ML20148J175

Study of Effects of Steam Jets & Water Jets on Emergency Sump. Concludes That Air Entrainment Due to Break at Steam Generator Nozzle Will Not Adversely Affect Functional Capability of Sump

ML20148J175
Person / Time
Site:	Sequoyah
Issue date:	09/30/1978
From:	TENNESSEE VALLEY AUTHORITY
To:
Shared Package
ML20148J126	List: ML20148J121 ML20148J175 ML20148J181 ML20148J188
References
NUDOCS 7811150176
Download: ML20148J175 (14)

Text

.

o.

n..

f 4

'A STUDY OF THE EFFECTS OF STEAM JETS-AND WATER JETS ON THE EMERGENCY SUMP AT SEQUOYAH NUCLEAR PLANT UNITS 1 AND 2 l

l i

l Tennessee Valley Authority )

i September, 1978 f

-l l

l 1

i 1

'/811150 h(o -

1

)

1 1

)

O $

CONTENTS Page A. Introduction . ... . . . . . . . . . . -. . . .. . . .. . .. 1 B. Phynical Arrangement . . . . . . . ' . . . . . . . . . .. .. 1 C. Steam Jet ... .

. . .. . . . . . . . . . . . . . 2 D. Air Entrainment .. . . . . . . . . . . . . . 3 j (1) Physical Considerations . . . . . . . .. .. .. . . 3 (2) Experimental Considerations ' . . . . . . . .. . 3 l

'4 (a). Water jet velocity . . . . . . . . . . .. 4 (b) . Nozzle size '. . .. . . .. . . . . . . -. 5 (c). Penetration. depth and bubble size .

. . . 5 (d) Scaling of penetration 4 l

depth and bubble size . .. . . . .. .. . .. 6 l

(e) Experimental observations in scaled model . . . . . . .. . . .. . . 6 1

(3) Conclusion . . . .- .. . . . .. . . .. .. .. 6 l l

i 1

l I

-i i

1 d.

E I g o -.

'1 1

1

/

e

/ '

. 9,

. ,f -

_3 .. !

.4

,.v!

e . . .

l

~

. l A. INTRODUCTION In re'sponse to discussions with the Nuclear Regulatory Commission (NRC), TVALhas completed an evaluation of the effects of steam and~ water jets on the operation of the emergency sump at the Sequoyah Nuclear' Plant during the recirculation mode of ECCS operation. The primary objectives were to confirm that:- (1) steam jets from pipe breaks will be condensed prior to reaching the sump screen,. and- (2) 'any air entrained in the water by a. jet impinging on-the water surface will not be drawn into the sump. . ' The overall objective was to assure that the performance of the pumps used to supply cooling water to the reactor core'and primary containment will not be degraded.

Two areas of potentia 1Lconcern were identified:

1. A steam jet of sufficient length could allow steam to enter the area under the sump cover and condense within that area, thereby causing pressure pulses in a confined region.

2. A pipe break above the minimum sump water level at the time of recirculation from the sump could entrain air in the effluent stream and transport the air to a location where it would be drawn into the. sump.

The effects on sump operation from the phenomenon discussed above have been' evaluated to show that the Sequoyah emergency sump will-perform as required..

B. PHYSICAL ARRANGEMENT The Sequoyah ' emergency sump is located as shown in Figure 1.

There are three Reactor Coolant System (RCS) pipe breaks in the vicinity of the sump which could potentially affect its operation during recirculation. Figure 2 provides a view of the break locations. The break locations were determined by Westinghouse as discussed in the Sequoyah FSAR, section 5.2, and Westinghouse Topical Report WCAP-8172-A HPipe Breaks for LOCA Analysis of the Westinghouse Primary Coolant Loop," January, 1975. The breaks at the reactor coolaut pump nozzle and at the crossover leg closure weld are under water when the Residual Heat Removal (RHR) and Containment spray pump suction is switched from the refueling water, storage tank to the sump. Since these two breaks are under water there :will 'not be any air entrainment, only a steam jet.

The effects-of steam jets have been studied for these breaks.

The break at the steam generator nozzle is above the post LOCA water. level during recirculation from the sump. The effects of (a) steam. jet,.and (b) air entrainment in the effluent from the break have been considered for this break. The effect of a steam

. jet for each of. these three breaks is discussed first.

pi s

C. STEAM JET All three breaks in the vicinity of the sump have the potential.

for a steam jet interaction with.the sump. The break at the crossover. leg closure weld was chosen for study.because its close proximity to the sump and orientation give it the greatest potential:for interaction with the sump. The crossover leg weld is approximately-26 inches from the sump screen as shown in Figure 3. The break at_the pump nozzle weld'is over twice as far away. The break at.the steam generator is~over 15 feet from.the )

sump. . Additionally the breaks. at the punp and steam generator j nozzles are oriented in such a manner that jets would not be directed tt.ard the sump (Figure 2) .

1 The mass and energy releases _ used to evaluate the jet effects are  !

based on the LOCA analysis releases provided in the,Sequoyah  !

FSAR, -Table 6. 2-11. (and duplicated in Table 1, attached) . The value of sh-am release rate chosen was the maximum rate after 900 seconds whia h is the earliest time that suction is taken from the sump. The flow rates out of the break are based on runout flow f rom the Ria pumps.

~

For a break in the crossover leg closure weld, and based on the conditions listed above, all the steam leaving the break will' condense prior to reaching' the sump cover (Figure 3). _The situation existing for this break at the start of recirculation from the sump is essentially that of a steam source (the broken l

pipe) venting to a suppression pool (the water collected in the lower compa:.tment) . There is considerable experimental evidence which shows that condensation of steam in a large pool of water is rapid and complete. The majority of this experimental work

-was performed in relation to the design of pressure suppression containments for boiling water reactors. The information obtained was conservatively applied to the sequoyah sump.

Examplessof the conservatism are:

1. The steam condensation.. tests were performed with mass and energyL release rates exceeding _ that from the sequoyah pipe -

break.

2. The steam _and subcooled water exiting the break at Sequoyah are assumed ~ not to interact with each other within the . broken

pipe -(Table 1) ,- therefore no condensation occurs until the steam exits the pipe.-

3. The-maximum bulk water temperature of the sequoyah sump water at the time of. recirculation is 1600.F.. . (See FSAR Figure

6. 2-31) . . There is therefore .a minimum' temperature difference of 520 F between the steam and the water. This is.within the bound of the. experimental conditions. Experiments.have shown

. that stationary condensation- during- high steam-flow densities runs quietly and completely up to bulk water temperatures of over 1758 F.

t * .

L

4 I The bulk of-the literature TVA reviewed deals with the jet exiting from the end of a pipe which yields a conical shaped jet (Figure 4) . - The jet from the break that may exist at sequoych will appear as a. disc. due to pipe restraints which physically restrict pipe. motion after a break. Figure 5 illustrates that shape of the. jet for. a crossover ~1eg closure weld break. The -

area of a cone! required to condense a' jet based on the sequoyah.

mass and energy relenses was determined. This area was then used ,

to determine the length of a jet based on the sequoyah I c onfiguration. The jet length was found to be much less than one I foot-for complete condensation to occur. Since a minimum condensation length of'26 inches is available, it is concluded that steam jets from pipe breaks will not affect the operation of the sump in a deleterious manner during the recirculation mode. j D. AIR EtFTRAINMEIR Underwater jets will not entrain air. However, the break at the steam generator outlet nozzle is above the post-LOCA water level.  !

The effluent from the break will entrain air and carry the air  !

into the water.

TVA has considered the location of the sump with respect to the orientation of the break at the steam generator nozzle and has concluded that air will not enter the sump. Additionally, tests were performed to show that even if air bubbles were introduced  ;

in the immediate vicinity of the sump they would not be drawn I into the sump.  ;

(1) Physical Considerations t

The break at the steam generator nozzle nearest to the sump is oriented so that the water jet is directed away from the sump (Figure 2) . Thus the air entrained by the jet will have a horizontal velocity component directed away from the sump and will rise to the surface without being drawn into the sump. Additionally any water jet from this break will impinge upon platforms and supports under the steam generator. This impingement will break up the jet and greatly reduce the penetration depth of the air. bubbles. Based on these considerations alone, there is no~ air entrainment problem with the Sequoyah sump.

(2). Experimental Considerations Tests were performed at TVA's Norris Engineering Laboratory to provide additional credence to the conclusion drawn above. Several tasks were performed in sequence:

(a) ' Calculate the maximum velocity of the water jet from the break.

f f

t (b) Determine. an equivalent diameter for a test nozzle.

(c) observe the_ depth.of-air penetration for a vertical water jet.and observe the air bubble size in a full scale test.

(d) Scale the bubble size and penetratior. depth and perform a test in the 1/4 scale model at the TVA

.Norris Engineering Laboratory.

(e) Visually determine if any bubbles would be drawn into the sump.

The details are as follows:

(a) Water iet velocity The water jet velocity out of the ascumed break is based on the separated flow model as presented by G. B. Wallist.. The bases for the use of the separated flow model are as follown:

1. The primary basis is that there is no interaction between the water and the sf.eam inside the crossover leg until both fluids are in the immediate vicinity of the break. This is the case for this break because only steam comes from the steam generator side of the break while subcooled water comes from the reactor coolant pump side. The effects of gravity, buoyancy, and pipe friction on the water flow with respect to the steam flow are therefcre non-existent away from the bronk itself. The flow rate of each phase is then purely a function of the pressure drop and inertia terms. (Thic would not be the case for a steam-water flow in a single pipe where the flow direction of both fluids in the same. )

2. The -volun7 in which the steam and water can interact is relatively small in the immediate vicinity of the break. The length over which the steam and water are accelerated out of the break is also very short. The acceleration is therefore due to the pressure drop and inertia that are already present in the individual fluid streams. The short time'that the wat'er 1"One Dimensional' Two-Phase Flow" by Graham D. Wallis, McGraw Hill, 1969.

_q_

5\

4 and. steam: masses are-in the vicinity of the break reduces significantly the interaction Detween the steam and water. As a result, the water, being.aubcooled by a considerable amount (340 to 600 F), will not change phase as it exits the break.

Based on the model just discussed the maximum ,

velocity of the water exiting the break is j 35 feet per second. 1 (b) Nozzle Size A 1-inch 1.d. nozzle was determined to be conservative and representative of the situation, l The basis for- this selection depends on several I considerations. The steam generator nozzle break has a break area of 317 in z. This in conjunction with a constant water flow rate of 1250 lbm/sec (Table 1) , would lead to a water jet' speed of 9.4 ft/sec, if no steam was considered and the entire 1 break area was available for water flow. However, '

the water jet speed was determined earlier to be 35 ft/sec due to the presence of steam. The break opening of the steam generator nozzle break is 2.6 inches-as determined from the break size, pipe diameter, and-pipe deflection. To provide an equivalent break opening (and hydraulic diamameter), the nozzle used in the experiment should, by application of the continuity equation, have a diameter of 2.6 x 9.4 = 0.7 inch i 35 A 1 inch diameter nozzle was selected for the full scale penetration test.

(c) Penetration Depth and Bubble Size Full scale tests were performed to determine penetration depth and size of the air bubbles. The test was performed.in a water tank 7 feet in diameter and 45 feet deep. The nozzle simulating the break was placed 2.5 inches from the water I surface. 'rhe jet was directed vertically downward to'be conservative. Air bubbles were photographed through side windows at various elevations. A l linear scale was installed inside the tank to provide- accurate indication of the depth of

' penetration of the air bubbles. Figure 6 shows the ,

effect of jet velocity on air penetration depth. l

.The test data show that for a water. jet velocity of 4

S l 1

_ _ _ _ _ _ _ ~_ --

35 feet per second the air penetration depth is less than 8 feet. The ' air bubbles were found to be uniform with the shape of an ' oblate spheroid (blimp) . . The dimensions of the major and minor axis were 0.20 and 0.15 inches,'respectively. The l uniform size of the bubbles was expected, which is in agreement with the work of others.

(d) -Scaling of Penetration Depth and Bubble Size The bubble size and depth of penetration were scaled for use in the 1/4 linear scale sump model test. The scaled bubbles were 0.8 mm in diameter, based on terminal rise velocity as presented by i Wallis. The penetration depth was scaled linearly which is consistent with the 1/4-scale model at Norris.

(e) Experimental Observation in Scaled Model l With the help of a calibrated, porous diffuser, scaled bubbles were introduced at various locations around the sump in the model, irrespective of the actual _ orientation of the jet from the break. The bubbles were released at various water depths.

With the model. sump water flowing at the scaled velocity as well as at the unrealistically high prototypical velocity, it was observed that the bubbles were not drawn into the sump even if the bubbles were released at floor level (13.2 feet full scale) and within 6 inches (2 feet full scale) ,

of the sump screen.

(3) Conclusion It is concluded that based on the considerations in parts 1 and 2 of this section, air entrainment due to a break at' the steam generator nozzle will not adversely affect the functional capability of the sequoyah sump.

One additional area with the potential for introducing air into the vicinity of the sump are the three ice condenser. drains located near the sump. The exits from these three drains will be rerouted so that their discharges will flow down the crane wall. This change will eliminate these potential sources of air entrainment.

=_.

l l

l 1

TABLE 1 POST REFLOOD MASS AND ENERGY RELEASE INFORMATION (MAXIMUM SAFETY INJECTION)

TIME STEAM FLOW WATER FLOW MASS ENERGY MASS ENERGY SECONDS lb./sec.. 103 Btu /sec lb./sec. 103 Btu /sec 155 134 162 1238 196

-200 126 153 1246 196 300 123 149 1249 154 400 122 147 1250 191 500 122 146 1250 187 i 600 122 146 1250 184 1 700 122 145 1250 181 l 800 123 146 1249 182 l Recirculation Begins 900 124 145 1248 178 930 125* 146 1247 182 935 87.3 101 1285 189 1000 85.9 99.3 .1286 187 1100 83.9 97.0 1288 187 1200 82.1 95.0 1290 186 1400 79.1 91.4 1293 184 1570 .76.8 88.9 1295 185 l

• Value used to determine steam jet length.  ;

4 d

J l

i l

l 1

-7~ ,

i s

l m a

l 270 l

l

... ' . . ' . ' . 'Q ORAIN HOLESw , : '- , 3. ) )

,.l--- sh :,

'p- .. HECIRCULATION PUMP 1

• g . s .

I

. . ,- . CONTAINMENT SUMP

~-

... . -- q

_ _._ __ J , _

\  %,+

- A~1 ,! ' ',-

STEAM GEN l LOOP ... LOOP -

l 3 , *. ' - 4

f. p ,

l[ 7 E,'. b -

AIR CLEANUP j.

_ .\

I

'. UNITS

'. '; ',- j 18 0 '-

'^ -

+ -

O l

',. ,l .. .-

CRDM COOLING

.- 4

'($

r' . UNITS LOOP LOOP -

2 i .

PRESSURIZER '

N.

'() IEF TANK [ -

h /

MODEL LIMITS REACTOR . , ,,...

SHIELD WALL

.,....3..'

CRANE WALL PRESSURIZER 90' FIGURE i MODEL LIMITS (SEQUOYAH UNIT I) a

1 l

i

[  :

l l

l STEAM

\ GENERATOR /

1 i

I REACTOR ,

COOLANT PUMP J MINIMUM WATER LEVEL STEAM GENERATOR l

! DURING SUMP RECIRCULATION OUTLET N0ZZLE I

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

LE

'.- N A CROSSOVER LEG

.i e CLOSURE WELO ,s

\

_.a , f CONTAINMENT FLOOR SUMP FIGURE 2 BRE AK LOCATIONS 8. ORIE NTATION

b O

O e

I i

l g CRO M t.EG p c8AME

n.. .

+

/ &

..,,c c -

\ r c

,,)

.. 1 @gets

.' 0_,-* ly,

' c p *,.:. cp.. g~

p\GOBE yggy of Su cBOSSU i

t I

g ep FIGURE 4 STEAM JET SHAPE FROM A SINGLE LINE o - ~

0, -== "

, == # -

L

.

FIGURE 5 STEAM JET SHAPE FROM A RESTRAINED PIPE l

p

/ \ ,

1 I

A'(B c.

l  !

\ l

\. )/ '

/" ~% l r

\

/ 2 \ .

/ \

l l r 1

i

( \

U /

/

\ /

's _,s '

\

WATER ^. ' -

SURFACE O C-JS U

oc 2 3

oc Test for 1" diameter pipe nozzle 3 4 2

E O '

g 6 m

5 5 N_

m 8 H

ur N

m e -

E 10 o

E -

t a 12 CONTAINMENT FLOOR I4 o 10 20 30 40 50 NOZZLE EXIT VELOCITY (ft/sec)

FIGURE 6 AIR PENETRATION DEPTH i