTR with either the APS or the PORV. Consequently, there appears to be little benefit in installing a PORV in ANO-2 for the purposes of mitigating a single SGTR since APS appears adequate wnen feedwater is available. 3.2 Dual Steam Generator Tube Rupture Analyses of a hypothetical accident initiated by a dual SGTR were performed. The dual SGTR was initiated by the simultaneous rupture of a single tube in both steam generators. The model and method used in the dual SGTRs calculations were identical to those used in the single SGTR calculations except for noding changes to represent the rupture of an additional steam generator tube. Two calculations of a dual SGTR were performed. The calculations were identical except that in one the RCS was cepressurized with the APS and in the other the RCS was depressurized with the PORV. The analysis with APS-initiated depressurization will be discussed first. The sequence of significant events occurring in the dual SGTR calculation with APS-initiated depressurization is shown in Table 4. The total break flow rate from the RCS to the steam generators was larger for the dual SGTR than for the single SGTR. Consequently, the RCS depressurized more rapidly with the dual SGTR which resulted in an earlier, 350 s versus 990 s, reactor trip. All significant events occurred earlier in the dual SGTR calculation than in the single SGTR calculation. The ADVs on both steam generators were opened at 950 s, corresponding to 10 min af ter reactor trip, in order to cool the RCS. The "B" steam generator was isolated at 1350 s when the hot leg temperatures had been lowered sufficiently to prevent any further discharge of radioactive steam through its safety relief valves. CE guidelines instruct reactor operators to isolate the steam generator with the highest radioactivity in the event 37

TABLE 4. SEQUENCE OF EVENTS FOR THE DUAL SGTR CALCULATION WITH APS Time (s) Event 0.0 SGTRs occurred 350 Reactor trip; LOSP; turbine trip 360 Secondary safety valves opened 380 HPSI started 390 EFW started 950 ADVs opened 1350 "B" steam generator isolated 1500 APS initiated 1740 Upper head voiding began 1970 HPSI termination criterion met 2160 HPSI initially terminated 2750 Calculations terminated ( 38

l tnat both generators are affected by SGTRs. Since both generators are equally affected in the symmetric dual SGTR, the radioactivity in each generator should be nearly the same. Consequently. the operators could isolate either generator in a dual SGTR. The "B" steam generator was isolated in the dual SGTR calculations in order to be consistent with the single SGTR calculations. The cooldown of the RCS was continued after 1350 s using the ADV on the "A" steam generator. Since the "A" steam generator was assumed to be affected by a SGTR, the use of its ADV to cool the RCS would release radioactive steam to the atmosphere. Depressurization of th'e RCS was begun at 1500 s when the hot leg subcooling reached (14 K) 25 F and APS started. The calculation was terminated at 2750 s. I l The calculated pressures in the RCS and steam generators are shown in Figure 22. The results were generally similar to those discussed previously except that the pressures of the RCS and isolated steam generators equilibrated near 2600 s. Since one of the operators' goals during a SGTR is to equilibrate the pressures to minimize leakage from the RCS into the isolated steam generator, the analysis indicated that this goal could be successfully achieved. The operators would begin blowdown of the isolated steam generator into the radiation waste system prior to any further depressurization of the RCS to prevent flow from the isolated steam generator to the RCS which would dilute the boron concentration of the RCS. An anomalous pressure rise was calculated at 1820 s in the isolated steam generator. The pressure rise was not thought to be real and was caused by a momentary numerical problem in the calculation. Hot and cold leg temperatures are shown in Figure 23.* The cold leg temperature in the "B" loop generally exceeded the corresponding. hot leg temperature after 1800 s as the steam generator began acting as a heat source rather than a heat sink. The cooling of the isolated steam generator by the RCS resulted in a subcooling of the liquid adjacent to the in the isolated steam generator. The liquid adjacent to the tubes ' 'c was 8 K (14 F) subcooled at 2750 s when the calculation was terminated. l 1 I 9 1 I

+ l t i 20000 i A RCS O A STEAM CEllERATCR r2500 l O 8 STEAM GENERATOR i Seco I ? d c. t 200c 9 a i s n ~ Pressures i ,S 2 Occo - Equi libra ted .c iS0c o s r, -] % - T wa L coc o c i w w w a u 5c00 L ~ C-g I / ~ 2 3 !500 B Steam Generator Isolated lo 0 0 500 1000 1500 2000 2500 30C0 Time (s) Figure 22. RCS cnd steam generef or pressures f or t he duc t SGTR wif h APS. 500 Y A hot LEG J O A CoLO LEG 600 0 B COLD LEG m 580 x ^ v u_ b \\ B Steam Generator Begins \\ Acting as Heat Source 3 ,g g 3 560 p-h j -550 2 3 I T-E a e

• 540 g~

J 4 g l500 1 520 r 0 500 ?000 1500 2000 2500 ~3000 Time (s) Figure 23. RCS fluid ten porof ures f or the duct SGTR with APS. ( 40

i i The calculated RCS average fluid temperature, pressurizer liquid level, and hot leg mass flow rates are shown in Figures 24 through 26. The { results are similar to those described in Section 3.1. The calculated mass flow rates from the RCS through the SGTRs to the f-steam generators are shown in Figure 27. The mass flow rate into the "B" steam generator decreased after the steam generator was isolated at 1350 s i due to the equilibration of pressures shown in Figure 22. At the end of the calculation, the flow was from the isolated steam generator to the RCS. The mass flow rate into the "A" steam generator was nearly constant after the ADV was opened at 1350 s. Opening the ADV allowed a direct flow path from the RCS through the "A" steam generator to the atmosphere. The calculated HPSI mass flow rate, APS mass flow rate, and hot leg succooling are shown in Figures 28 through 30, respectively. HPSI and APS l were initiated and terminated according to the same control systems used in I the single SGTR calculations. Consequently, the results were similar to those described in Section 3.1. f Calculated steam generator liquid levels are shown in Figure 31. The level in the isolated steam generator was nearly constant at'the end of the i calculation because the RCS and steam generator pressures had equilibrated. An additional dual SGTR calculation was performed in which the RCS was depressurized with the PORV. This additional calculation was restarted at 1350 s from the previously described dual SGTR calculation. The results of the additional dual SGTR calculation are documented in Table 5 and in i Figures 32 through 41. The results are similar to those shown in Figures 22 through 31 for the dual SGTR with APS-induced depressurization. Consequently. the results are presented only to document the calculation. A discussion of these results will not be presented. The dual SGTR calculations showed that ANO-2 would respond'similarly snether manual depressurization was conducted with the APS or PORV. The operators appeared to have adequate control over the pressure, temperature, and level in the RCS in either case. Consequently. it does not appear l 41 i

---

,,n

,,n. ,,w, 7 ~,

I 1 i i i 530 - RCS ] 580 REFERDICE l 570 p 560 ^ ^ 4 )*- M v v 8

• 560 '

\\ b i \\ '.=s0 \\ I~ [ 3 i [550p { e s e t $;0 8 j l \\ ~ 54C - t 4 500 4 I 530 C 500

• 000 1500 2000 2500 30C0 Time (s)

Figure 24. Average R2S fluid temperature f or the dual SGTR with i APS. f 6 a 1 200 -5 E 4[ 9 v -150 v i e e I e l _e e 10 0 J-f m_ t 4-s 3 1 7 T 2 l \\ 'i a _.s , 50 1 i I l d i N i 1 I t 0 g 0 500 10C0 1500 2000 2500 3000 Time (s) Figure 25. Pressurizer liquid level f or the cual SGTR with APS. ~ ] 4 42 J 3 e e

5C00 i** A A LOOP C 0LU 150C0 7 6C00 - o I s m N G3 1 , 10000 0 v 3 ] [ 4CCC - r i E i t '5000 e O J 2 2000 - 2 \\. 3

. 2Mt.

k I b 0 C C 500 'OCO 1500 2000 2500 3000 Tim. (s) Figure 26. Hot leg rnass flow rates f or the duct SOTR with APS. 30 g a A SGTR 60 0 B SGTP 7 20 x n r 40 x ) Q&sp~,D c E l 5 } /. W'~JdLj'i -- 20

• 6 "iM U.

i.it ? 3 iC '- o i B Steam Generator \\o'y = 1 ] [g, t { g g l Isolated -0 2 06 l e i 4

v j

i l-i l -20 4 - 10 0 500 '000 1500 2000 2500 30C0 Time (s) Figur e 27. Sreck muss flow rates f or the cuoi SGTR with AFS. I, 5 I i 43 l.

_ _ _. ~ _ - _. i l i i J 60 ; l -12 0 1 10 0 Q m b 40 m 1 E 4 8 l g I i, rSO e y p, [ i I ;! !! fc. y o I o = 1 l' , 60 ,o I $ 20 b . i I e o i t i 40 a 3 { O li 2 l'l'ij A ,I i i 5 l l i1 i ig l 0 O O 500

• 0C0 1500 2C00 2500 3000 Tim. (s)

Tigure 28. Total HPSI mass flow rcle for the duct SGTR with APS. .1

l 10 i

4 -20 u si r r r' l 7 i t l i l e ! I ! 'l i I l I-15 D 1i ,t i l e o i $$h .j lj. I 3 I { 4 il i i l ~ i i it : I 'l [ t ll , 10 o, l 8 4ht ] 4 l i s

l. i i

l [ l l a i t 4 l l l! i o 3 l.' 5 3 2p 11 l i' i; l t !; ;4 3,! .i 1 I 1 lR ll !!!!!! ll.il o j 0 0 0 500 'OC0 1500 2000 2500 30C0 Time (s) l'igure 29. APS mass flow role f or the duct SGTR. a 4 .) i ('k l 44

30 ) b50 A A "0T LEO I l,1 ,O B FOT LEO [ E 9 .* b 3 20 'f Y o ] j C = 0 i = o 30 o 2 o u t A o bl 20 7 10 '- t 3 3 \\ j 3 4 i l t0 l \\. t ,- C Cg 0 O C $00 '000 1500 2C00 2 00 30C0 Time (s) Figure 30. Hot leg succoolf r ; f or the cual SGTR with ADS. I 14 A A STEAM GENERATCR C B S~EAM GENERATOR j500 12 l-e c 2 d 10 "i-gM ,400 i I 3.~ d, ~ m I t a 300 t L:9 t1 y ~ c-S; i r h,C. _C' m l a f i.*l -200 J I l l I. jt i 46 nQ j -i. O 4 a f -tCO 2 0 500 '0C0 1500 2000 -2500 30C0 Time (s) Figure 31. Stecm generatcr licuid levels f or the duct SGTR with AFS. 45

necessary to install a PORV on ANO-2 to mitigate the effects of a dual SGTR given the current CE guidelines. However, the addition of a PORV and a change in guidelines to cool down the RCS by primary feed and bleed might recuce the off-site release of radioactivity when both steam generators are affected by SGTRs. Of course, the addition of a PORV would also provide a benefit as a redundant depressurization system, e I i 46

TABLE 5. SEQUENCE OF EVENTS FOR THE DUAL SGTR WITH PORV Time (s) Event 0.0 SGTRs occurred 350 Reactor trip; LOSP; turbine trip 360 Secondary safety valves opened 380 HPSI started 390 EFW started 950 ADVs opened 1350 "B" steam generator isolated 1500 PORV opened 1740 Upper head voiding began 2040 HPSI termination criterion met 2250 HDSI initially terminated 2750 Calculations terminated 1 47

j l t 20C00 fA RCS 2 <C A STEAM GENERATCR 2500 l 'O 8 STEAM GENERATOR k i '5000 1 e -2000 0 mo 1 I c. l O'" .x i ) 8 'OC00 "- 1500 o 3 l .b ~ .C C C s'._--- p -g -Q l'1000 M i m C w p., w, c., ~~ 5C00 - 3 l 500 1 0 O 0 500

• 0C0 1500 2000 2500 2000 Time (s)

Figure 32. RCS cnd steam generator pressures f o-the dual SGTR with PCRV. 4 600 A FOT LEG O A COLO LEG 600 0 B COLD LEG l m 580 -t n x I l F. e_ -3 v v 3 e 4 t w = a r + 3 c-550 "60 - O u o e ~ a I V~N j

• $40 p F

b 500 520 O 500 "000 1500 2000 2500 3000 Time (s) Figure 33. RCS fluid ter-ceratures f or the dual SGTR with PCRV. ~48 i I t i

580 - - , 580 e j RCS I I

--- RETEPENCE 570

• j (560

.i ^ n i E x { v v

• $60 -

\\ u s .=. \\ -540 t c 3 l i. o egg ll-e c. O. - E [ 6 r$20

• y'.,

I ) 540 - t \\

l 500 f

530 O S00

• 0C0 1500 2000 2500-30C0 Tim (s)

Figure 34. Average RCS fluid tempercture f or the dual SGTR, wit h PORV. l 6 i I / '= 1 ] i e i 2 j 54h ~ -150 [ e l / Y100 t t a t 3 3& 2 ~ g J _J 50 ..l .I ~ 0 0 0 $00 1000 1500 2000 2500 30C0 Tim. (s) Figure 35. Pressurizer liquid level f or the cuot SGTR n th PORV. 49

300C A A LCOP I.is000 O 8 Lo0P J l 6C00 l-7 n l l C 3 l n a - 100C0 O i d [ 4c00 - 3 .2 = l I {5000 n S 4 I ,E 2000 - i d \\, \\. 0 sc h g i 0 0 S00 0C0 1500 2000 2$00 3000 Tim. (s) Figure 36. Hot leg mots flow rates for the duci SOTR nith PORV. (. 30 A A SGTR - 60 O B SGTR n 20 d. 7 ( f40 N

c f

1. qf?le*.vpa O-g 3

q,4' ' h I.* c k 'C '- ) '." I ~ l j fD E 6 3 I 1 0 2 0 kr 4 4 f - / l0 2 l -20 -10 ' 0 $00

• 000 1500 2000 2500 30C0 Time (s) rigur e 37. Brook mess flow rates f or tre cwol SGTR with PCRY.

[' ' j: d 50

60 I i20 N +a.! s s 1' 10 0 Q g m ( 40 l-A N \\b e ,I -50 3 c, / l' i t .:s / !,) i aj O i a r q f t y h60 [ } i $ Ze '-- e b 40 m c i I L l 2 1 l 2 i } 1 ll -20 l I i O 0 O SCO '0C 0 1500 2000 2500 30C0 Tim. (s) Figure 38. To*cl HPSI mass flow rate for the duct SGT9 alth PORY. 15 r3a ] i i I l -25 Q n ) 6 i r20 e i a -l l. l k i V 4 e u.). I 3 3 6 6 j o i ft5 } l.!. o = 1 l j - 1.' l l. a d e 5.r - 1.j - 1 ij e .jo a i c C 2 i 2 I f 3 . ii -5 4 i: i, : I ..i. ) :-

4. 4[j i.

i .I o i i t.. 0 500 C00 1500 2000 2500 30C0 Time (s) rigure 39. PCRV mcss flow rcle f or the duol SGTR. J^ d i' 51 t

I i 1 I i i i 1 30 I \\ a A *CT LEG 50 ] 0 B HOT LEG 4 m m D 6 6 40 3 20 v o i .5 l .5 l , 30 -o r 3 -s e o i o u i / 1 o Ih) &u 7n "J 10 - t t t g _3 3 t U [ P 10 j l i l \\\\ d / \\ O' 0 m C S00 CCO 1500 2000 2500 3C00 4 Tim. (s) rigure 40. l-ot leg subceoling 'or the dwet 50T3 with 80RV. i 14 i 6 lF500 A UNAFFECTED STEAW GENERATOR i

i 4

O AFFECTED STEAu CENERATOR t i 12 ' [ O-T ^ I b 10 *- ~4C0 0 ~ i 0 e i I e } 3 '- i a #. 8 '300 d ~ 1 1 .t. I, 4 e e i c d 3 l d 3 i 7 6 >~ I.., '.T l 200 J q s.j J 4 G~ l I.,fil! 4' 1 A i 'i 1 10 0 4 I 2' O Soo 10C0 1500 2000 2500 30C0 Time (s) i Figure 4L Stoom generof or liquid levels f or ite duel SGTR wirp PCRV. i I 52

.~. - _ _ = _ _ = _ _ _ - - _ _ _, _ i ~ 4 CONCLUSIONS Conclusions derived f rom the ANO-2 SGTR analyses are presented below. 1. The reactor cperators have adequate control over RCS pressure, temperature, and level in the event of a single or dual SGTR. j Analysis of RELAP5 calculations indicated that the operators l could successfully mitigate the effects of SGTRs. 2. The installation of a PORV in ANO-2 would provide little benefit for the mitigation of SGTRs since the existing APS system is adequate when feedwater is available. i The RCS can be successfully depressurized by either APS or a PORV. The results presented in Sections 3.1 and 3.2 show that the calculated thermal-nydraulic response of the plant was similar whether depressurization was initiated by the APS or a .I 1 PORV. l 3. The operation of HPSI significantly affected RCS cooldown. RCS fluid temperatures increased significantly when HPSI was terminated, as illustrated by Figure 3. The termination or throttling of HPSI would require an increase in the flow through the ADV to maintain a given cooldown rate. 4. Cold leg temperatures should be monitored closely after a steam generator is isolated. ) In the calculations, the affected steam generator acted as a heat source after it was isolated causing the cold leg temperatures to exceed the hot leg temperatures in the affected loop. Consequently, the minimum loop subcooling, a key parameter during cooldown, generally occurred in the cold legs of the affected inop after the affected steam generator was isolated. 1 53

5. REFERENCES 1. V. H. Ransom et al., RELAP5/M001.5: Models. Developmental , Assessment, and itser Information, October 1982, EGG-NSMD-6035. 2. Arkansas Power and Light Company, Arkansas Nuclear One. Unit 2, License Aoplication FSAR, Maren 1974 Docket-50368-63. 3. American Nuclear Society Proposed Standard, " Decay Energy Release Rates Following Shutdown of Uranium-Fueled Thermal Reactors," Approved by Subcommittee ANS 5.1, ANS Standards Committee, October 1971, Revised October 1973. 4. Baltimore Gas and Electric Company, Calvert Cliffs Nuclear Power Plant, Units 1 and 2, Preliminary Safety Analysis Report, January 1971, Docket-50317-21. 4 5. United States Nuclear Regulatory Commission, Arkansas Nuclear One Unit 2 Technical Specification, Appendix "A" to License No. NPF-6, July 1978, NUREG-0336. 6. Combustion Engineering, SGTR Guideline, CEN-152 Revision 01. ( 54

APPENDIX A CODE UPDATES 1 1 l l l l 55

APPENDIX A CODE UPDATES The code updates which were added to RELAP/M001.5 cycle 25 for the steam generator tube rupture calculations are listed in Table A-1. e 56

l s

• e i

r TABLE A-1 Code Updates

• C C.iP A 6 E D ic Itt i: s.i E G O IR
• 1 DINT JALZ32'

'I MJJTNT.d3 --1 F4 4J4t S447-,4T.--4.-.4 NG. - 0T SF-, rriv4 T4V4.440,- - -

• HTD Tdd( J +L I

.LT. 5 ATT (K1)GO T C SJJO

• I M03THT.296 0

OHcCA F0h hiRY LOW VLID C ONDE N5 ATIJN O A55 .-c - ( i0 0 e.1:..-2 ) 6 0-T C-eive-- -

• 1 T3AA22b.29 6190 C3N T IN UE
• 1 J AT X 21w. l(,

-- i+44-.tT. -J44G-T-G-14 I I e 57

f e 'O " U.S. NUCLEAR REGULATORY COMMISSION EGG-NTAP-6226 BIBLIOGRAPHIC DATA SHEET 4 TITLE AND SUBTITLE

2. (Leave elai 41

+ l Analyses of Steam Generator Tube Ruptures in Arkansss Nuclear One Unit 2 3 RECIPIENT S ACCESSION uo. 7 AUT HO54lSt S. DATE REPORT COMPLE TED C. B. Davis, J. E. Blakeley appy) ivl983 N 9 PE RF ORMING ORGANIZATION N AME AND M AILING ADORESS (IncIvor 20 Codel DATE REPORT ISSUED MONTM l YEAR April 1983 EG&G Idaho, Inc. ,t,,,,,,,,, Idaho Falls, ID 83415

8. (Leave Sank) 12 SPONSORING ORGANIZATION N AME AND MAILING ADDRESS Itaciude 2,0 Codel o PROJECT / TASK / WORK UNIT NO.

Division of Systems Integration Office of Nuclear Regulatory Research ii. riN NO. U.S. Nuclear Regulatory Commission Washington, DC 20555 A6354 13 TYPE OF REPORT PE RIOD COVE RE D (inesus,ve carrs/ j i 15 SUPPLEMENTARY NOTES

14. (Leave dashi 16 ABSTH ACT (200 *cwas or sess)

Analyses of hypothetical steam generator tube ruptures (SGTRs) in Arkansas Nuclear One Unit 2, a Combustion Engineering pressurized water reactor, were ( performed using the RELAPS computer code. Hypothetical accidents initiated by single and dual SGTRs were analyzed. Calculations were performed to com-pare the effectiveness of the auxiliary pressurizer spray versus a pressurizer power-operated relief valve in depressurizing the reactor coolant system. The analyses were performed to determine if the reactor operators could control the thermal-hydraulic response of the plant during a SGTR. 17 KE Y WORDS AND DOCUMENT AN ALYSIS 3 7a DESCRiPTORS 17b :DENTIFIERS OPEN ENDED TERMS 18 AV AIL AB'LITY ST ATEMENT 19 SE CURITY CLASS ITM,3 report 1 21 NO OF PAGES Unclassified Unlimited 20 pCuRiry CtASS <rn.s, ages

22. PR.CE dnclassified S

N RC F ORY 335 sin get}}