Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 11.3.2 - Phase Fraction Measurement in Multiphase Flow Using an Improved Electrical Method
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00282
 Material Information
Title: 11.3.2 - Phase Fraction Measurement in Multiphase Flow Using an Improved Electrical Method Experimental Methods for Multiphase Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Ye, J.
Feng, Y.
Guo, L.J.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: phase fraction
multiphase flow
electrical method
 Notes
Abstract: The measurement of fraction of each phase has been a problem of petroleum industry for many years. In this paper, two kinds of improved electrical probes are introduced to satisfy the experimental requirement. Conductance technique is applied to measure water fraction since water has significant electrical property from gas or oil. The self-developed double-ring probe is considered to measure water fraction. On the other hand, capacitance method is used to estimate the void fraction in gas-oil mixture. The capacitance system consists of a capacitance probe with two helical brass electrodes which are surface-mounted to the outer wall of the pipe and an advanced electric circuit. The calibration results shows that the improved double-ring probe is not sensitive significantly to the phase distribution and has good dynamic resolution in gas-water two-phase flow. Besides that, the measurement error of double helical capacitance probe are around 5% in steady stratified flow, wavy stratified flow and 10% for slug flow in gas-oil two-phase flow. Furthermore, the water fraction is measured for oil-water by conductance method. As a result, it may be good for investigating interfacial structure characteristics especially slug flow and the propagation of gas/oil/water fraction waves. Meantime, a series of experiments are carried out in gas-water flow for void fraction wave investigation and in oil-water flow for phase inversion point identification, respectively, using double-ring conductance probe. Measurement of water fraction in gas-water-oil three-phase flow is performed to evaluate the response to water fraction of double helical capacitance probe in three-phase flow. The results confirm the practical value of the two probe system in both research work and industry.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00282
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1132-Ye-ICMF2010.pdf

Full Text

Paper No 7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


Phase Fraction Measurement in Multiphase Flow Using an Improved Electrical Method


J. Ye, Y Feng and L. J. Guo*

Xi'an Jiaotong University, State Key Laboratory of Multiphase Flow in Power Engineering
No.28, Xianning West Road, Xi'an, Shaanxi, 710049, P. R. China
*E-mail: lj-guo@mail.xjtu.edu.cn


Keywords: Phase fraction, multiphase flow, electrical method




Abstract

The measurement of fraction of each phase has been a problem of petroleum industry for many years. In this paper, two kinds
of improved electrical probes are introduced to satisfy the experimental requirement. Conductance technique is applied to
measure water fraction since water has significant electrical property from gas or oil. The self-developed double-ring probe is
considered to measure water fraction. On the other hand, capacitance method is used to estimate the void fraction in gas-oil
mixture. The capacitance system consists of a capacitance probe with two helical brass electrodes which are surface-mounted
to the outer wall of the pipe and an advanced electric circuit. The calibration results shows that the improved double-ring probe
is not sensitive significantly to the phase distribution and has good dynamic resolution in gas-water two-phase flow. Besides
that, the measurement error of double helical capacitance probe are around 5% in steady stratified flow, wavy stratified flow
and 10% for slug flow in gas-oil two-phase flow. Furthermore, the water fraction is measured for oil-water by conductance
method. As a result, it may be good for investigating interfacial structure characteristics especially slug flow and the
propagation of gas/oil/water fraction waves. Meantime, a series of experiments are carried out in gas-water flow for void
fraction wave investigation and in oil-water flow for phase inversion point identification, respectively, using double-ring
conductance probe. Measurement of water fraction in gas-water-oil three-phase flow is performed to evaluate the response to
water fraction of double helical capacitance probe in three-phase flow. The results confirm the practical value of the two probe
system in both research work and industry.


Introduction

Multiphase flow exits in petroleum and chemical industry
applications and a great deal of problem needs to be solved,
one of which is phase fraction measurement. Because of the
complex phenomenon in multiphase flow, it is important to
find out that the average cross-sectional phase fraction in
the flow. For instance, to learn how gas-water interface form
and develop, void fraction measurement is needed (Fossa
1998). However, when there are two or three phases
contained in the pipeline, the flow structure and phase
distribution have great difference from single phase flow
and they are so complicated that it is hard to determine the
phase fraction. There is little information available in
literature about measurement in three-phase flow. Besides
that, a real-time on-line instrument is also the requirement
of multiphase measurement. From the scientific view, fast
measuring techniques are becoming more and more
important for the experimental investigation and modelling
of multiphase flow (Silva et al. 2007).
Electrical technique consisted of conductance and
capacitance method is based on the principle that each phase
has different electrical conductivity or relative permittivity.
Electrical technique can be high frequency response, low
cost and easy to install, although it strongly depend on the
type of the probe used and liquid or gas material to be
investigated.


According to Coney (1973), the theory of conductance
probe which could solve the electrical problem of flat or
circle electrodes contacting with a layer of liquid film had
established based on Laplace equation and conformal
transformation. In early studies, using conductance
technique, Asali et al. (1985), Andressi et al. (1988), Fossa
(1998), Angeli & Hewitt (2000), Gu & Guo (2008) had
successfully employed ring or wire electrodes for liquid
film measurement, respectively. To eliminate temperature
errors, Kim et al. (2009) used a three-ring conductance
probe. Merilo et al. (1977) proposed to use a rotational
electrical field with three groups' electrodes in order to
minimize the void distribution effect.
Capacitance method can be used when the flow media are
not electro conductive and unlike conductance probes, these
kinds of probes do not need to contact with the fluids. Shu
et al. (1982) employed a simple plate capacitance sensor for
void fraction Measurement in gas-water two-phase flow and
obtained good agreement to the prediction. Elkow &
Rezkallah (1996) got the void fraction in gas-water flow
combining a helical capacitance sensor with a concave plate
probe. Besides these, Tollefsen & Hammer (1998) studied
the helical capacitance sensor to reduce errors in phase
concentration measurement. They found out that when
relative permittivity was close enough, for example gas and
oil, the sensor was not sensitivity to the flow pattern.
However, Ahmed (2006) determined the signal-to-noise






Paper No


ratio, sensitivity, and time response of the ring capacitance
sensors in gas-oil two-phase flow. It was found that
ring-type sensors were more sensitive to the void-fraction
signal than the concave type for the same spatial resolution.
In this paper, both conductance technique and capacitance
method are applied to measure phase fraction. The two
probes are carefully calibrated before use. For conductance
technique, a new size of double-ring probe and double
helical capacitance probe are designed. Calibrations are
performed to measure phase fraction in gas-water, oil-water
and gas-oil two-phase flow. In the end, application
experiments are carried out to verify the practical value of
the two probes.

Nomenclature


gravitational constant (in -i
inner diameter of the pipe (mm)
dimensionless voltage
normalized voltage
capacitance (F)
normalized capacitance
velocity (m/s)


Greek letters
a void fraction
(p water fraction
p density (kg/m3)
p kinetic viscosity (Pas)

Subscripts
g gas
o oil
w water
W void fraction wave
mix mixture
s superficial


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


Experimental Facility and Instrument

The experiment was carried out in the State Key Laboratory
of Multiphase Flow in Power Engineering in Xi'an Jiaotong
University. The facility is shown in Figure 1. The working
fluids are air, tap water and LP-14 white oil. Air is
compressed by a dual screw rod compressor into a buffer
vessel and then measured by a set of orifice system after
filtering. Water is pumped from a water tank and then
measured by an electromagnetic flow meter (accuracy
0.5%). Oil is pumped by a gear pump from an oil tank and
then measured by a mass flow meter (accuracy 0.1%). In
horizontal test section, the three phases are mixed in a mixer
in which a horizontal plate is fixed in order to get stratified
flow at the initial part of the test section. In vertical part, the
water is measured by mass flow meter through the by-pass
valves and then the gas-water two phases are mixed in a
mixer chamber in order to form steady bubble flow, as
shown in Figure 2. The mixture is separated in a separator
which is open to atmosphere. The air is vented to
atmosphere and the liquid is returned to the three phase
separator tank, from which oil and water flow back to their
storage tanks respectively.
The test section of experimental facility consists of two
parts. One is a 16.5m long horizontal pipe and the other is a
2.58m height riser. All segments of the test section are made
of plexiglass for visual observation. The inner diameter of
the test section is 50mm.
A high speed multifunction board NI PCI-6259M connected
with a PC is used to do data acquisition. The data
acquisition and analyzing program is developed under the
environment of G-programming system of LabView 8.2.
The sampling frequency is 2000Hz for all the signals and
sampling time is 120s.


Figure 1: Schematic diagram of experimental facility
1-water tank; 2-oil tank; 3-centrifugal pump; 4-gear pump; 5-air compressor of dual screw rod; 6-magnetic flowmeter;
7-mas flowmeter; 8-orifice flowmeter; 9-gas-water mixing chamber; 10-three phase mixer; 11-three-phase separator






Paper No


A f H
A air inlet




1 water inlet
Figure 2: Gas-water mixer chamber

Probe system Design

The system consists of a pair of double-ring conductance
probes and a double helical capacitance probe and proper
electronic circuits for each probe.

Conductance probe system

The principle of conductance probe is that gas or oil phase is
poor conductor while water phase is good one. When
conductance probes contact with water, the current of the
loop is great outputting high voltage level. This is contrary
to that the probes contact with gas or oil. So, average
cross-section void fraction can be measured by the
conductance probe. As two fluids flow through the
conductance probe alternatively, the instrument will show a
continue change of voltage signals over time.
A pair of double-ring conductance probe was flush-mounted
in both horizontal and vertical test section, away from the
inlet 13.72m long and 1.96m high, respectively. Figure 3
shows the schematic diagram of the double-ring
conductance probe used in this paper. The electrodes were
made of chrome-plated brass ring in order to prevent
corrosion and rust. The width of the ring was 3.65mm and
the distance between the two rings was 16.5mm. The pair of
ring electrodes then was fastened by an insulation tube with
diameter equal to the outer diameter of the test pipe. On
vertical section, there were two pairs of probe with 100mm
away from each other.

G 073D


C.33D




Figure 3: Schematic diagram of double-ring
conductance probe

In this study, the double-ring probe was excited by
sinusoidal AC. The frequency was 6 kHz which was
proposed by Guo (2002). The output of the probe was AC
signals, so a signal conditioning circuit was needed. The
measurement circuit consisted of four parts: current-voltage
conversion circuit, full-wave rectifier circuit, second-order
low-pass filter circuit and signal amplification circuit. After
passing the four parts, the signal was associated with the


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

void fraction or water fraction.

Capacitance probe system

Different type of capacitance probe has strongly effect on
the electric field, thereby affecting the value of the
capacitance. In capacitance method, the two phase mixture
is considered as dielectric. At a certain distance between the
two electrodes, changes in phase fraction will lead to change
of the dielectric, the value of capacitance then also change.
As Tollefsen & Hammer (1998) suggested, double helical
capacitance probe was used in this study, as shown in Figure
4. Two plates of spiral brass electrodes twisted on the pipe,
whose length was 200mm, about 4 times of the inner
diameter. The electrodes could not be too short; otherwise it
would not lead to uniform electric field affecting
measurement accuracy. The width and the thickness of the
electrodes were 63mm and 1.2mm, respectively. The twisted
angle of the electrode was 180 degree. 2mm thick layer of
insulating material was deposited on outside of the
electrodes and then a layer of shielding material was
covered.


elecides


-Ol


R ,,- S- e layer



Figure 4 Schematic of helical capacitance probe

For gas-oil flow, the value of capacitance when the pipe was
full of oil was nearly 8 pF, while 5 pF when there were
nothing but air. It was so small value that a very high
precision circuit for measuring weak capacitance was
necessary. In this study, a chip AD7746 was used to
determine the weak capacitance, whose output was digital
signal. Trough the SDA, the signal was sent to a measuring
instrument based on Easy-ARM development board, which
was designed and manufactured in our laboratory. The
signal was converted to analog voltage signal. For flows
contained water, because of the very different permittivity
of water from air and oil, the output capacitance value of
flows with water was significant lager than that with air or
oil. To measure the water fraction of gas-water or
gas-oil-water flow, another measurement circuit based on
the chip CAV424 was designed. The output was voltage
signal. Finally, all the voltage signals were sent to a PC,
saved and processed.


Experimental Calibration

In this study, calibrations of the probes consisted of both
static and dynamic method. During static calibration, liquid
was increasingly added into the test section forming a
stratified state and the outputs of the voltage and how much
water had been added were recorded manually. A number of
tests conducted with the probes were done in order to test
the repeatability and hysteresis characteristics of the probes.






Paper No


Quick-closing valve was used for exact water fraction in
dynamic calibrations, which was installed on the vertical
section and horizontal section for double-ring conductance
probe and double helical capacitance probe, respectively,
see Figure 5 and Figure 6. The preliminary static and
dynamic calibrations showed that the output was nonlinear.
In addition, the sensitivity of the probe had a little different
between a static test and a dynamic test.

4--E

Water outlet



I5 I
8 -





S Mixing
chamber
Air inlet


t Water inlet
Figure 5: Schematic diagram of dynamic calibration of
double-ring conductance probe
Drain hole
J
Helical capacitance probe


Quick-closing valve U I



Figure 6: Schematic diagram of dynamic calibration of
double helical capacitance probe

Calibration result for double-ring conductance probe

The calibration result of gas-water two-phase flow showed
that the repeatability and hysteresis characteristics of the
probe were better. The plots of void fraction of gas-water
two-phase flow from the static method as a function of the
dimensionless voltage output from the double-ring
conductance probe can be seen in Figure 7.


1.0

0.8

0.6

0
L 0.4
0


0.2 0.4 0.6
Dimensionless Voltage (U*)


0.8 1.0


Figure 7: Void fraction measured as a function of
dimensionless voltage for double-ring conductance probe


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

The fitted curve showed that the probe was nonlinear
response and void fraction was satisfied to the following
equation:
a = 0.98 + 0.20U -2.71U *2 +1.53U *3 (1)
where a is void fraction and U* is dimensionless output
voltage.
During dynamic calibration, a bubble flow state was
selected, because bubble flow is the most homogeneous
flow pattern. The calculated water fractions by equation (1)
in static calibration were compared to the measured values
using quick-closing valve. The results of this comparison
are shown in Figure 8 where it can be seen that most of the
measured data are within + 5% of the actual values. Since
the probes were calibrated in stratified and bubble flow,
which are the most inhomogeneous and homogeneous,
respectively, very small difference had been found between
them. As a result, in this work, the output of the double-ring
conductance probe was likely not very sensitive to the flow
pattern in gas-water two-phase flow.


0.8

0.6
n

o 0.4

0.2
t 0.2


02 04 06
a (dynamic calibration)


08 10


Figure 8: Calibrated water fraction from the double-ring
conductance probe as a function of the water fraction
measured by quick-closing valve method in gas-water flow



1.0
+10% -
0.8 -
c .'_' -'- -10%

.o_
+s []o


S0.2 -"


0.0
0.0 02 04 06 08 10
qp (dynamic calibration)

Figure 9: Calculated water fraction as a function of the
water fraction by quick-closing valve in oil-water flow

To detect the response of the probe in oil-water flow, a
number of dynamic calibration measurements were done in
many sets of oil and water velocities. Measured water
fraction by quick-closing valve was compared to the
calculated water fraction using equation (1) which was
obtained in stratified situation of gas-water flow, the same


+5%,




-.






Paper No


with that in oil-water stratified flow. The result is shown in
Figure 9. It can be seen that most of the calculated data were
within + 10% of the measured values at high value of water
fraction. In addition, some plots were away from -10%. This
was considered to be due to the transition of flow structure
from stratified flow or oil-in-water flow to water-in-oil flow
which was observed in the experiment at this water and oil
velocities. This phenomenon is called phase inversion.
Phase inversion is the phenomenon whereby the phases of a
dispersion of two immiscible liquids spontaneously
interchange under conditions determined by the properties,
phase volume hold-up and energy input (Yeo 2002).
Although oil-water stratified calibration was almost the
same with gas-water stratified calibration, there were some
differences in oil-water dispersed flow. Because oil-water
flow is more complex than gas-water flow and conductance
technology is an intrusive method, so the double-ring
conductance probe has some difficulties to deal with the
measurement in oil-water two-phase dispersed flow. But
when the flow structure starts to change, the output of the
probe will have significant difference, especially during
water-in-oil flow changes to oil-in-water flow or opposite
situation. This is a great help of identifying the phase
inversion phenomenon. Figure 10 shows the static
calibration of oil-water emulsion. An instantaneous value
was recorded when water and oil were strongly mixed
during calibration. As shown in Figure 10, at a water
fraction of about 0.3, the output voltage of the probe was
approximately zero. This could be attributed to the transition
of flow structure, which was varied from oil-in-water flow
to water-in-oil flow. In water-in-oil flow, water as dispersed
phase was diffused in the oil and oil became continues phase.
The conductance probe contacted with oil, so the output was
zero. It is also can be found that water fraction in oil-water
flow had somewhat liner response to the dimensionless
voltage.


0.2 0.4 0.6
Dimensionless Voltage (U*)


0.8 1.0


Figure 10: Static calibration for oil-water emulsion

Calibration result for double helical capacitance probe

Calibrations of the double helical capacitance probe were
done in gas-oil and oil-water flow. In general, the response
was much more linear. However, the precision of calculated
water fraction compared with measured water fraction using
quick-closing valve is not very ideal.
The result of static calibration of double helical probe in
gas-oil flow is shown in Figure 11. The fitted curved was


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

obtained by the following equation:
a = 0.02 + 0.29C +1.76C *2 -1.07C *3 (2)
where a is void fraction and C* is normalized capacitance
which followed by the equation:
C -C
C* = C (3)
Co Cg
C is the measured capacitance, Co is the capacitance of all
oil in pipe and Cg is the capacitance of all air in pipe. All the
capacitance values are transformed outputting digital value
by AD7746. The linearity is 5.15%.


0.e
0
LL 0.4
:2
0


Normalized Capactance C*


Figure 11: Calibration result of the double helical
capacitance probe in gas-oil flow


1.0

0.8

S0.6

S0.4
0.2
S0.2


0.0 02 04 06 08 10
a (quick-closing valve value)

Figure 12: The comparison between calculated void
fraction and quick-closing valve value in gas-oil flow

Figure 12 shows the deviation between the calculated void
fraction using equation (2) and the measured void fraction.
It can be found that in different flow patterns, the deviation
of the calculated and measured value was with 10%. Most
of the value measured in stratified flow was of high
precision and nearly with 5% accuracy. However, some
plots in slug flow were beyond 10%. This is because that the
phase distribution is very different from the stratified flow
and liquid slug and elongated bubble within the length of
the test section can be formed in this situation. When liquid
slug and elongated bubble pass the probe alternatively, the
instantaneous value obtained by the probe is not exact the
average value of the two parts of the void fraction, and
thereby a little higher than that measured by quick-closing
valve. Bur finally, in a certain way, the effect of flow
structure can be ignored in gas-oil flow.


oe
ra
rS
I0
1S






Paper No


Oil-water flow measurement using double helical
capacitance probe was also performed. Both static and
dynamic calibrations were done to estimate the response of
the probe to the water fraction. Compared with double-ring
conductance probe, phase inversion phenomenon might not
cause the output to be zero, as Figure 13 shows. It can be
found that in the stratified flow the dynamic response also
had good agreement with the static calibration. However,
the difference between static and dynamic calibration was
significant obvious in oil-water dispersed flow. The
measured water fractions were seriously drifted off the static
calibration line. The concave upward part presented the
oil-in-water flow while the concave downward part
presented the water-in-oil flow. As a result, different flow
patterns need different treatment in oil-water flow using
capacitance method.


02 04 06
Normalized Voltage V*


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

velocity and mixture velocity. It is interested to find that the
wave velocity was linear with gas superficial velocity while
it had a nonlinear response to water superficial velocity. And
a linear relationship of wave velocity and the mixture
velocity could be expressed by the following equation using
drift flux model:
V = 1.16U_ +0.67ggD (6)
where V, is the void fraction wave velocity and Umi is the
velocity of gas-water mixture.


20
S4 Usg=0 24m/s
1 o Usg=0.35m/s
"12 1 A Usg=0 5m/s

514
10o. 0
12
a Uswo010nis
AA Uswm0.40mis s
'0 0.2 0.4 0.6 o0 02 0 o4 06 0.8 10
Usg (1nV) Usw (mIs)


00 02 04 06 08
Umix (rnfs)


1.0 12 1.4


Figure 14: Variation of void fraction wave velocity


08 10


Figure 13: Calibration of the capacitance probe in
oil-water flow



Application Experiments and Results

Void fraction wave measurement

The propagation of void fraction in two-phase flow will
form void fraction wave, which is also named density wave
or concentration wave. It can be expressed by the
fluctuation of void fraction varied by time. The problem
about fraction wave is that whether fraction wave is a
dispersion wave or not has not reach a common
understanding. Void fraction wave was measured in this
study using double-ring conductance probe in gas-water
vertical flow to determine the wave velocity and main
frequency characteristics.
A pair of double-ring conductance probe apart from each
other 10cm was used to collect the void fraction signals.
Transition time to of the two signals detected by the two
probes was obtained by cross-correlation method in the
limited time as following equation shows:

R ,()= Y(t)-X(t-r)dr (4)

So the velocity of the void fraction wave will be obtained
using following equation:
v =L/to (5)

where L is the distance between the two probes.
Figure 14 shows the variation of calculated void fraction
wave velocity with gas superficial velocity, water superficial


0 20 40 60 80 100
Frequency (Hz)
Figure 15: FFT spectrum diagram of voltage signals


0


0.6 0.8 1.0 1.2 1.4 1.6
Wave velocity (m/s)


18 2.0


Figure 16: Main frequency of void fraction wave as a
function of wave velocity


^0




'" Static calibration
o Dynamic calibration for stratified flow
SA Dynamic calibration flor dispersed flow


a Usw=0.lmfs
o Usw=0.2mVs
A Usw=0.4rrs
9 Usw0.8mins





---,,--n-a
0 g
"o u- no






Paper No


The main frequency of the void fraction wave was obtained
using the method shown in Figure 15. The filtered voltage
signals subtracted the average value were processed by FFT.
The peak on the spectrum diagram was considered to be the
main frequency of the void fraction wave. Figure 16 shows
a primary study on the changes of main frequency as a
function of wave velocity in different water superficial
velocities. It can be seen that the points were divided into
three parts. An initially understanding of the diagram could
be that the main frequency of the wave velocity was nearly
the same when water superficial velocity was low
(Uw,=0.lm/s, Usw=0.2m/s), and increased in a certain water
superficial velocity, but remained unchanged (Uw=0.4m/s),
and finally jumped to a higher frequency becoming unstable.
This may imply that the void fraction wave is a slightly
dispersive wave which was also reported by Matuszkiewicz
(1987).


Phase inversion point identification

As mentioned before, phase inversion is a phenomenon that
the continuous phase and dispersed phase interchange
abruptly in two immiscible liquid. When phase inversion
occurs, the theological characteristics and pressure drop will
change significantly. Phase inversion point is defined as the
critical input water fraction when the dispersed phase
become continuous phase, which may be a major factor to
be considered in the pipelines design.
As Figure 10 shows, phase inversion phenomenon could be
found by double-ring conductance probe which was used to
study the oil-water two-phase horizontal flow. To determine
the phase inversion point clearly, the trend of the
dimensionless output voltage variation is shown in Figure
17, where water superficial velocity was 0.15m/s, while oil
superficial velocity was from 0.70m/s to l.Om/s. It can be
found that, when oil velocity was lower than 0.90m/s, the
fluctuations of the dimensionless voltage are relatively large,
indicating that water phase maintained a continues state in a
sort of sense. When oil velocity was higher than 0.9m/s, for
example when it was 1.Om/s, the output dimensionless
voltage remained zero and the fluctuation almost
disappeared. So according to the static calibration in Figure
10, we can imagine that phase inversion has occurred. Water
phase has dispersed as droplets in continuous oil phase.

Uso=o.7mls
02 WA k,,Ak JW,1--1Li1LCIWIAI~ll~l


S
c
(U
0
C


o02 uso=o.mrs 20 30 40 50
S0.1 r L, 'Vi4r U1
0.0


10 20 30 40 50
0.2 Uso=0.9mls

I 10 20 30 40 50
021 Uso=0.ml/s
0.1

0.0
0 10 20 30 40 50
02 Uso=1.0m/s
0.1-

0 10 20 30 40 50


Time (s)
Figure 17: Dimensionless voltage output for different oil
superficial velocities


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010


The variation of averaged dimensionless voltage with oil
superficial velocity as shown in Figure 18 was trying to find
out the phase inversion point. Every fitted curve had a point
that the dimensionless voltage was 0, for example, when
Uw=0.15m/s, the corresponding value of 0 was
Uo=0.935m/s. As mentioned above, Uo=0.935m/s could be
regarded as the critical oil superficial velocity at
Uw=0.15m/s when phase inversion occurred. It can be also
found that as water superficial velocity increased, the
critical oil superficial velocity also increased.


0.0 02 04 06 08 1.0 12
Uso (m/s)

Figure 18: Output dimensionless voltage as a function of
oil superficial velocity

Table 1 show the comparison of critical input water fraction
with other researchers' data available from literature when
phase inversion occurs. These data were obtained for
oil-water flow in pipes of D=50mm, pw=998kg/m3,
po=835.4kg/m3, /w=lmPas, ou=18.8mPas, U,=0.15m/s,
Uo=0.935m/s. This implies that the conductance probe
method can identify the phase inversion point reasonably. It
supplies a simple and easy method to account for a complex
phenomenon.

Table: Comparison of phase inversion point

Lit e Critical input
Literature
water fraction
Present study 13.8 %
Nadle & Mewes (1997) 23.9 %
Yeh(1964) 18.8 %
Arirachakaran (1989) 35.9 %
Decarre & Fabre (1997) 58.4 %
Brauner & Ullmann (2002) 25.6%

Water fraction measurement in three-phase flow

Preliminary exploration on water fraction measurement was
carried out in gas-oil-water three-phase flow. Thinking
about that the permittivity of water is much larger than air
and oil, air and oil were considered as one phase.
Three-phase flow was simplified to two-phase flow with a
great difference of permittivity constant of the two phases.
Double helical capacitance probe could be used to measure
the water fraction of the three-phase flow approximately.
Three kinds of oil-water ratio were selected, which were 1:3,
1:1 and 3:1, respectively. Static calibration was done in
gas-oil-water stratified state. Figure 19 shows the result of






Paper No


calibration. To compare three calibration results, a new
normalized voltage was calculated as follows:

V* (7)
015 g
where Vg is the output voltage when the pipe full of air, VO 15
is the output voltage when the water accounts for 15% of the
total volume which is the sum of air, water and oil. Doing
this was to facilitate combining the three results together. It
can be found that three results matched well, especially
when total water fraction was between 0.05 and 0.25 and
generally were consistent with the same law. Seemingly, the
normalized voltage was not sensitive to the oil-water ratio
and only concerned with water fraction. This implies that
the hypothesis with considering air and oil as one phase is
reasonable. However, with further study on practical use in
three-phase flow, the results were not encouraging, see
Figure 20. The complicated three-phase flow resulted in that
the error of the two methods was much lager. To study
measuring water fraction of multiphase flow needs deeply
exploration.

1.0
= 75% water fraction of oil-water mixture
o 50% water fraction of oil-water mixture
A 25% water fraction of oil-water mixture
0.8

C
2 0.6

a0.4

i- 0.2

0.0 A
0.0 0.5 1.0 1.5 2.0 2.5
Normalized Voltage V'=(Vx-Vair)/(Vo.is-Vair)
Figure 19: Static calibration of gas-oil-water three-phase
flow using double helical capacitance probe

0.7

0.6

>0.5 -

S0.4

E 0.3
E
0.32




0.0 X1 -
0.0 0.1 0.2 0.3 0.4 0.5 0,6 0.7
(p (quick-closing valve value)
Figure 20: Comparison of water fraction between static
and dynamic calibrations in gas-oil-water three-phase flow


Conclusions

The system including probes and proper circuits for water
fraction measurement was presented in this paper. Two
types of probes, which were the double-ring conductance


7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

probe and double helical capacitance probe, had been
designed and calibrated in gas-water, gas-oil and oil-water
flows. The calibration results showed that the double-ring
conductance probe was not sensitive to the flow pattern
including bubble flow and slug flow in gas-water flow and
the accuracy could reach 5%. Nevertheless, the accuracy
was around 10% or more, when it was used to detected
water fraction in oil-water stratified or dispersed flow. The
response to void fraction in gas-oil flow was better with
accuracy of about 10% when double helical capacitance
probe was used. However, in oil-water flow, the dynamic
calibration result could not match the static calibration result
very well. But by using quick-closing valve, it was possible
to measure water fraction using a new calibration curve.
Three experiments were performed to verify the practicality
of the two probes. Void fraction wave characteristics and
phase inversion point were investigated using double-ring
conductance probe. In general, this kind of probe was better
sensitive to water fraction variations, and was likely to be
more suitable for evaluating the mean void fraction. Water
fraction measurement was taken preliminary to test the
sensitivity of double helical capacitance probe, which were
only associated with water. However, it would be a long
way to use this probe practically in experiment and industry.


Acknowledgements

The study has been supported by National Natural Science
Foundation of China (Grant No. 50823002 and No.
50536020). We thank the referees of this paper for their
valuable suggestions to improve the quality of this paper.

References

Ahmed, W.A. Capacitance sensors for void-fraction
measurements and flow-pattern identification in air-oil
two-phase flow. IEEE Sensor Journal, Vol. 6,
1153-1163(2006)

Arirachakam, S., Oglesby, K.D., Malinowsky, M.S.,
Shoham, O. & Brill, J.P. An analysis of oil/water flow
phenomena in horizontal pipes. SPE, 155-167(1989)

Asali, J.C. Hanratty, T.J. & Andreussi, P. Interfacial drag
and film height for vertical annular flow. AIChe. Journal,
Vol. 31, 895-902(1985)

Andreussi, P. Di Dofrancesco A. & Messia, M. An
impedance method for the measurement of liquid hold-up in
two phase flow. International Journal of Multiphase Flow,
Vol. 14, 777-785(1988)

Angeli, P., Hewitt, GF. Flow structure in horizontal
oil-water flow. International Journal of Multiphase Flow,
Vol. 26, 1117-114' ?i,2'i1

Brauner, N. & Ullmann, A. Modeling of phase inversion
phenomenon in two-phase pipe flows. International Journal
of Multiphase Flow, Vol. 28, 1177-1204(2002)

Coney, M.W.E. The theory and application of conductance
probes for the measurement of liquid film thickness in two






Paper No 7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30-June 4, 2010

phase flow. Journal of Physics E: Scientific Instruments, Vol.
6, 903-1002(1973) Yeh, GC. Phase-volume relationship at the point of phase
inversion in liquid dispersion. AIChE Journal, Vol. 10,
Da Silva, M.J., Schleicher, E. & Hampel, U. A novel needle 260-265(1964)
probe based on high-speed complex permittivity
measurements for investigation of dynamic fluid flows.
IEEE Transactions on Instrumentation and Measurement,
Vol. 56, 1249-12 5.,i "')

Decarre, S. & Favre, J. Phase inversion prediction study.
Journal of L' Institute Francais du Prtrole, Vol. 52,
415-424(1997)

Elkow, K.J. & Rezkallah, K.S. Void fraction measurements
in gas-liquid flows using capacitance sensors. Measurement
Science and Technology, Vol. 7,1153-1163(1996)

Fossa, M. Design and performance of a conductance probe
for measuring the liquid fraction in two-phase gas-liquid
flows. Flow Measurement and Instrumentation, Vol. 9,
103-109(1998)

Gu, H.Y & Guo, L.J. Experimental investigation of slug
development on horizontal two-phase flow. Chinese Journal
of Chemical Engineering, Vol. 16,171-177(2008)

Guo, L.J., Li, GJ. & Chen, X.J. A linear and non-linear
analysis on interfacial instability of gas-liquid two-phase
flow through a circular pipe. International Journal of Heat
and Mass Transfer, Vol. 45, 1525-1534(2002)

Kim, J., Ahn, YC. & Kim, M.H. Measurement of void
fraction and bubble speed of slug flow with three-ring
conductance probes. Flow Measurement and
Instrumentation, Vol. 20, 103-109(2009)

Matuszkiewicz, A. Flamand, J. & C Bour6, J.A. The
bubble-slug flow pattern transition and instabilities of void
fraction waves. International Journal of Multiphase Flow,
Vol. 13,199-217(1987)

Merilo, M. Dechene, R.L. & Cichowlas, W.M. Void fraction
measurements with a rotating electric field conductance
gauge. Journal of Heat Transfer, Vol. 99, 330-332(1977)

Nidler, M. & Mewes, D. Flow Induced Emulsification in
the Flow of Two Immiscible Liquids in Horizontal Pipes.
International Journal of Multiphase Flow, Vol. 23,
55-68(1997),

Shu, M.T, Weinberge, C.B. & Lee, YH. A simple
capacitance sensor for void fraction measurement in
two-phase flow. Industrial and Engineering Chemistry
Fundamentals, Vol. 21, 175-181(1982)

Tollefsen, J. & Hammer, E.A. Capacitance sensor design for
reducing errors in phase concentration measurements. Flow
Measurement and Instrumentation, Vol. 9, 25-32(1998)

Yeo, L.Y, Matar, O.K., De Ortiz, E.S.P. & Hewitt, GF. A
simple predictive tool for modelling phase inversion in
liquid-liquid dispersions. Chemical Engineering Science,
Vol. 57,1069-1072(2002)




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs