ORNL-TM-4122.txt

   

| ORNL-TM-4122

DEVELOPMENT OF A
VENTURI TYPE BUBBLE GENERATOR FOR USE
IN THE MOLTEN-SALT REACTOR XENON
REMOVAL SYSTEM

C. H. Gabbard

MASTER

 

 

 

 

OPERATED BY UNION CARBIDE CORPORATION = FOR THE U.S. ATOMIC ENERGY COMMISSION

 
 

 

This report was prepared as an account of work sponsored by the United
States Government., Neither the United States nor the United States Atomic
Energy Commission, nor any of their employees, nor any of their contractors,
subcontractors, or their employees, makes any warranty, express or implied, or
assumes any legal liability or responsibility for the accuracy, completeness or
usefulness of any information, apparatus, product or process disclosed, or
represents that its use would not infringe privately owned rights,

 

 
ORNL-TM-4122

Contract No. W-Th0S5-eng-26

Reactor Division

DEVELOPMENT OF A VENTURI TYPE BUBBLE GENERATOR
FOR USE IN THE MOLTEN-SALT REACTOR XENON REMOVAL SYSTEM

C. H. Gabbard

Molten-Salt Reactor Program

 

 

 

 

 

 

NOTICE NOTICE
::tllfr:ealf;t wiznti?sa;f; rrxr-ljatmnfloffa preliminary This report was prepared as an account of . work
at the originatinp i:stallafioflali i 011; 1n_tfrna1 1se sponsored by the United States Government, Neither
vision or correc%ion and theré-f r]sdsu 1e°tto re- the United States nor the United States Atomic Energy
sent a final report. It is passf;é)tg thoes noi ;‘Epre- Commission, nor any of their employees, nor any of
confidence and should not be a-bstract?a;?)i}t? e[:;lin their contractors, subcontractors, or their employees,
disclosed without the a, roval‘ f th igri l:? r makes any warranty, express or implied, or assumes any
installation or DTI Extpps' 5 k e inating legal liability or responsibility for the accuracy, com-
ension, Oak Ridge. pleteness or usefulness of any information, apparatus,
product or process disclosed, or represents that its use
would not infringe privately owned rights.

 

 

December 1972

OAK RIDGE NATIONAL LABORATORY
Cak Ridge, Tennessee 37830
operated by

UNION CARRIDE CORPORATION .
for the ASTE R
U.S, ATOMIC ENERGY COMMISSION
SYSTOTRUTION OF THIS DOCUNENT IS LH&H&E%

it Y Snatenpk
15 sna MEG Guniraciors |

  
iii

TABLE OF CONTENTS

ABSTRACT
I. INTRODUCTION
ITI. BUBBLE GENERATOR DESIGN
IIT. OPERATING CHARACTERISTICS AND TEST RESULTS
IIT.~-1. Bubble Size
IIT.-2. Gas Injection Pressure Characteristics
IV. CONCLUSIONS AND RECOMMENDATIONS
V. ACKNOWLEDGEMENT
NOMENCLATURE
REFERENCES

APPENDIX

14
27
28
29
30

31
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

lo'

11.

12.

LIST OF FIGURES

Bubble Generator Design Configurations which were
Given Reduced Scale Evaluation Tests

Bubble Generator Design for the Gas Systems
Technology Facility

Bubble Size Produced by 2.1 in., Threoat Diameter
Bubble Generator as a Function of Liquid Flow Rate

Bubble Size Produced by GSTF Bubble Generator
as a Function of Surface Tension

Surface Tension as a Function of Sodium Oleate
Concentration for Laboratory Batch Samples and
for Loop Samples

Bubble Size Correlation for GSTF Design Bubble
Generator

Simplified Flow Diagram of the Gas Systems Technology
Facility

Gas Injection Pressure and Overall Pressure Drop
of Prototype Bubble Generator as a Function
of Gas Flow Reate

Geometry of Bubble Generator Used in Analysis of
Gas Injection Pressure and Overall Head Loss

Pressure Drop of Gas Feed Passages as a Function
of Gas Flow Reate

Pressure Drop Across the Gas Plume Interface
as a Function of Throat Void Fraction

Correlation of Plume AH (H6) to Throat Liguid
Velocity and Throat Void Fraction

10

11

16

16

17

19

22

2k

25
DEVELOPMENT OF A VENTURI TYPE BUEBLE GENERATOR
FOR USE IN THE MCLTEN-SALT REACTOR XENCN REMOVAL SYSTEM

C. H. Gabbard

ABSTRACT

A venturi type bubble generator was developed for appli-

cation in the xenon removal system proposed for a molten-salt

_breeder reactor. Gas injected into the high velocity liquid
‘at the“venturi throst is formed into bubbles by the fluid
turbulence in the diffuser cone. Tests were conducted using
aqueous solutions tc determine the various pressure drops of
the bubble generator as a function of liquid and gas flow
rates and to determine the bubble diameter produced.
Empirical relaticnsktips were developed which could be used in
combination with the more conventional fluid flow equations
to predict the overell head loss and the gas injection pres-
sure of the bubble generator. A dimensionless correlation
for predicting the tubble diameter was developed for bubble
generators of similsr geometry.

Keywords: Bubble Generator, Bubbles, Bubble Size, Gas
Injection, Fused Salts, MSBE, MSBR, Performance, Xenon, Fluid
Plow. o —

 

I, INTRODUCTION

In a nuclear reactor operating in the thermal energy range, the con-
tinuous removal of the geseous fission product poison xenon-135 is neces-
sary to obtain a breeding ratio greater than 1.0. In a molten-salt
breeder reactor (MSBR), the xenon-135 circulates in solution with the
molten flucride fuel salt. A proposed method of removing this xenon is
to continuously inject helium bubbles into the salt stream to a gas
volume fraction of 0.2 to 1.0 percent at the reactor core midplane. The
xenon-135 would transfer by turbulent diffusion and would be stripped
from the salt when the bubbles were removed. Calculations indicate that,
even with this low gas volume fraction, adequate xenon-135 removal would
be cbtained by stripping the bubbles from a bypass stream which is about
10 percent of the main salt flow. Little advantage would be gained by
stripping larger flows. A more complete discussion of xenon removal from

a MSBR by this method is presented in Reference (1).
This report describes the design, development, and operating char~
acteristics of the bubble generator proposed for use in a 150 MW(t)
molten-salt breeder experiment (MSBE).(z) A full scale Plexiglas model
of this bubble generator was stfiaied in a test facility using water,
glycerin-water mixtures, and CaCl2 aqueous solutions. A prototype model
of Hastelloy "N" will be further evaluated with molten salt as part of

the test program of the Gas System Technology Facility (GSTF).(B)

ITI. BUBBLE GENERATOR DESIGN

The uwltimate goal of the development program was to obtain informa-
tion which could be used to design a full scale bubble generator which
could be tested in the GSTF using molten salt., Several design criteria
that have evclved during the development of the bubble generator are
listed in Table I. Devices requiring auxiliary power or having moving
parts were considered originally but were rejected as being unnecessarily
complex for a high-temperature molten-salt system. Fluid powered devices
basically resembling flow venturi appeared to satisfy the criteria and
three configurations were selected for continued development. The dif-
ferent configurations, shown in Figure 1, are variations in the method of
forming the high velocity throat region. Helium injected into the high
velocity salt stream at the throat forms small bubbles as a result cf the
fluid turbulence in the diffuser section.

Reduced scale tests were performed on these three configurations and
each performed satisfactorily. Initial testing of the "teardrop" design
indicated that the resulting bubble size was about one-fourth of the salt
flow gap over the range of liquid flows tested.(u) Consequently, a flow
passage of 0,080 in. would be required to produce 0.020 in. diameter
bubbles. The "multivane' design was an extension of this principle to
provide a more uniform bubble distributicon over larger pipe sizes and to
avoid the large diameter that would have been required in a full scale
teardrop design with a 0.080 in. annulus. Tests of a single vane prototype
revealed a flow oscillation around the trailing edge of the vane. In
addition, the gas distribution along the width of the vane and btetween
the flow passages on either side of the vane was difficult to control.
Reduced scale tests on the "venturi" design were performed using 3/k in.

and 1 1/2 in. pipe size commercial jet pumps that were modified to more
Table I
Bubble Generator Criteria
The bubble generator shculd be sized for application in the
MSBE.
Nominal salt flow rate = 500 gpm,
Gas flow rate = ¢ - 0,65 scfin helium.
The generated bubble diameter should be 0.020 in. or less.

The gas bubbles should be uniformly dispersed in the flowing
salt stream.

The bubble generator should be simple, reliable, and
maintenance-~-free.

The bubble generator should operate from pressure drop inherent
in the overall system design and should not require a gas com-
pressure for the injection cof gas.
ORNL-DWG 71-10220

GAS FEED-EH
VANES |

 

 

 

H
O

 

 

 

 

 

RECTANGULAR .
CROSS SECTION —g[

PN
FEEDE;}}

 

 

e

FLow) |

 

 

 

 

 

 

 

 

 

 

 

 

MULTIVANE VENTURI " TEAR DROP

FIGURE 1

BUBBLE GENERATOR DESIGN CONFIGURATIONS WHICH WERE
GIVEN REDUCED SCALE EVALUATION TESTS
closely provide a venturi geometry. These tests showed that well dis-
tributed bubbles of abcut the desired size could be produced.

Because of its simplicity and ability to meet the other requirements,
the venturi design was selected for full scale development. Figure 2 shows
the final design chosen for further testing with molten salts at high tem-
perature in the GSTF, This design is a modified venturi with the 2.10 in.
diameter throat stepped to 2.18 in. at the gas feed holes. The gas is
injected through 18 - 1/8 in. diameter radial holes into the high velocity
region at the venturi throat. An annular gas cavity forms between the wall
of the bubble generator and the flowing liquid in the 2.18 in. diameter
cylindrical mixing chamber. The length of this cavity depends on the gas
flow rate, and at full gas flow the cavity extends into the 15° diffuser
section. The actual bubble formation occurs in the fluid turbulence in

the entry of the diffuser cone.
ITI. OPERATING CHARACTERISTICS AND TEST RESULTS

A full scale model of the proposed bubble generator with a 2.1 in.
diameter throat and with 4 in. diameter inlet and outlet piping connec-
tions was fabricated of Plexiglas for complete testing and evaluation,
Tests on this bubble generator were conducted to determine the bubble
size produced, various pressure drops, and general operating characteris-
tics. The tests were run with demineralized water, L41.5 wt percent

glycerin in water, and 31 wt percent CaCl,. agueous solution. The glycerin-

2

water mixture and the Ca012 solutipn have the same kinematic viscosity as
fuel §alt and provided dynamic similarity. Tests were also conducted with
up to about 200 ppm n-butyl alcochol or sodium oleate added to demineralized
water, The n-butyl alcohol, a surfactant; stabilized small bubbles and
inhibited coalescence but had little effect on the density, viscosity,

or surface tension of the bulk fluid. The sodium cleate, also a surfactant,
decreased the surface tension by about a factor of two and inhibited bub-

ble coalescence, but did not alter the density or viscosity of the bulk

fluid.
ORNL-DWG 72-9609
_-GAS INJECTION LINE
" Y~in, OD x 0072-1in.
WALL TUBING

18 Y-in-diam
GAS INJECTION HOLES —._

 

5047 in.
diam

 

 

 

 

 

ANNULAR GAS D|STR|BUT|ON/‘/ : |
CHANNEL 3/4 in. x 5/8 in, — n

- 4'%g in—

I
-

 

— 9% in. —

FIGURE 2

BUBBLE GENERATOR DESIGN FOR THE
GAS SYSTEMS TECHNOLOGY FACILITY
I1IT.,-1. Bubble Size

I1T1.-1.1 Test Condition

In the proposed xenon removal system, helium bubbles are to be
injected and removed in a 10 percent bypass loop. The bubbles on the
average are expected to circulate several times around the primary cir-
cuit of the reactor before being processed in the side stream. During
this circulation, the bubbles will be affected by solution and dissclution
as they pass through different regions of pressure and temperature, and
by breakup and ccalescence as they pass through high and low shear regions
(e.g., the pump). Consequently, the circulating bubble size is likely to
be controlled by the system dynamics rather than by the bubble generator
itself. However the size generation characteristics of the bubble gen-
erator should be of general interest for other systems and for possible
unanticipated modes of operation, such as full flow gas injection and
removal, In addition, the size produced may serve as an "initializing"
condition for monitoring changes as the bubbles pass through the system.
Consequently, some analysis and some limited tests were made to obtain
an indication of the bubble size produced by the bubble generator as it
is affected by flow and fluid properties. Flow rate was varied from
200 gpm to 550 gpm and surface tension was varied from 72 dynes/cm to
30 dynes/em by adding different amounts of sodium oleate. An antifcaming
agent, G.E. Silicone Emulsion AF-T2, was also added at concentrations of

10 percent of the sodium oleate.

III1.-1.2 DBubble Size Measurements

The bubble size distributions produced by the bubble generator were
determined by taking still photographs at the discharge of the diffuser
cone. A conventional studio camera with a 12 in., focal length lens was
used to take the photographs on 4x5 Polaroid film. A strobe light with a
1/30,000 second duration was used to "stop" the bubble motion and to pro-
vide back lighting.

The photographs, which were about actual size, were enlarged to
obtain a total magnification of 8. Enlargements to greater magnification
resulted in a loss of resolution. The bubble size distributions for each

condition were determined by scaling bubble sizes directly from the
enlargements. The diameters were measured by comparison with a plastic

template having drilled holes ranging from 1/32 to 3/4 in. in increments
of 1/32 in. A volume averaged bubble diameter as defined below was cal-
culated for each distribution:

1/3
L(n, 4,

 

where: n. is the number of bubbles of a given diameter,

di,per unit area of the photograph.
The resolution of the photographs was adequate to measure bubble diameters
in the 0.008 in. range (1/16 in. on the enlargement), but no bubbles
could be identified in the 0.004 in. diameter range. The results of these
tests are shown on Figures 3 and 4.

Figure 3 shows the volume average bubble diameter produced by two
bubble generator designs plotted as a function of liquid flow rate at
several values of surface tension. The data are compared with a slcpe of
-0.8 power dependence discussed in greater detail later in this report.
There was a high degree of scatter in some of the sets of data at constant
surface tension. Consequently, only selected data sets having low scatter
are shown on fihe plot. Although there were differences in the slope of
the various lines, the data tend to support a -0.8 power dependence.
Similar data taken previously alsc support a -0.8 power, and none of the
data have suggested a slope significantly different from -0.8.

Figure L is a plot of the bubble diameter as a function of surface
tension at three flow rates. The measured surface tension data from
locp samples taken during the course of this experiment were scattered
and did not agree with the data from previous laboratory scale samples
which were in general agreement with the sodium cleate supplier's litera-
ture. The values of surface tension used in Figure L were obtained from
the calculated concentrations in the test loop and the surface tension
vs concentration data from the laboratory semples as shown on Figure 5.
The measured surface tension data from the loop samples are also shown
on Figure 5. The discrepancy between these is not fully understocd.

However, the actual circulating concentration of sodium cleate could change
BUBBLE DIAMETER , VOLUME AVERAGED (in.)

ORNL-DWG 73—1541

 

 

GSTF STEPPED BORE DESIGN

DEMINERALIZED WATER
10 ppm SODIUM OLEATE

O
®
A 20 ppm SODIUM OLEATE
A 75 ppm SODIUM OLEATE

 

 

 

 

 

 

 

SMOOTH BORE DESIGN
V DEMINERALIZED WATER

 

v

0.050 |

o

 

 

v

 

 

> @1+— 0

N

v
0.020 \g
. TN

v

 

 

 

0.010

 

 

 

 

 

 

>

 

 

 

 

FLOW RATE (gpm)

FIGURE 3

BUBBLE SIZE PRODUCED BY 2.1 IN. THROAT DIAMETER BUBBLE GENERATOR
AS A FUNCTION OF LIQUID FLOW RATE
BUBBLE DIAMETER, VOLUME AVERAGED (in.)

0.060

0.050

0.040

0.030

0.020

0.010

0.009

ORNL—DWG 731542

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A LIQUID FLOWRATE = 200 gpm
@ LiQUID FLOWRATE = 350 gpm
O LIQUID FLOWRATE = 500 gpm
L’
/ °
/ : |
@
(d,) =c (0/p)0-B A @
O
ei O
A A ®
O
)
A
A O
®
O
O
10 20 30 40 50 60 70 80 90

SURFACE TENSION {dynes/cm)

FIGURE 4

BUBBLE SIZE PRODUCED BY GSTF BUBBLE GENERATOR
AS A FUNCTION OF SURFACE TENSION

100

0T
SURFACE TENSION (dynes/cm)

80

70

60

50

40

30

20

ORNL—-DWG 73-1539

 

 

 

 

® LABORATORY SAMPLES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O O LOOPSAMPLES
o .
\
.\ 5
o O
° L
© q
-\.
\\
e ————
\’
20 | 40 60 80 100 120 140 160 180 200

CONCENTRATION OF SODIUM OLEATE (ppm)

FIGURE 5

SURFACE TENSION AS A FUNCTION OF SODIUM OLEATE CONCENTRATION
FOR LABORATORY BATCH SAMPLES AND FOR LOOP SAMPLES

1T
12

during a given test run because the sodium oleate, being a surfactant,
would be stripped from the circulating loop along with the bubbles. The
concentration in the loop samples could then be less than the calculated
average concentraticn for the entire loop depending on the time the samples
were taken. The bubble photographs were taken immediately after gas flow
was started following -an hour's circulastion without gas flow. This proce-
dure should have provided a concentration of sodium cleate essentially
equal to the.calculated average 8t the time the photographs were taken.

The bubble diameter data of Figure 4 are too scattered to accurately
determine the actual power dependence. However, the data tend to support
a value cf 0.6 as predicted by the theoretical considerations discussed

below and as illustrated on Figure L.

IIT.-1.3 Analysis of Bubble Size Data

The bubbles produced by the bubble generator are apparently formed
in the entrance region of the conical diffuser as a result of fluid tur-
bulence. The following equation has been proposed to predict the size of

gas bubbles produced by fluid turbulence.

- 2
e, 3/5 /5
1{ o €g

(5) to calculate droplet diameters produced

Equation (1) was used by Hinze
by emulsification of one ligquid in another in an isotropic-turbulent

flow field. Assuming turbulent flow in a conduit with conditions such
that the friction factor would be constant, the power dissipation per

(6)

TR

e =k Re (2)
2 2 L *
p D2 Ea

unit volume (e) can be expressed as:

Substituting this relationship for the power dissipation, Equation (1)
gives:

gp D, g "1V, D5 o
2
—2 ¢ 2 (3)

2
2 u H

 
13

The bubble size data presented in Figures 3 and 4 generally confirm a

3/5 power dependence for the surface tension term, but indicate an expon-
ent of -0.8 for the Reynolds Number term rather than -1.2 as indicated by
Equation (3). This would tend to confirm the form of Equation (1), but
suggests a relation different from Equation (2) for the power dissipation
rate in the bubble generation region of our device. Equation (3) might
apply when power is added to the fluid continuously as in an agitated tank
or in pipeline flow where the friction losses represent a continuous energy
dissipation within the fluid. In the present bubble generator, the fluid
may receive an "energy impulse' as some of the kinetic energy of the high
velocity fluid in the thrcat is converted to fluid turbulence in the dif-
fuser, and the above equations may not apply specifically for this mechan-
ism.,

An alternate expression for the power dissipation rate based on the
wall shear stress has been proposed by Kress* for the GSTF bubble genera-
tor design. Using his prcpesed relation for power dissipation, Equaticn
(1) gives the following relationship predicting a 3/5 power dependence
cn surface tension and a -4/5 power dependence on the Reynoclds number

as observed.

 

3/5 -L/5
gp D 1/3 68/3 g V, D. p
g_}z C 2 C t 72 (L)
D, L/3  2/3
Ue Ug Ue

At the present time, there are insufficient data to verify Equation (L)
because only the liquid velocity and liquid surface tension have been
varied. Therefore, we have elected to empirically correlate the data
using the dimensionless groups that appear in Equation (3). These same
dimensionless groups have been obtained independently by dimensionsal
analysis.

The recommended form of the equation for the
GSTY bubble generator is then:

-
D, 8J3/5 v ol

_ 272
<%}-—KD2 5 J " (5)
=2 H
where K = L.54 x 10 ~.

 

*Personal communication, T. Kress to C. H. Gabbard, Dec. 4, 1972.
1k

The comparison of this correlation with the data is shown in Figure 6.
Several data points which were not used in determining the wvalue of K are
- indicated on the plot. These were the points on Figure L4 that did not
fall on the lines representing the 3/5 power of surface tension. Based
on this correlation, the bubble diameter produced by the GSTF bubble
generator operating with fuel salt flowing at 500 gpm should be about
0.01, The value of "K" given above is believed applicable only to
bubble generators that are geometrically similar to the GSTF design. This
is shown by the data on Figure 3 for the smooth bore design which had the
same throat diameter, but had a 7° diffuser cone instead of the 15° cone
in the GSTF design. A larger value of "K" would be required for the smooth

bore design.
I1TI.-2. Gas Injection Pressure Characteristics

To appreciate the importance of the gas injection pressure, anlunder—
standing is needed of the relationship of the bubble generator to other
portions of a reactor system. Figure 7 is a simplified flow diagram of
the GSTF which is representative of a reactor system in regard to the oper-
ation of the bubble generator. The gas injected into the fecllowing salt at
the bubble generator is removed by the bubble separator and is recycled
back to the bubble generator via the bulk salt separator, the drain tank,
and the gas holdup tank. The gas holdup tank including the throttle valves
on either end simulates the delay time and flow restriction of a L8-hr
charcecal trap which in a reactor system, would allow radioactive decay of
the Xe-135 concentration to an acceptable level prior to reinjection of
the helium sweep gas back into the salt system. If the pressure required
to inject the gas into the bubble generator were sufficlently below the
pump tank (or drain tank) pressure to provide the pressure drops for the
48-hr charcoal bed and for the gas flow control valve, a compressor for
highly radiocactive gas would not be required. This concept has been shown
to be feasible and the necessary design features have been incorporated
into the final GSTF bubble generator and system designs for continued
evaluation with hot fuel salt.

The measured pressure differences vs gas flow rate of the final GSTF
prototype bubble generator are shown in Figure 8 for a liquid flow rate of

500 gpm. These pressure differences are expressed as zero-void liquid head
GSTF BUBBLE GENERATOR «d, (in.)

0.050

0.040

_Cl
<
)
o

0.020

0.010

ORNL—-DWG 73—1546

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIQUID FLOW RATE
A =200gpm
A =350 gpm
@ - 50(1 gpm | 0.020
O DATA POINT EXCLUDED FROM /
CALCULATION OF K A
| ! — 0015
3/5 1-4/5
<dv>=4.54x10-202[‘_’_"f_2_9£ [VD?p
u? H
J
N
I | 2
A A —] 0010 -
- o
. M ~
o ® LA & |Oo oo
® o
,F(r)r !L ——3 0.005
L PREDICTED OPERATING POINT
MSR FUEL SALT AT 500 gpm
| | .
0.05 0.10 0.15 0.20 0.25 0.35 0.40

BUBBLE SIZE CORRELATION FOR GSTF DESIGN BUBBLE GENERATOR

[0 PD2 90]3/5 [V Do p
u2 M

] ~4/5

FIGURE 6

qT
ORNL.— DWG 71-13437R2
STACK

  
  

  
    
    
   
 
  

 

 

 

 

 
  

 

 

 

 

 

 

 

 

  
 
 
   
     
 
   

 

 

 

 

 

 

 

 

 

 

BLOWER
SAMPLER
INSTRUMENT N
PENETRATIONS f ENCLOSURE
PORTS SALT PUMP LANCE
STATION
fl H”/] GAS FLOW
SALT (, ATOR
MONITORING | __BULK SALT SEPARA FLOW
STATION e MEASURING
TTT] £ VENTURI
FUTURE |5 f
EXPERIMENTS \ GAS-SALT MIXTURE
T\‘l LOAD ORIFICE T
TIT :
BUBBLE
1 EESEEETOR/////P SEPARATOR
SAMPLER
| e—p4— HELIUM SUPPLY
Oik“——GAS FLOW CONTROL VALVE
48- HOUR
GAS HOLD- i HELIUM EXHAUST TO STACK
UP TANK

 

 

GAS FLOW

FIGURE 7
SIMPLIFIED FLOW DIAGRAM OF THE GAS SYSTEMS TECHNOLOGY FACILITY

9T
17

ORNL—-DWG 73-1545

 

12

10 O

Od)o//

8 0
6 Qfif:w
O DISCHARGE TO INLET

0 52—

 

 

 

 

 

 

 

 

BUBBLE GENERATOR DISCHARGE

 

 

 

 

 

—16 /

DISCHARGE TO GAS
—-18 — INJECTION LINE

 

 

 

LIQUID HEAD DIFFERENCE (ft of liquid)

 

~22

 

 

/00

/ 00 CALCULATED

—24
& O  MEASURED
26 £ o

 

 

 

 

 

 

 

 

 

 

 

 

0 1 2 3 4 5

GAS FLOW RATE {cfm at throat conditions)

FIGURE 8

GAS INJECTION PRESSURE AND OVERALL PRESSURE DROP OF PROTOTYPE
BUBBLE GENERATOR AS A FUNCTION OF GAS FLOW RATE .
18

and are referenced to the bubble generator discharge because, in the pro-
posed piping system, this pressure is more closely related to the system
reference pressure in the pump tank gas space. The increase in gas injec-
tion pressure with gés flow rate was greater than would be indicated by
the increase in diffuser losses and by the increase in the gas passage
pressure drop.

A study of the measured pressure drop data and the various hydro-
dynamic mechanisms of the bubble generator indicated that the pressure

differences could be described by six terms:

Hl The inlet to throat head difference.

H2 The mixing losses and head recovery across the sudden
enlargement from 2.1 in. to 2.18 in.

H3 The mixing losses and head recovery across the 15°
diffuser cone.

Hh The liquid head equivalent to the gas compression
work between throat and discharge pressure,

H5 The liquid head equivalent to the pressure drop in
the gas passages.

H6 The liquid head difference between the liquid and the

gas plume.

As a convenience in comparing different fluids, each of the above terms
were expressed as feet of zero-void liquid head. With the exceptions of
Hl’ which is dependent only on the liquid and of H5’ which is dependent
only on the gas, the pressure differences are a function of both liquid
and gas flow rates, The procedures used in evaluating these six terms
are discussed below.

Figure 9 shows the bubble generator geometry used for the following
analysis and the location of the various pressure drops outlined above.

The fluid head, H_, between the inlet and the throat may be calculated

l!
from the conventiconal venturi equaticn:

1/2
Q= Fa Ft A2 cV (2ng)
Q 2
1 1
H = = |————— (6)
1 2g A2 Fa Ft Cv
ORNL-—DWG 73—-1544

THROAT PRESSURE TAP
© (; Qe @ <%)
Hy - -

o i_ Hy AND Hg _1
fi//////////?//////////)
S ) L

 

 

 

 

Dg = Dy
D4 C5
A
1 ] NN ;
V; D3 A 4=D3
v2 V3T V2 v
1 e 4

 

 

 

GAS CAVITY

GAS INJECTION LINE

FIGURE $

GEOMETRY OF BUBBLE GENERATOR USED IN ANALYSIS OF GAS INJECTION
PRESSURE AND OVERALL HEAD LOSS

6T
20

The observed inlet-throat head difference agreed with the calculated
value within about 2%.

The change in fluid head across the sudden enlargement from a diameter
of 2.1 in. to 2.18 in. H2,can be calculated theoretically by a mcmentum
balance across the length of the 2.18 in. cylindrical bore. The effect
of the gas volume on the fluid velocity was also included in the momentum
balance. The boundaries for the momentum balance are taken Just within
the 2.18 in, diameter at each end. The upstream velocity at Station No. 3
is assumed uniform and equal to the average velocity at Station No. 2.

The liquid and any injected gas are assumed mixed and at uniform velocity
at Station No. 4. The increase in mass flow rate due to the gas addition
was negligible compared to the mass flow of liquid and was not included

in this calculation.

 

 

M (v, - V.)

T - L 3
gC
PA, - P)A = P2 Vs 0o Wy = V)
373 LTk g,
A3 = A
i (P3 - Ph) ] A2 v3 (Vh - v3)
H, = - = o (7)
e 3 ¢

The value of H2 obtained from the momentum balance includes a mixing
loss as well as the change in velocity head that would be predicted by the
Bernoulli equation.

The pressure recovery and head loss in the diffuser cone can be cal-

culated by the Bernoculli equation,

where "n" is the "Borda-Carnot" loss:
K (Vu - v_)°

1 5
2g

h =
C
21

A value of Kl = 0.317, determined experimentally for the existing
bubble generator, agrees closely with the conventional textbook value for
a 15° diffuser.(T)

calculated head rise of the diffuser secticon in terms of zerc-void fluid.:

A void fraction correction was applied to express the

> 5 2

) Vh - v5 Kl (Vh - v5)

H. = - 1 - x|, (8)
3 2gc 2gc

 

In addition to the normal hydraulic losses in the diffuser, the work
required to compress the gas is supplied by the kinetic energy of the
liquid and decreases the head rise in the diffuser. The work required for
a polytropic compression of the gas is given by the equation:

n-1

p n
_ n RT il

W o (Ph) -3
The work of compression can be converted to equivalent liquid head

by multiplying by the ratioc of the mass flow rate of gas to that of liquid.

\ = T U e (9)
The pressure drop through the gas passages of the bubble generator was

determined experimentally as a function of the gas volume flow rate,

FPigure 10 shows the results of the tests. The results expressed as feet

of gas head vs volume flow rate are applicable toc any gas. The gas pres-

sure drop expressed as feet of liquid head is given by the following equa-

tion which applies specifically to the geometry tested:
P

Ho=C @ (&) (10)
Pe

where C = 59.4 mine/ftB; Q@ = volume flow rate of gas, cfm,

During the various pressure drop tests of the bubble generator, the
gas feed pressure was observed tc be higher than the sum of the static
throat pressure and the pressure drop across the gas feed passages.

This "Plume D/P" apparently represents the pressure difference required

to divert the liquid around the gas cavity similar to an impact pressure
GAS PASSAGE AH (ft of gas)

22

ORNL—-DWG 73—1543

10,000

5000

2000

1000

500

200

100

50

20

 

10
0.1 0.2 0.5 1 2 5 10

GAS FLOWRATE {cfm)

FIGURE 10

PRESSURE DROP OF GAS FEED PASSAGES AS A FUNCTION
OF GAS FLOW RATE
23

cn a solid object. Direct measurements of this pressure difference were
made on the original smooth bore bubble generstor design by using one of
the gas feed holes as a static pressure tap for the throat liquid. The
pressure drop across the interface of the gas plume was obtained by

subtracting the gas passage pressure drop "H_." from the measured pressure

difference between the gas injection line ang the static tap. The results
of the measurements are shown in Figure 11.

The "Plume D/P" was found to be directly proportional to the liquid
specific gravity and was found to be a function of both the liquid and the
gas flow rates. An empirical correlation relating the plume D/P, the
liquid velocity in the throat, and the void fraction at the throat was
determined which gave a good representation of the data from various flow
rates for two fluids. A coefficient "Kz" was defined as the ratio of the
plume D/P, expressed as feet of liquid head, to the liquid velocity to
the 2.5 power. The value of K2 vs the void fraction for the existing
data was fit to a polynomial by the least squares method. The value of
K2 was best described by a cubic equation, and the results of the fit are
shown in Figure 12. The value of the plume D/P would then be calculated

as fcllows:
2.5

= )
He = K, v, (11)
2 3
where K2 = (A+BX+CX +DX°)
Q
X = B
9 +Q
e g 6
and A = -1.84825 x 10~
B = -1.26802 x 10‘2
C = 0.171324
D = -0.885819
V, = Liquid velocity at throat (ft/sec).

Equation (11) gives & negative value of He consistent with the sign
convention used in the computer program, BGNDGN, discussed in the Appendix.
The pressure in the gas relative to the liquid is actually positive.

The pressure distribution of the bubble generator can be obtained by
the summation of the above six terms. A BASIC language computer program,

BGNDGN, was written to calculate the pressure distribution of the GSTF
PLUME AH ({ft of liquid)

10.0

5.0

2.0

1.0

0.5

0.2

0.1
0.001

 

ok

ORNL-DWG 731538

—
_§

 

 

 

GAS FEED |

500 gpm
|

O CaCl, SCLUTION SG = 1.3
® DEMINERALIZED WATER

0.02 0.05 0.01 0.2 05 0.1

THROAT VOID FRACTION QQ/(QQ + Qq)

FIGURE T1

PRESSURE DROP ACROSS THE GAS PLUME INTERFACE
AS A FUNCTION OF THROAT VOID FRACTION
(x 104

2.5
Hg/V2

K=

25

ORNL—-DWG 73-1540

 

—4

 

 

-1 ——

 

 

 

K=A+BX+CX2+DX3
A=—184825x 106
B =—1.26802 x 102
C=0.171324
D = —0.885819

| |

Hg = PLUME D/P {ft of liquid)

 

 

—— V5 = THROAT VELOCITY (ft/sec)

| |

 

|

O CaCly$G = 1.3 500 gpm
® CaCl, SG=1.3 400 gpm
A CaCly SG=1.3 300 gpm
A H,0 SG=1.0 500gpm

 

 

 

0.02 0.04 0.06

THROAT VOID FRACTION X = Qg/(Q, + Q)

FIGURE 12

0.08

0.10

CORRELATION OF PLUME AH (HG) TO THROAT LIQUID VELOCITY

AND THROAT VOID FRACTION
26

bubble generator as a function of the liquid and gas flow rates. Although
the procedure for calculating the value of H6 was based on data from an
earlier bubble generator design, this procedure was used in the BGNDGN pro-
gram and the results appear to be applicable to the final GSTF design.

A listing of this program and sample output for the GSTF operating with fuel
salt are included in the Appendix. The above procedures and the computer
program were developed specifically for the bubble generator design for the
GSTF and were checked against data from the prototype in the water test
locop. However, with the exception of the gas passage pressure drop, the
procedures are believed to be applicable to various fluids and sizes
assuming a reasonable geometric similarity. The calculation of the gas
passage pressure drop could be revised to use conventional pressure 4rop
calculation procedures for any other particular design.

A comparison of the calculated and measured pressure distributiors for
the GSTF prototype design is shown in Figure 8, These pressure distributions
apply specifically to the prototype bubble generator in the water test loop
and would differ slightly from the distributions of the actual GSTF bubble
generator because of the difference in pipe size and the difference in
absolute pressure. The calculated pressures for the GSTF bubble generator
operating at design conditions with water and two types of molten-salt
are shown in the Appendix.

The calculations in the Appendix for fuel-salt indicate the gas flow
for normal gas recycle operation will be limited to about 1 scfm by the
various pressure changes inherent in the system. These calculaticns were
based on a pump tank pressure of 15 psig, a salt pressure of 28 psig at
the bubble generator discharge, and a pressure drop of 7.5 psi across the
48-hr holdup tank at 0.8 scfm gas flow rate. Operation of the GSTF at
higher gas flows up to 1.3 scfm can be achieved by either opening the
throttle valves at the L8-hr holdup tank cor by operating on an open cycle
with the gas supplied from an external source at somewhat higher pressure.
The gas flow capacity of the GSTF was specified a factor of 2 greater than
for the MSBE to provide a margin for experimental purposes, and the maxi-
mum gas flow of 1 scfm would not be a limitation in the MSBE. The lower
than predicted gas feed pressure was probably caused by a local flow

disturbance at the step in throat diameter. This belief is supported by
27

the facts that the measurements of the throat pressure just upstream

of the step were in good agreement with the calculated value of Hl’ and
at low gas flow rates the overall pressure drop of the bubble generator
was in good agreement with the summation of Hl’ Hg, H, and H .

3 b

The calculated values of HE’ H3 and Hh are subject to some degree
of error because in an actual bubble generator it is impractical to pro-
vide a throat mixing length long enough to complete the momentum transfer
assumed in the calculations. Part of the mixing losses assigned to the
mixing section occur in the diffuser and could account for the differences

in slope between the calculated and measured pressures shown on Figure 8.

IV, CONCLUSIONS AND RECOMMENDATIONS

The bubble generator design developed for application in the Gas
Systems Technology Facility and the Molten-Salt Breeder Experiment is
expected to successfully meet the criteria specified in Table I.

The gas flow limit at about 1 scfm would be a valuable safety
feature in the event of a malfunction of the gas flow control system at the
maximum flow position. However, an initial transient at a higher flow
could occur depending on the location and size of holdup volumes and pres-
sure drops in the gas system. The design of a reactor gas-system should
attempt to minimize the rate and duration of this transient.

There are uncertainties in regard to the mechanism of bubble formation
in the bubble generator and a relatively extensive program would be required
to fully evaluate the proposed mechanisms. However, the bubble size pro-
duced by the bubble generator is believed to have a minor influence on the
overall operation of a reactor circulating system because of the bubble
degradiation and compression in passing through the pump and the other
changes in size that may occur because of coalescence, gas solubility,
and pressure changes. Therefore, an effort to fully evaluate the proposed
mechanisms of bubble formation does not appear to be justified at this
time. However, plans have been made to check the viscosity dependence of
the reccommended correlation.

The calculation procedures and computer program developed for esti-
mating the pressure distribution as a function of liquid and gas flow
rates appears to be sufficiently accurate for most applications. The

suitability of these design calculations to cover operation in a high
28

temperature salt system will be evaluated from the operating data of the
GSTF »

The calculation procedures for various pressures and bubble diameter
are believed to be applicable to other sizes, but we have no experimental
verification of this. The reduced scale tests completed early in the pro-
gram were survey type experiments and insufficient data were taken to
evaluate scale effects. Therefore any bubble generator of significantly
different size or geometry should be checked experimentally against the

calculations prior to use in any critical application.

V. ACKNOWLEDGEMENT
The author is indebted to many individuals for their cooperation and
assistance in completing this work. In particular, the contribution of
T. S. Kress in suggesting mechanisms for the bubble size correlation and
the contribution of G. M. Winn for this operation of the test loop and

his execution of the various experiments are gratefully acknowledged.
29

NOMENCLATURE
Cross sectional areas of bubble generator (see Fig. 9)(ft2)
Diameters ¢f bubble generator (see Fig. 9)(ft)

2,-1/2
Approach area factor [l~(A2/Al) ]

Thermal expansion factor [1 + a(T—SBO)]2

Various differential heads associated with bubble
generator (see Fig, 9) (ft of zero void liquid)

Molecular weight of gas

Reynolds number, VDp/u

Static pressures in bubble generator (see Fig. 9)(lbf/ft2)
Volume flow rate of liquid or gas (ft3/sec)

Universal gas constant 15L45.3 ft/1bf/# mole-°R
Temperature, °R

Velocities in bubble generator (see Fig. 9)(ft/sec)

Void fraction

Velocity coefficient assumed = 1.0

Bubble diameter (ft)

Volume averaged bubble diameter (ft)

Gravitational conversion factor (lbm—ft/lbf—secg)

Mass flow rate of liquid or gas (lbp/sec)

Polytropic gas compression constant

Coefficient of thermal expansion per °F = 8 x 10_6
Gas film thickness (ft)

Power dissipation per unit volume (ft~lbf/ft3-sec)
Viscesity of liquid or gas (1bp/ft-sec)

Density of liquid or gas (1lby/ft3)

Surface tension (lbg/ft)
(7)

30

REFERENCES

M. W. Rosenthal, P. N. Haubenreich, and R. B. Briggs, The Develop~-
ment Status of Molten-Salt Breeder Reactors, ORNL-4812, August
1972, pp 2Th-297.

J. R. McWherter, Molten Salt Breeder Experiment Design Bases,
ORNL-TM-3177, November 1970,

R. H. Guymon, System Design Description of the Gas Systems
Technology Facility, ORNL-CF-72-3-1, March 30, 1972.

MSR Program Semiannual Progress Report Feb. 28, 1969, ORNL-4396,
p 95.

J. 0. Hinze, Fundamentals of the Hydrodynamic Mechanism of Splitting
in Dispersion Processes, AIChE Journal, Vol. 1, No. 3, 1955, p 295-
298,

T. S. Kress, Mass Transfer Between Small Bubbles and Liquids in
Cocurrent Turbulent Pipeline Flow, ORNL-TM-3718, April 1972, p 58.

J. K. Vernard, Elementary Fluid Mechanics, John Wiley and Sons,
1947, p 176.
31

APPENDIX

COMPUTER PROGRAM, BGNDGN, FOR CALCULATING GAS INJECTION PRESSURE AND OVER-
ALL PRESSURE DROP OF VENTURI TYPE BUBBLE GENERATOR

A computer program, BGNDGN, was written in BASIC language to cal-
culate the overall pressure drop and the gas injection pressure of the
bubble generator. This program, which uses the relationships discussed
in Section ITI-2, was written specifically for the GSTF bubble generator
design. The program should be valid for different sizes with geometric
similarity. The program is listed below along with output covering the
bubble generator operation in the GSTF with the proposed fluids.

The required data input for running the program are in statements
450 and L51 as follows:

450 Data Fl, G, G9, Mo, N

Fl = liquid flow rate {gpm)

G = liquid specific gravity

G9 = gas density (lbs/ft3)

MO = molecular weight of gas

N = pelytropic constant for gas

451 Data D8, P9, P8, T2, K9

D8 = inlet and discharge pipe ID (in.)
D9
P8
T2 = salt temperature (°F)

pressure at discharge (psig)

pump tank pressure (psig)

K9 = thermal expansion factor (1 + o A T) for Hastelloy "N"

The gas flow rates are input into statement 140 as scfm. Different
throat and mixing chamber diameter {inches) could be entered in statement
30 and 31, respectively.

There are four lines of output for each gas flow rate as foliows:

Line 1., a. H1l

lead difference across the diffuser cone
(ft of zero void liquid)

b. HZ2 = head difference across the mixing chamber
(ft of zero void liquid)
c. H3 = head equivalent to gas compressicn work

(ft of zero void liquid)
32

d. HL = total head difference between the gas
injection line and the salt at throat
(ft of zero void ligquid).

e, H5 = head difference across the the gas passages

(ft of zero void liquid).

Line 2. a. The gas flow rate (scfm).

b. The gas flow rate (cfm at throat pressure and
temperature).

¢. The salt static pressure at the throat (psig).
d. The static pressure at the gas injection line (psig).

Line 3. a. The pressure in the GSTF gas line 210 upstream
of the flow control valve (psig).

b. The pressure drop across the GSTF gas flow control
valve (psig).

Line 4, a. The overall head loss of the bubble generator
(ft of zero void liquid).

b. The head difference between bubble generator dis-
charge and the gas injection line (ft of zero void
liquid).

A negative value of the value D/P in output line 3 indicates the
pressure drop available in the GSTF gas-system 1s insufficient to provide
recycle operation at that flow rate and at the design flow restriction
of the L8-hr holdup tank. This calculation assumes a 7.5 psi pressure
drop across the 48-hr holdup tank at 0.8 scfm and that the pressure

drop varies with the square of the volume flow rate.
1 R
2 R
10
20
21

20
100
110
120
130
131
132
140
150
160
161
170
180
190
200
210
220
230
240
249
250
260
261
270
280
290
300
310
320
321
322
323
332
340
360
370
380
381
382
383
399
400
401
402
403
404
405
450
451
500

33

EM RGNDGN CHG &6/8/72 CALC RURRLE GEN PRESS DIST
EAD Al1,A2,A3, 04
READ F1sG»G2sM0sN
READ D8sP9,P8, T2,K9
H1=0
H2=0
H3=0
D9=2.10
Di=2.18

PO=14.7

PRINT *"LIQ SG=""G,"M@L wWT="M0,'"'DIS PRESS="P9
PRINT "LIQ FLAW="'F1,"THR DIA="D9,"B@RE DIA="D1
p2=pg8/12

D=D1/12

DO=D9/12
P2=P9+14.7
T=T2+ 460
K1=N/7C1-N)

K2=(N=1)/N

Q0=F1/7(T7+48%60)

MI=R0*G%x62+ 4

VO=Q00/(0.785%x(DO*XKI) *2)
V=Q0/7C(0.785%(D2%K9) 12)
H=(V012-yt2)/64.4

FOR F2=0 T? 1.41 STEP .2
M2=F2%G9/ 60

Z=0

PO=P2-(H1+H2+H3)Y*x62.4%G/ 1t 44
CI1=0Q0+F2%T/493%14.T/{PO*x60)
A2=QO0+F2%T/493%14.7T/(P2%60)
V1=01/C0.785%(D*K9)12)
Ve=R2/¢(0.785%x(D2%xKI)t2)
FO=F2%14.7/P0*T/ (49 3%60)
X2=F9/(Q0+F93)
H1=C((VV112-Y21t2)/64¢4~0+31T7/644%(V1-yY2)1t2)%(1.0-X2)
G1=G*00/0Q1

F3=F2%T/493%14.7/P0
H2==V1*Gi* (V1 =V0)/(32.2%()-F3%0. 4167
PO=P2-(H1+H2+H3)%62. 4G/ 144

IF PO<0O THEN 404
W=K1*%1545%T/MO({P2/PO)tK2=-1.0)
H3=WxM2/M1
PO=P2-(HI+H2+H3)*62.4%G/1 44
F3=F2xT/493%14.7/P0C
GEB=GI9%493/T*PD/14.7
H5=2=59« A8B8%F 31 2% G/ (5% &2 4)
X3=F3/(F1/T7448+F3)
K3=A1+A2%X3+A3%X312+04%X31t3
HA4a=K3*%VO012.50+HS

P3=P0-14.7
PO=P2=-(HI+H2+H3+HA) %62 . 4%G/ 144

Z=2+1

IF Z<3 THEN 170
P7=P8-3.58%(F2+0.08476)1t2
P&=P7~(P0-14.7)

Li=sHI+H2+H3+HA4

L=H~-(H1+H2+H3)>

PRINT

PRINT H1,H2,H3sHAasHS
PRINT"SCFM="F2,"TCFM="F35s"T PR="P3,"G PR="P0C-14.7
PRINT"L 210 PR="pPT,"VALVE D/P="'Pé&
PRINT "I1-@ D/H="L,»"0~G D/H=""L1

NEXT F2

DATA ~1.84825E~62,-1.P6802E~2,0+1713245,-0.885519
DATA S00,3:28365001125451067

DATA S.047s2B8515,13C0,1.009

END
34

Table A-1

GSTF Bubble Generator Operation on Fuel-Salt and Helium

LIQ SG= 3.2836
LIG FLOW= 500 THR DIA= 2.1

20.9288 430198
SCFM= 0O TCFM= O
L 210 PR= 14.9312

1-@ D/H= S.97787

211368 320759
SCFM= 0.2 TCFM= 1.25849
L 210 PR= 14.2232

I-@ D/7H= 7.06068

21.3703 2.50307
SCFM= C. 4 TCFM= 2.23567
L 210 PR= 1247488

I1-0 D/H= 7.7982

2145242 1.8788

SCFM= 0.6 TCFM= 3.05499
L 210 PR= 10.508

I-0 D/H= B.44298

2146677 1.29773
SCFM= 0.8 TCFM= 3.76738
L 210 PR= 7.50077

I1-9 D/H= 9.03121

21.802 De 754595
SCFM= 1 TCFM= 4.,40476
L 210 PR= 3.72717

1-0 D/H= 9.57257

219267 0250605
SCFM= 1.2 TCFM= 4.98902
L 210 PR==0.812R27

I-0 D/7H= 10.0705

22041 ¢ ~0e21224
SCFM= 1.4 TCFM= 5.53574
L 210 PR=-6.11923

[-2 D/7H= 10.5269

MAL WT= 4 DIS PRESS= 28
BORE DIA= 2.185
0 -2e¢58365E-2 G

T PR=2=7.90077 G PR==T7.86401
VALVE D/P= 22.7952

B-G D/H= 25.205

~0.256411 -2+5625 -2+ 1B4S1E-4
T PR==6.36006 G PR==-2.7139

VALVE D/7P= 169371

@-G D/H= 21.5855

-De 462339 -3+ 67663 -2.90790FK-3
T PR==-5.31064 G PR=+-T91799E~-?

VALVE D/P= 12.828

A-G D/H= 19.7339

«0.637301 ~4423783 -S¢96038E-3
T PR==4439319 G PR= 1+.6368¢6

VALVE D/7P= 8.87111

-G D/H= 18.5278

~0.787922 -4455608 =3 +8003€E-3
T PR=-3.5562 G PR= 2926672

VALVE D/P= 4.57415

Q-G D/H= 17.6214

-0.,920421 ~4.77094 ~1043230CE~2
T PR=-2.7359 G FR= 4«00P€4

VALVE D/P==-0.275467

@~-G D/7H= 16.8652

-1e03916 ~-4435226 =10 46T4E-2
T PR==-2.07739 G PR= 4.96915

VALVE D/7P=-5.78198

@-G D/H= 16.1859

-1.14751 -5« 13854 =2.52009E-2
T PR==1+442803 G PR= 583367

VALVE D/P=-12.0028
@-G D/H= 15.5432
35

Table A-2

GSTF Bubble Generator Operation on Flush Salt (66~3L Mole % LiF-BeF»o) and

 

Helium
LIQ SG= 1942 M@L WT= 4
LIQ FLBW= S500 THR DIA=s 2.1

209288 430198
SCFM= 0 TCFM= O
L 210 PR= 14.9312

I1-3 D/H= 5.97787

214108 357596
SCFM= Q.2 TCFM= 0625965
. 210 PR= 14.2232

I-8 D/H= 6&.7026

21.2619 2.95381
SCFM= 0.4 TCFM= 121356
L 210 PR= 12.7488

I-@ D/H= 7.33261

2144023 2.38724
SCFM= 0.6 TCFM= 1.77081
L 210 PR= 10.508

I-2 D/H= T.90747

215347 185413
SCFM= 0.8 TCFM= 2.30244
L 210 PR= 7.50077

I-0 D/H= B.44564

216615 134347
SCFM= 1 TCFiM= 281173
L 210 PR= 3.72717

1-0 D/H= B +9566E

21.7843 O0eB497 62
SCFM= 1.2 TCFM= 3.30129
L 210 PR==0.812827

1-0 D/H= 944572

2190364 0370441
SCFM= 1.4 TCFM= 3.7733
L 210 PR==-6.11923

I-0 D/H= 9.91553

DIS PRESS= 22.69
RGRE DIA= 2.18
0 ~2«58365E~-2 0
T PR= 1.45742 G PR= 1.47916
VALVE D/P= 13.452
-G b/H= 25.205
-0.177838 -1 47477 —6.88325E~-4
T PR= 2.0673 G PR= 3.30837
VALVE D/7P= 10.9148
@-G D/H= 23.0313
~0.339626 ~2¢ 4979 ~2.66891E-3
T PR= 259748 G PR= 4.69954
VALVE D/P= B8.04923
@-G D/H= 21.3782
-~0e 4838344 =322396¢ ~5+84167TE~-3
T PR= 3.08124 G PR= 5.79431
VALVE D/P= 4.71367
2~G D/H= 20.0773
T PR= 353413 G PR= 6€.62143
VALVE D/P= 0.819344
Q-G D/H= 19.0231
-0.75297S ~ 4. 108 49 = 1¢54592E-2
T PR= 396418 G PR= 7.42162
VALVE D/P=-3.69445
-G D/H= 18.1435
-0.871064 -4+376 “2.17810F-2
T PR= 437573 G PR= 8.05827
VAL VE D/7P=-8B8711
@-GC D/H= 17.387
~0e9K0R826 ~4.5775 ~2e90444F-2
T PR= 477109 G PR= B.6232
VALVE D/P=-14.74724
@-G D/H= 16.7157
36

Table A-3

GSTF Bubble Generator Operation on Water and Helium

LIG SG= 1 MIOL WT= 4
LIG FLOW= 500 THR DIA= 2.1
21.6925 40 45896
SCFM= O TCFM= O
L 210 PR= 14.9312

I-2 D/H= 6.136

21.9089 3.5988

SCFM= 1t TCFM= Q+.695957
L 210 PR= 3.72717

-9 D/H= 7.07409

22.1104 279977
SCFM= 2 TCFM= 1.37067
L 210 PR=-26.6368

I-8 D/H= T7.88634

223024 2.03906
SCFM= 3 TCFM= 2.02683
L 210 PR==-76+.1608

[0 D/H= 8.65218

22. 488 130439
SCFM= 4 TCFM= 2.66€37
L 210 PR=-144.845

I-0 D/H= 92.3325

22 6688 O« 589443
SCFM= 5 TCFM= 3.29081
L 210 PR=-232.689

I-9 D/7H= 10.0833

22.8454 -0+ 108465
SCFM= 6 TCFM= 3.90148
L 210 PR=-339%9.693

I-@ D/H= 10.7578

DIS PRESS= 18.959
RORE DIA= 2.18
0 ~2.70202E-? 0
T PR= 7+6267 G PR= 7.63841
VALVE D/P= 7.29277
@-G D/H= 26.1245
-0.23437) ~1.69285 -7+ 43097E-3
T PR= 8.0072 G PR= 8.74077
VALVE D/P=-5.0136
0-G D/H= 23.5805
-0.449001 -2.87493 -2.92702E-2
T PR= B.35917 G PR= 9.6049R
VALVE D/P=-3642418
@-G D/H= 21.5862
-0.64615 -3.70827 - 6.490235E-2
T PR= 8.69104 G PR= 10.298
VALVE D/P=-86.4588
@-G D/H= 19.987
-0.827413 ~4.29173 -0-113879
T PR= 9.00751 G PR= 10.8637
VALVE D/P=-155.715
9-G D/H= 1846677
~0e994064 ~4.7P 431 =0.175685
T PR= 9.3112 G PR= 11.3584
VALVE D/P==244.047
B-G D/H= 1745399
~1.14722 ~5.05471 -0.249944
T PR= 9.60347 G PRP= 11.793K
VALVE D/P=-351.487
G-G D/H= 164535
37

TNTERNAL DISTRIBUTION

1. 5. K. Beall

2. M. Bender

3. C. E. Bettis

. E. S. Bettis

5. R. B. Briggs

6. C. J. Claffey

7. C. W. Collins

8. W. B. Cottrell
9. D. E. Ferguson
10. A. P. Fraas

11.-12 C. H. Gabbard

13. A. G, Grindell
14, P. N. Haubenreich
15. H. W. Hoffman
16. P. R. Kasten
17. M. I. Lundin
18. R, N. Lyon
19, H. G. MacPherson
20. R. E. MacPherson
21. H. C. McCurdy
22. A. J. Miller
23. A, M. Perry

2l,-25., M. W. Rosenthal
26. Dunlap Scott
27. M, Sheldon
28, M. J. Skinner
29. I. Spiewak
30. D. A. Sundberg
31. D. B. Trauger
32. G. D. Whitman
33.-34, Central Res. Library
35. Document Ref. Section
36.-38. Laboratory Records
39. Laboratory Records (LRD-RC)

EXTERNAL DISTRIBUTION

Lo, M, Shaw, AEC-Wash.
hi1.-Lk2, ©N. Haberman, AEC-Wash.

43, D, F. Cope, AEC-0OSR

L, David Elias, ARC-Wash.

b, J. E. Fox, ARC-Wash.

L6, E. C. Kovacice, AEC-Wash.
4L7.-49., Director, Division of Reactor Licensing, AEC-Wash.
50.-51. Director, Division of Reactor Standards, ALC-Wash.

52,-68. Manager, Technical Information Center, AEC-For ACRS Memiers

A9. Research and Technical Support Division, AEC, ORO
70.-71. Technical Information Center, AEC