Experimental investigation of hydrodynamic phenomena in vertical-upward adiabatic two-phase flow conditions
Nuclear Science and Technology, Vol.10, No. 3 (2020), pp. 49-59
Experimental investigation of hydrodynamic phenomena in
vertical-upward adiabatic two-phase flow conditions
Dinh Anh Tuan*, Pham Tuan Nam, Nguyen Tu Oanh
Institute for Nuclear Science and Technology, Vietnam Atomic Energy Institute,
179 Hoang Quoc Viet, Ha Noi
(Received 12 September 2019, accepted 14 July 2020)
Abstract: In order to investigate hydrodynamic phenomena in two-phase flow conditions in nuclear
safety analysis, a series of two-phase flow experiments were conducted using a single flow channel in
which air and water were simultaneously injected into the test section. The experiments under
atmospheric pressure conditions were carried out with the water velocity and the air velocity covering
the ranges from 0.2 to 1.5 m/s and 0.05 to 0.2 m/s, respectively. The technique of two-sensor
conductivity probe was used for the measurement of bubble parameters. The experimental results
presented and analyzed in this study are the local time-averaged void fraction and bubble velocities at
three axial positions L/D = 14.4, 51.2 and 71.3.
Keywords: Two phase flow regime, Two phase flow experiment, Conductivity probe method.
conducted that are based on the assumptions of
fully developed flow patterns and without heat
addition to the flow, the so-called adiabatic two-
phase flow. Through these research works, the
flow structure of air-water two-phase flow was
figured out, and many flow-pattern maps have
been proposed using dimensional coordinates
based on the liquid and gas superficial
velocities. All of these maps are based on
experimental data; and, therefore, the big
question is whether these maps can be
extrapolated to a wider range of tube diameters,
fluid properties, and flow patterns or not.
I. INTRODUCTION
In order to determine the heat transfer and
pressure drop characteristics of a given flow in
nuclear power reactors under various
operational transients or accident conditions,
e.g.,
LOCA
(Loss-of-Coolant-Accident),
studying two-phase flow plays a very important
role. The hydrodynamics and heat transfers are a
coupled thermodynamic problem in which
single phase flow is changed. Therefore, phase
distribution, flow pattern and heat transfer
characteristics are also affected. Furthermore,
due to the shape change in large bubbles, two-
phase flow in a channel can hardly become fully
developed at low pressure. So, with the inherent
high complexity, a local description of the
adiabatic two-phase flow is insufficient if there
is not knowledge of the previous “history” of
the flow. The hydrodynamic instabilities and the
thermodynamic equilibrium between the phases
are extremely complicated until now [1]. To
avoid such complexities, a relatively large
amount of experimental work has so far been
For a few decades, a considerable
amount of works on the measurements of local
two-phase flow parameters has been
successfully performed by many investigators
since Neal & Bankoff’s work (1963) on the
measurement of local void profile in air-water
flow condition [2]. However, there are not
enough data to adequately support a wide
range of continuing efforts in the calibration
and validation of advanced models and codes.
©2020 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute
EXPERIMENTAL INVESTIGATION OF HYDRODYNAMIC PHENOMENA IN…
Based on the current practice in
experimentation, modeling, and analysis of two-
phase flow in nuclear reactor safety, the present
work is an experimental investigation of various
local parameters of concurrent air-water two-
phase flow, flowing upward in a vertical circular
tube with inner diameter 25.4 mm under nearly
atmospheric pressure. Due to constraints on the
financial and human resource issues, emphases
are put on the following: 1) Measurement of the
local parameters of two-phase flow which are
the local time-averaged void fraction and bubble
velocities at three axial positions using two-
sensor conductivity probe. 2) Description of
II. EXPERIMENTAL APPARATUS
The experiment was installed at
VINATOM (Vietnam Atomic Energy Institute)
for studying two-phase flow regime, transition
phenomena, and measurement of the two-phase
flow parameters such as void fraction, bubble
velocity. The principle of the experiment is
shown in Figure 1. It consists of test section,
bubble generator, water supply system, air
supply system, and data acquisition system. The
test section is a vertical transparent tube with an
inner diameter of 25.4 mm and a height of 2 m.
Two-sensor conductivity probe is located at
three positions L/D = 14.4, 51.2, and 71.3 for
measuring the two-phase flow parameters [3]. In
order to measure the pressure of the system,
absolute and differential pressures are installed.
The Rosemount 3051C absolute pressure is
connected to the L/D = 14.4, while the
Rosemount 2880 differential pressure is
connected to the L/D = 51.2 and 71.3.
some
statistical
and
hydrodynamic
characteristics of flow patterns as bubbly and
slug in a vertical-upward air-water flow; 3)
Checking the effect of changing the air and
water flow rate as in Run 4 (jf = 1 m/s, jg = 0.05
m/s), Run 9 (jf = 1 m/s, jg = 0.1 m/s) and Run 13
(jf = 0.5 m/s, jg = 0.2 m/s) on the results.
Fig. 1. Schematic of test facility
50
DINH ANH TUAN et al.
Air is supplied by an air compressor with
determine the void fraction and bubble
velocity in air-water two-phase flow, and
thereafter many researchers have developed
this method [4,5]. The two-sensor
conductivity probe is based on the
continuous value of local conductivity in
two-phase flow with each sensor tip acting as
an electrode. The circuit made up of sensor
tips is in the "Open" or "Close" state
depending on the sensor tip contacted with
the water or gas.
a capacity of 1 HP (Horse Power). After passing
the test section, the gas is released to the
environment through a separator while the water
is returned to the water tank. The water flow
rate is measured by a Coriolis flow meter with
the range of 0-3000 kg/h and the error of 0.2% -
0.5% for the entire operating range, while the air
flow rate is measured by a rotameter with the
range from 0.2 - 30 l/m and the error of ± 2%.
The rotameter also measured the bypass water
flow with the measuring range of 6 l/h to 60 l/h.
The data collection and processing system
In this study, the two-sensor
conductivity probe is mounted L/D= 14.4,
51.2 and 71.3 upstream of the bubble
generators. As shown in Figure 2, it is
possible to change the radial position of the
probe in the cross-section using a traversing
mechanism. By performing measurements at
different radial locations, radial profiles of
time-averaged two-phase flow parameters
can be measured.
includes
a
signal
conditioner,
A/D
(Analog/Digital) signal converter, and
a
computer with LabVIEW software. The
visualization of the two-phase flow is achieved
through the high-speed camera with a maximum
frame rate of 1000 and Xenon lamps.
III. SIGNAL PROCESSING TECHNIQUE
OF LOCAL TWO-PHASE FLOW
PARAMETERS
A sketch of the two-sensor probe is
shown in Figure 3. Two sensor tips are made
by thermal couple wire type K with a cross-
section of 0.2 mm. Sensor tips will be
sharpened to the cone shape.
A. Two-sensor conductivity probe
The conductivity method was first
proposed by Neal and Bankoff [2] to
Fig. 2. Instrumentation set-up
51
EXPERIMENTAL INVESTIGATION OF HYDRODYNAMIC PHENOMENA IN…
Fig. 3. Sketch of the two-sensor conductivity needle-probe
make
a
signal-processing program for
B. Signal processing
The sensor probe is initially placed in
obtaining exactly the necessary information
from the raw signal [7,8]. The signal-
processing program was developed on
the water with a low voltage signal [6].
When the sensor contacts with the gas, a
higher voltage level is obtained. The signal
form is very different from the ideal square
wave since each sensor has a finite size and
time delay due to the wetting and rewetting
phenomenon. Therefore, it is necessary to
LabVIEW
software
combined
with
MATLAB language with four main parts:
signal reading & normalization; making cut-
off level; trans-rectangle & filtering; and
bubble statics as illustrated in Figure 4. [9].
Signal
TransRectangle
&
Bubble Statics &
Display
Signal Reading
& Normalization
Make Cut-off
Levels
Filtering
Fig. 4. The structure of the signal-processing program
First, signals from the rear and front
sensors will be read and standardized. The goal
is to eliminate high-frequency noise signals.
Then the cut-off value is set to trigger a square
signal [10,11]. As Yun [12], it is crucial to
determine the appropriate cut-off level for
obtaining the value of the void fraction and
bubble velocity accurately. The threshold level
is calculated based on the standard level of
pulse amplitude and standard level of the slope
edge, in which these standards can vary with
each bubble instead of being assigned by a
given constant for all bubbles.
After determining the cut-off value, the
signal is converted into square signal and
filter out unsuitable parts of a square pulse in
the part of Signal TransRectangle & Filtering.
This process is based on the algorithms of
52
DINH ANH TUAN et al.
Yun [12] and Euh [13]. Finally, the two-phase
flow parameters are calculated in the part of
Bubble Statics.
Where, Nt is the number of bubbles that
strike the sensor; (tTF – tTR)j is the time that the
sensor is exposed to the gas phase.
From the square wave signal, the
number of bubbles that hit the sensor can be
measured by counting the number of pulses in
the signal. The interfacial velocity of each
interface can be obtained by using the
distance to the different tips of the two-sensor
probe and the time delay between the
upstream and downstream signals. The
parameters of the two-phase flow have been
calculated as below [14].
- The time-averaged interfacial velocity:
The interfacial velocity can be computed
by taking into account the span among the tips
of the front and rear sensor and the time
difference between the front and rear signals.
Thus, the time-averaged interfacial velocity is
given as:
| ⃗ |
∑
- The time-averaged void fraction:
Where Δs is the distance between the
front and rear sensor; tRR – tTR is the relative
time between the bubble striking the front and
rear sensor.
The time-averaged void fraction is a
function of the total sampling time - Ω, and the
accumulated pulse widths of the upstream
sensor during the sampling period. Thus, this
time-averaged void fraction is simply the
accumulated time the sensor is exposed to the
gas phase divided by the total sampling time of
the sensor.
IV. PRELIMINARY RESULTS AND DISCUSSION
Experiment data was collected at 8 radial
measurement points with each point distance of
1.5 mm along three axial positions (L/D =
14.4, L/D = 51.2, L/D = 71.3). All the flow
conditions are summarized in Table I.
(1)
∑
̅
Table I. Experimental flow condition
Parameter
Superficial
Run
1-5
Run
Run
Run
Run
Run
Run
6 -10
11-13
14-16
17-19
20-22
23-25
gas
0.05
0.10
0.20
0.30
0.50
0.80
1.00
velocity, jg [m/s]
Superficial water
0.2
0.3
0.5
1.0
1.5
0.2
0.3
0.5
1.0
1.5
velocity, jf [m/s]
0.2
0.3
0.5
0.2
0.3
0.5
0.2
0.3
0.5
0.2
0.3
0.5
0.2
0.3
0.5
53
EXPERIMENTAL INVESTIGATION OF HYDRODYNAMIC PHENOMENA IN…
Experiments carried out are represented
in the flow regime map shown in figure 5 [15].
According to the flow map it can be seen that,
when studying measurement of two-phase flow
parameters, the authors measured with wider
ranges based on the liquid and gas superficial
velocities. However, it is still not possible to
cover the entire. Therefore, the experimental
system built at VINATOM with the
measurement range shown in Figure 5 is
expected to contribute to the missing
experimental data on the flow regime map.
Fig. 5. Test conditions on flow regime map
presents the results of the comparison of the
velocity measured by the imaging technique and
conductivity probe. The difference between the
two measurement techniques is within ± 15%.
This result is good and suitable for use in the
two-phase flow experiment.
A. Signal processing verification
In order to ensure the quality of
conductivity probe and the signal processing,
the imaging technique is applied using the high-
speed camera with xenon lamp for observing
and recording the time of bubbles passed
through the conductivity probe [16,17]. An
independent small-scale experimental system
was built. Test Section is a shape square acrylic
box with a size of 2 × 2 cm2 and a length of 40
cm. A small steel tube is used to generate a
single bubble with a diameter of 2 to 5 mm. It is
possible to determine the velocity of each
bubble through the image processing software
developed by the research team of Hanoi
University of Science and Technology. Figure 6
B. Local time-averaged void fraction
In order to present better local
parameter
distribution
and
transport
characteristics, the results of bubbly flow test
condition Run 4 is selected. The time-
averaged local void fraction and bubble
velocity profiles at all three axial locations
are given in Figure 7. Each row from top to
bottom represents the result at L/D = 14.4,
51.2 and 71.3, respectively.
54
DINH ANH TUAN et al.
Fig. 6. Comparison of bubble velocity obtained by Imaging Technique and Conductivity Probe
(a)
(b)
Fig. 7. Local profile of void fraction for run 4: jf = 1 m/s, jg = 0.05 m/s
(a) Front Sensor; (b) Rear Sensor
In Run 4, the void distribution
bubbles due to the turbulent effect and forced
toward the wall by lift force, resulting in the
wall peak void distribution. This bubble
interaction mechanism also results in small
amount change of void fraction at two upper
axial positions. This phenomenon matches the
measurement result in the work of Dang [5].
experiences a change process of transition
(flat) (L/D = 14.4) – wall peak (L/D = 51.2 and
71.3). This reason is explained by the bubble
breakup mechanism. At inlet position (L/D =
14.4), wake entrainment mechanism is
dominant, small bubbles coalescence to larger
bubble, and drive bubbles toward pipe center.
Along the flow path, the liquid velocity is high,
and bubbles will be break up to smaller
When the flow rate of gas is higher in
run 9, the bubbles are fluctuated at the first
55
EXPERIMENTAL INVESTIGATION OF HYDRODYNAMIC PHENOMENA IN…
measurement position. Therefore, the void
fraction at Front Sensor and Rear Sensor are
quite different. Along the flow path, the
bubble breakup to form smaller bubbles and
concentrates at the wall region (Figure 8).
Therefore, the void fraction in the wall region,
at two upper measuring positions are higher
than that of position measurement first.
(a)
(b)
Fig. 8. Local profile of void fraction for run 9: jf = 1 m/s, jg = 0.1 m/s
(a) Front Sensor; (b) Rear Sensor;
(a)
(b)
Fig. 9. Local profile of void fraction for run 13: jf = 0.5 m/s, jg = 0.2 m/s
(a) Front Sensor; (b) Rear Sensor;
The local profiles of Run 13 are given in
This experimental data was located near the
boundary between bubbly and slug flow
regime in Figure 5. Therefore, the difference
can be explained by the fluctuation of air flow
rate during the experiment, thus the flow
regime was shifted from bubbly to slug flow.
At the inlet position, bubbles coalescence to
Figure. 9. In this flow condition, the major
bubble shape is a slug and they cover the entire
flow channel. Thus, group void distribution
matches the shape of a slug bubble. However,
under the same flow condition, the bubbly
regime was recorded in previous study [15].
56
DINH ANH TUAN et al.
form larger bubbles. At second measuring
mechanism starts to contribute to the number
of small bubbles near-wall region. According
to shear off mechanism, when slug bubbles are
large enough, they are sheared at the rim, and
many small bubbles show up.
position, the void distribution is affected by
slug bubbles that small bubbles follow the slug
bubbles, distributing in the slug bubbles’ wake
regions. Besides, the effect of shear off
C. Bubble Velocity
(a)
(b)
(c)
Fig. 10. Local profile of bubble velocity:
(a) Run 4; (b) Run 9; (c) Run 13
In run 4 and run 9, the interfacial
enter the wake region, they will accelerate
and may collide with the leading one.
Therefore, the bubble velocity at first
position is slightly higher than the two
upper measuring positions. In addition, near
to the wall region, the velocity of the
bubbles is strongly fluctuated due to wall
friction and turbulent intensity.
velocity is distributed corresponding to the
single-phase velocity profile and this agrees
with Hibiki [18]. The liquid velocity profile
is flattened when the gas is added. As
shown in Figure 10, the value of bubble
velocity is approximately equal to the sum
of superficial velocities. When bubbles
57
EXPERIMENTAL INVESTIGATION OF HYDRODYNAMIC PHENOMENA IN…
In Run 13, the channel-averaged velocity
at the higher measuring positions are gradually
decreased and are distributed almost uniformly
in the radius. This phenomenon occurs in the
case of low water superficial velocity, and drag
force is domination. As the bubble grows, the
bubble is normally accelerated by the effect of
buoyancy force. However, due to the effect of
drag force, velocity in run 13 is suitable.
be helpful to study and comprehend the
fundamental two-phase flow phenomena. For
practical application in nuclear reactor safety,
they are envisaged to be further improved and
applied to verification and validation of
calculation methods and codes, e.g., CFD
codes, in modeling of two-phase flow.
NOMENCLATURE
D – pipe diameter
g – acceleration of gravity
L – length
lchord – chord length
N – number of bubble
p – pressure
t – time
tdelay – delay time
u – velocity
V. CONCLUSIONS
The experimental investigation on local
interfacial parameters for vertical upward air-
water two-phase flow was performed in this
study. Two-sensor conductivity probe was used
for the measurement of 25 flow conditions that
covers from bubbly flow to slug flow. The
local parameters included are a time-averaged
void fraction and bubble interfacial velocity.
The data acquisition frequency of 10 kHz and
the sampling time of 60 s were applied.
V – Volume
Greek Symbols
α – void fraction
µ – viscosity
From the local experimental results, the
profiles of void fraction and interfacial velocity
along the axial and radial of the flow channel
were discussed in detail. The bubble
interaction mechanisms caused the differences
in local parameter distribution. For high liquid
superficial velocity (jf), the effect of buoyancy
force is dominant, and bubble break up
phenomena is observed. On the contrary, for
low liquid superficial velocity (jf), the effect of
drag force is dominant and bubble interaction
mechanism will change from bubble break up
to bubble coalescence.
ρ – density
σ – deviation standard
Ω – total sampling time
Subscripts
b – bubble
f – liquid
g – gas
ACKNOWLEDGEMENT
This work was performed at Institute of
Nuclear Science and Technology, Vietnam
Atomic Energy Institute under the project
“Construction of Thermal-hydraulic Experiment
and Investigation of Hydrodynamic Phenomena
When changing the flow conditions, the
void fraction distribution changes from wall
peaking at Run 4 and Run 9 to center peak at
Run 13. The reason might be due to the
uncertainties in the measured results that need
to be carefully considered and reduced in
future works. In conclusion, the test facility,
the experimental methods, and the preliminary
experimental results obtained in this study can
of
Two-phase
Flow”,
grant
code
DTCB.17/16/VKHKTHN. The authors would
like to thank Ministry of Science and
Technology for their support on this project.
REFERENCES
[1]. L.S. Tong and Y.S. Tang., Boiling Heat
Transfer And Two-Phase Flow (Series in
58
DINH ANH TUAN et al.
Chemical and Mechanical Engineering), CRC
[10].W.H. Leung, S.T. Revankar, Y. Ishii and M.
Ishii., Axial development of interfacial area
and void concentration profiles measured by
double-sensor probe method, Int. J. Heat Mass
Transfer (38) 445-453, 1995.
Press (book), 1997.
[2]. L.G. Neal and S.G. Bankoff., A high resolution
resistivity probe for determination of local
void properties in gas-liquid flow. AIChEJ (9)
490-494, 1963.
[11].X. Zhao, G. Lucas, and S. Pradhan., Signal
Processing method in using four-sensor probe
for measuring the velocity vector of air-liquid
two phase flow. Journal of the Japanese
Society of Experimental Mechanics (JSEM)
(9) 19-24, 2009.
[3]. M. Walter., Study on interfacial area transport in
vertical bubbly flows, Ingenieur Thesis, Korea
Atomic Energy Research Institute, Korea, 2008.
[4]. A. Manera, B. Ozar, S. Paranjape, M. Ishii, H.-
M. Prasser., Comparison between wire-mesh
sensors and conductive needle-probes for
measurements of two-phase flow parameters,
Nuclear Engineering and Design (239) 1718–
1724, 2009.
[12].B.J. Yun, K.H. Kim, G.C. Park, C.H. Chung,,
A study on the Measurement of Local Void
Fraction. Journal of the Korean Nuclear
Society 24(2) 168-177, 1992.
[13].D.J. Euh, B.J. Yun, C.H. Song, T.S. Kwon,
M.K. Chung, U.C. Lee., Development of the
five-sensor conductivity probe method for the
[5]. Z. Dang, G. Wang, P. Ju, X. Yang, R. Bean,
M. Ishii, S. Bajorek, M. Bernard.,
Experimental
study
of
interfacial
measurement
concentration, Nuclear Engineering and
Design (205) 35-51, 2001.
of
the
interfacial
are
characteristics of vertical upward air-water
two-phase flow in 25.4 mm ID round pipe.
International Journal of Heat and Mass
Transfer (108) 1825–1838, 2017.
[14].I. Kataoka and A. Serizawa., Averaged bubble
diameter and interfacial area in bubbly flow,
Proc. 5th Two-phase Symp. Japan, Kobe,
Japan, 28-29 November 77-80, 1984.
[6]. K. Sekoguchi, H. Fukui, M. Tsutsui and K.
Nishikawa., Investigation into the statistical
characteristics of bubbles in two-phase flow
(2nd Report, Application and establishment of
electrical resistivity probe method), Bull.
JSME (18) 397-404, 1975.
[15].Y. Taitel, D. Bornea and A. E. Dukler.,
Modelling Flow Pattern Transitions for Steady
Upward Gas-Liquid Flow in Vertical Tubes,
AlChE Journal (Vol. 26, No. 3) 345–354, 1980.
[7]. M.S. Hoffer and W. Resnick., A modified
electro resistivity probe technique for steady-
and unsteady-state measurements in fine
dispersions – I. Hardware and practical aspects,
Chem. Engng. Sci. (30) 473-480, 1975.
[16].M. Ishii and S. Kim., Micro four-sensor probe
measurement of interfacial area transport for
bubbly flow in round pipes. Nuclear
Engineering and Design 205 123-131, 2001.
[8]. J.M. Burgess and P.H. Calderbank., The
measurement of bubble parameters in two-
phase dispersions-I. The development of an
improved probe techniques, Chem. Engng Sci.
(30) 743-750, 1975.
[17].M.
McCreary.,
Design
and
Testing
Conductivity Probes for the Measurement of
Two-Phase Flow Parameters. MS Thesis,
Oregon State University, US, 2001.
[18].T. Hibiki, M. Ishii, Z. Xiao., Axial interfacial
area transport of vertical bubbly flows,
International Journal of Heat and Mass
Transfer (44) 1869–1888, 2001.
[9]. M. Behnia, S.J., Gillespie., Void fraction
measurement by a computerized double-point
resistivity probe. Flow. Meas. Instrum. (2)
243-247, 1991.
59
11 trang
10800
Free
Bạn đang xem tài liệu "Experimental investigation of hydrodynamic phenomena in vertical-upward adiabatic two-phase flow conditions", để tải tài liệu gốc về máy hãy click vào nút Download ở trên
File đính kèm:
- experimental_investigation_of_hydrodynamic_phenomena_in_vert.pdf
