Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 14.3.2 - High sensitivity capacitance probe for liquid and gas dispersions in a conductive liquid
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00348
 Material Information
Title: 14.3.2 - High sensitivity capacitance probe for liquid and gas dispersions in a conductive liquid Experimental Methods for Multiphase Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Strazza, D.
Demori, M.
Ferrari, V.
Poesio, P.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: liquid-liquid
gas-liquid
dispersion
capacitance probe
 Notes
Abstract: A capacitance sensor for hold-up measurements in liquid-liquid and gas-liquid dispersion is presented. When the continuous phase is a conductive fluid sensitivity issues arise, moreover the conductive fluid give rise to capacitance couplings between the pipe and the metallic parts around it. Through a tailored electronic and proper designed guard electrodes the proposed sensor system is capable to detect variations of small amounts of the dispersed phase even if the continuous one is conductive. Preliminary liquid-liquid hold-up data are compared with homogeneous model ans Quick Closing Valves measurements showing promising results.
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: VID00348
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1432-Strazza-ICMF2010.pdf

Full Text



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


High sensitivity capacitance probe for liquid and gas dispersions in a conductive
liquid


D. Strazza*, M. Demorii, V. Ferrarii and P. Poesio*

Dipartimento di Ingegneria Meccanica e Industriale, Universith di Brescia, Via Branze 38, 25123, Brescia, Italy

t Dipartimento di Elettronica per l'Automazione, Universith di Brescia, Via Branze 38, 25123, Brescia, Italy
domenico.strazza~ing.unibs.it and pietro.poesio~ing.unibs.it

Keywords: Liquid-liquid, gas liquid, dispersion, capacitance probe




Abstract

A capacitance sensor for hold-up measurements in liquid-liquid and gas-liquid dispersion is presented. When the
continuous phase is a conductive fluid sensitivity issues arise, moreover the conductive fluid give rise to capacitance
couplings between the pipe and the metallic parts around it. Through a tailored electronic and proper designed guard
electrodes the proposed sensor system is capable to detect variations of small amounts of the dispersed phase even if
the continuous one is conductive. Preliminary liquid-liquid hold-up data are compared with homogeneous model ans
Quick Closing Valves measurements showing promising results.


Nomenclatu re


homogeneous one having effective properties Brauner
(2004). Two key aspects for the quantification of disper-
sions are drop (or bubble) size, distribution, and hold-up
(or void fraction for gas-liquid flows). While to mea-
sure drop size there are well known laser techniques -
Simmons & Azzopardi (2001) hold-up measurements
in dispersed flow are still under investigation. In some
cases, the amount of the dispersed phase is very small
(less than 5%) and measurement of such small quanti-
ties requires high sensitivity probes.
Different kind of probes have been manufactured for
the estimation of void fraction/hold-up and among them
there are nuclear sensors, conductance probes, and ca-
pacitance probes. Nuclear sensors combine non intru-
siveness and high sensitivity and so are the most suit-
able for dispersed flow investigation, see for example
Kumara et al. (2010). On the other hand this kind of
sensor involve safety issues and they are expensive.
Electrical probes (either conductive or capacitive) are
also used to measure the hold-up. Conductance probe
- Fossa (1998) have been widely used in gas-liquid
flows. They work when one of the two fluids is conduc-
tive and the hold-up is obtained by measuring the resis-
tance of the mixture inside the test section and compar-
ing it with the value of single fluids. Such probes are
relatively simple to manufacture and, since electrodes
are flush mounted on the inner pipe wall, they are al-
most non intrusive. On the other hand, they have low
sensitivity and the detection of small percentages of the


Symbol
D pipe inner diameter (m)
4 pipe inner area (m )
Q; volumetric flow rate of the i-phase (m3/s)
H; hold-up of the i-phase (-)
ui, sup. velocity of the i-phase = Q;/24 (m/s)
o conductivity (S/m)
,; relative permittivity of the i-phase (-)
C' Capacitance
R Resistance
Subscripts
c continuous phase
d dispersed phase
w water
ooil
g gas



Introduction

Dispersion phenomena are often encountered in indus-
trial and environmental applications. As an example a
dispersion of two immiscible liquids, where one of the
liquids is the continuous phase and the other is the dis-
persed one, is typically found in chemical and food pro-
cesses. In the modelization of such a flow regime it
is in general possible to consider the two-fluids as an







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


Glass box
QCV









two
phase
mixture


Figure 1: Sketch of the experimental set-up.


dispersed phase may prove tricky Van der Meulen et
al. (2009).
Capacitance probes have the advantage to be rela-
tively cheap and easy to install because the electrodes are
on the external part of the pipe, not in contact with the
fluids, and they are normally used for non-conductive
fluids.
The capacitance technique applied to conductive flu-
ids shows some problems; when the conductive liquid
is the continuous phase, small amounts of the dispersed
phase do not give rise to \ignlilk~.lll signal variations,
Caniere et al. (2008).
In a previous work, Demori et al. (2009) developed a
capacitance sensor for oil-conductive water core-annular
flow hold-up estimation. In this work the aim is to de-
velop a capacitance sensor for non-conductive and con-
ductive fluids capable to detect also small amounts of the
dispersed phase.


Experimental setup

The experimental facility consists of an L = 9 m long
glass pipe with internal diameter D = 0.022 m (L/D =
428). A sketch of the experimental facility is given in
Fig. 1.
The inlet devices (A and B in Fig. 1) allow to study
two-phase liquid-liquid and gas-liquid flows and also
three-phase flows. Through nozzle A oil, air or both flow
together vertically for 0.7 m before entering the horizon-
tal test pipe passing from nozzle B, where water is intro-
duced.
Oil has an 886 kg/m3 density and 0.9 Pa-s dynamic
viscosity. Oil and water flow rates are measured by a tur-
bine and a magnetic flow-meters respectively. Oil flow-
meter is specifically calibrated for high viscosity fluids.


Air flow rate is measured by a thermal mass flow meter,
Oil screw-type pump is controlled by a mechanical
reducer, water is supplied by a centrifugal pump con-
nected to a frequency inverter that assures control of the
pumped flow-rate. Pressure drop data are collected 6 m
downstream the injection point by a differential pressure
transducer (PD) on a 1.5 m tract of pipe. The capaci-
tance probe (CP in Fig. 1) is placed between the pressure
plugs of the pressure transducer. Quick Closing Valves
(QCV) are placed after the capacitance probe on the last
tract of the experimental pipe.
The pipe end is at atmospheric pressure; the fluids are
discharged into a storing vessel where water and oil sep-
arate by gravity in order to recirculate them to the en-
trance of the flow facility. A glass box is inserted to re-
duce optical distortion and to allow a correct observation
of the flow. All the movies and the pictures are collected
through the observation window.

Experimental procedure
Flow rates, pressure drops and the capacitance probe
voltage outputs are acquired for 10 s at the frequency
of 1 kHz. The capacitance output is averaged and from
the obtained value the predicted hold-up is computed.
The preliminary hold-up values measured with the ca-
pacitance probe have been compared with a reference
value obtained through QCV, see for example Oddie et
al. (2003). Because of the small quantities involved in
the measurement the obtain hold-up values from QCV it
is necessary to work with an high accuracy. The proce-
dure we adopted for oil in water dispersions is described
in the following:

*water flow-rate is set at a fixed value (u,, = 1.58
m/s);







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


Guard
section


Figure 2: Block diagram of the complete sensor system.


oil is injected from the minimum flow rate (us =
0.03 m/s);

observations and data acquisitions for the present
oil and water flow-rates are completed;

the flow mixture in a 1.5 m tract is trapped through
QCV and emptied in a beacker;

the beaker is weighted in a balance having a sen-
sitivity of 0.01 g. The obtained value is compared
to the weight of only water trapped and hold-up is
computed;

the procedure is repeated 3 times in order to check
data repeatability;

keeping fixed water flow-rate, oil flow-rate is in-
creased; observations, data acquisitions for the new
oil and water flow-rate pair are completed 3 times;

the previous step is repeated until a 20 %b oil in wa-
ter dispersion is reached;

Further discussions on the QCV method applied to
hold-up measurements for dispersion phenomena is pro-
vided in the Results section.


Sensor description

Our sensor can be divided into three principal parts as
shown in the block diagram of Fig. 2:

1. the measurement head with the sensing and guard-
ing electrodes placed on the pipe;

2. the electronic circuit;

3. the data acquisition system.


Abouelwafa & Kendall (1980) carried one of the first
investigations on the different shapes of the electrodes:
with only two electrodes it is not possible to obtain a
spatial resolution of the fluids in the measurement sec-
tion. Gas-liquid dispersion are in general eccentric and
localized in the upper part of the pipe, while liquid-
liquid dispersion, especially when the density difference
is small, can be distributed in all the cross section: the
same amount of dispersed phase placed in different po-
sitions of the cross section could give rise to different
outputs while it is necessary to have an unique solution
for each specific hold-up. The double helical solution
resulted the best one for all flow regimes Abouelwafa
& Kendall (1980); however this solution needs a larger
axial length of the probe, it is less sensitive than the con-
cave configuration and it is not easy to manufacture -
Caniere et al. (2008). For flow regimes with a conduc-
tive fluid near the pipe wall such as gas and oil disper-
sion in tap water it is necessary to have a high sensitiv-
ity, otherwise the conductive fluid acts as a shield for the
variations of the dispersed phase.
For the measurement head the concave configuration
is chosen with the electrodes that are directly placed fac-
ing each other on the external surface the pipe. Guard-
ing electrodes are placed before and after the sensing
electrodes as suggested for example by Canibre et al.
(2008).
Special care has been devoted to electrodes design, by
electromagnetic Finite-Element Method (FEM) simula-
tions using COMSOL MULTIPHYSICS. From previous
simulations on core-annular flow regime Demori et al.
(2009) we found that as the opening angle decreases,
measured capacitance becomes less sensitive to the spa-
tial resolution of the fluids, as also mentioned by Dos
Reis et al. (2008). Assuming these considerations valid
also for the dispersed flow regime, a covering angle 8e
90" is used. As the opening angle decreases the sensor







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


area decreases as well and the only way to have a good
sensitivity, is to increase the electrode length Lnc, but
the increase of the probe axial length makes the probe
measurements less local. A final compromise was to set
Lac to 0.1 m.
For void fraction measurements, Geraets & Borst
(1988), Abouelwafa & Kendall (1980) and Elkow &
Rezkallah (1997) used working frequencies between 1
and 2 MHz so for the transduction electronic circuit we
choose the working frequency f, of 2 MHz. Increasing
the working frequency the sensitivity increase because
the shielding effects of the conductive fluid decrease, but
at the same time it become more difficult to work with
standard electronic.
With reference to Fig. 2, polarizing electrodes on
the right side of the sensor head are connected together
to the terminal P. The two guard electrodes on the left
side are connected together to terminal GND, while
the measurement electrode is connected to terminal M.
The circuit excites terminal P with a sinusoidal voltage
V,(t) A,cos(w,t), while terminal GND is grounded.
The terminal M is connected to the virtual ground of
a transimpedance amplifier, which senses the current
collected by the central left-side measurement electrode
while holding it at the same null potential of the neigh-
boring guard electrodes. The transimpedance amplifier
is composed by an inverting-configuration operational
amplifier where the feedback impedance is made by the
parallel between the resistance RF and the capacitance
CF. This configuration allows to eliminate the effects of
stray capacitances of the connecting cables that can be
much higher than the effective capacitance C that has to
be measured Huang et al. (1988). The signal Vo ex-
iting the transimpedance amplifier is then demodulated
and filtered.
The DC output voltage Vmeas is then fed to the data
acquisition and visualization system which is based on a
16-bit A/D data acquisition board, National Instruments
PCI-6259, connected to a personal computer. The sys-
tem is set to work in negative sensitivity: in general the
gaseous or liquid dispersed phase have a lower dielec-
tric constant respect to water so their addition cause a
decrease of the measured capacitance C. The negative
sensitivity is convenient: the condition with only water
flowing is set to give an output of 0 V (reference value)
so the addition of the dispersed phase give rise to an in-
crease of Vmeas.
The overall sensitivity of the probe is 2 V/pF and
this mean that the output voltage variations for dispersed
flow regime are in the order of hundreds of mV.
In Demori et al. (2009) we found that in presence of
a conductive fluid inside the pipe, to have a correct mea-
surement it is necessary to consider not only the phe-
nomena in the cross section, but also the ones along the


L


Figure 3: Measurement section and external couplings
along the pipe.


pipe itself. With reference to Fig. 3, the measurement is
influenced by external couplings. The first possible cou-
pling mechanism is the contact between the conductive
water wire with the metallic parts of the hydraulic set-
up, such as pumps. The second possible mechanism is
the capacitive coupling between water and metallic parts
that can be present nearby, such as mounting fixtures and
supports. These external couplings represent an addi-
tional path from which currents It can leak out from
the measurement section in the axial direction along the
pipe. The larger are these current losses, the lower is
the current collected by the measurement electrode, de-
creasing the sensitivity and affecting the correct mea-
surement. For a given system configuration, these losses
could be virtually corrected for. However, when the dis-
persed phase is also present in the pipe together with
the conductive continuous phase, the section of the con-
ductive path varies with the hold-up, which is the quan-
tity we want to measure, and therefore the estimation
of the external couplings to correct the measurement is
not straightforward. A solution for minimizing current
losses is to increase the guarding electrodes length De-
mori et al. (2009). The length of the guarding elec-
trodes was experimentally tuned to a value LG of 0.4 m.
It is important to note that from our analysis the guard-
ing electrodes have not only the function to control the
electric field (as suggested in previous works), but also
to minimize the influence on the measurements due to
the current losses.

Calibration curve
We now present a calibration curve for liquid-liquid oil
in water dispersions. The calibration curve is obtained
considering the continuous and the dispersed phase as
an homogeneous equivalent fluid having opportunely
weighted permittivity and conductivity as shown in Fig.
4. Different modelizations have been proposed to calcu-
late the average properties of this equivalent fluid. Ger-







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


0995~


~ 09r
O
rr

098~


Figure 4: Approximation of the dispersed flow regime
electrical properties.


aets & Borst (1988) suggested to use the simple relation
proposed by Maxwell (1881) that consider spheres of
uniform size


0 0 02 0 04 0 06 0 08 H 0 12 0 14 0 16 0 18 D 2


Figure 5: Calibration curve obtained with FEM simula-
tions.


Results


In this section we present the preliminary results ob-
tained with the probe. First of all we show the ability
of the probe to detect small amount of oil and air dis-
persion in conductive water and then we present prelim-
inary hold-up results for oil in water dispersions.


2(1 Hd)
-r,eq or,c e


2(1 Hd)
2e + He


2(1 Ho)
- or'" 2 +Ho


2(1 Ho)
2 H


Dynamic response
The dynamical response of the probe is firstly tested
starting from only water flowing (reference condition)
and inserting a small amount of oil, Fig. 6(a), or air, Fig.
6(b) in the pipe. Results are reported in Fig. 7.


The equivalent fluid properties has the same properties
of the continuous phase weighted by the hold-up, ne-
glecting the characteristics of the dispersed phase. This
modelization is valid when properties of the dispersed
phase (in this case oil) are small compared to the contin-
uous ones. In our case we have a, = 0.03 S/m, Er,w,
78, and Er,o = 2.7.
Eq. 1 and 2 are valid for low hold-up values (Ho <
1I'. .) that is when the spheres can be considered as not
interacting, Geraets & Borst (1988), but we extended it
until Ho = -l's '. The calibration curve through FEM
simulations introducing Eq. 1 and 2.
Since oil, water, and glass permittivity can change due
to material inhomogeneities or temperature differences,
it is better to express the capacitance measurements in
a somewhat different way, by introducing the Relative
Capacitance Variations (ROD), defined as

C C ,
ROD = o ROD < 1, (3)
C, Co

where C is the capacitance to be measured and Co and
C, represent the capacitance when the pipe is filled by
oil and water respectively Geraets & Borst (1988).
In this way it is possible to make straight comparison
between experimental data and numerical simulations.
In Fig. 5 the obtained calibration curve is shown. As
can be seen a 20 %b variation of the value Ho corresponds
to a small variation in ROD thus confirming that for the
analysis of dispersed flow regimes it is necessary to have
high sensitivity probes.


Figure 6: Pictures of the oil in water (a) and air in water
(b) dispersions used to test the dynamic re-
sponse of the probe.

In spite of the small amount of dispersed phases in-
serted in the pipe and the water conductivity, the sensor
shows a rapid response to the flow regime variations. For
air in conductive water dispersions Canibre et al. (2008)
found no sensitivity of their capacitance probe, while in
our case there is a clear difference between the only wa-
ter reference signal and the dispersed one thus confirm-
ing the sensitivity of the probe. The air water dispersion
proposed in Fig. 6(b) is characterized by a slight inter-







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


0 6 -o/wv d spers


0 2 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (s)

Figure 7: Dynamic response of the probe respect to oil
and air injection in the test water continuous
phase.


mittent behavior where agglomerations of small bubbles
are separated by a finer dispersion. The sensor is also
able to detect also this intermittent behavior with oscil-
lations of the output signal Vmeas around the mean value.
From a qualitative point of view it is also possible to note
that oil in water dispersion is uniformly distributed in
the cross section, while air in water dispersion consists
of the mentioned agglomeration of small bubbles in the
upper part of the pipe. This consideration entails that for
gas-liquid dispersions the modelization of the equivalent
liquid to obtain a calibration curve require a different ap-
proach because air bubbles cannot be considered as non
interacting.

Hold-up data
We now show the preliminary hold-up measured values
for liquid-liquid oil in water dispersions obtained using
the calibration curve previously described. The results
are compared with the homogeneous model prediction
and QCV data.
The homogeneous model neglects the slippage be-
tween the in situ velocities and so the dispersed phase
is assumed to flow at the same velocity of the continu-
ous one; in general this can be considered valid for fine
oil water dispersions Brauner (2004). Under these as-
sumptions the oil hold-up can be computed by the water
and oil flow rates


Ho (4)
Go + Q=
In Fig. 8 we show the results of the comparison be-
tween sensor measurements and homogeneous model
predictions, while in Fig. 9 we show the comparison
between sensor and QCV measurements.
As can be seen in Fig. 8, the sensor overestimate the
oil hold-up Ho respect to the previsions obtained by Eq.
4. This can be mainly due to the model used for calculat-
ing eq, and seq and to the fact that from a fluid-dynamic


Figure 8: Comparison between oil hold-up Ho mea-
sured values and predicted ones. The solid
line represent perfect agreement.


0.1
Ho probe (-)


Figure 9: Comparison between oil hold-up Ho mea-
sured values and QCV. The solid line repre-
sent perfect agreement.







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


ferent oscillations around the averaged value following
the physical hold-up variations.
A calibration curve for liquid-liquid dispersion is ob-
tained by numerical simulations. Experimental results
are compared with reference values obtained by the ho-
mogeneous model and QCV, showing that the probe
overestimate the hold-up. In spite of the overestimated
results, the sensor is able to detect small amounts of the
dispersed phase with a monotonic response allowing the
possibility to obtain hold-up measurements once tuned
the calibration curve.


References

Abouelwafa M.S.A., Kendall E.J.M., The use of capaci-
tance sensors for phase percentage determination in mul-
tiphase pipelines. IEEE Transactions on Instrumentation
and Measurement 29, pp. 24-27, 1980

Brauner N., Liquid-Liquid Two-Phase Flow Systems.
Modeling and Control of Two-Phase Flow Phenomena,
Ed. V. Bertola, CISM Center, Udine, Italy, 2004

Caniere H., T'Joen C., Willockx A., De Paepe M., Ca-
pacitance signal analysis of horizontal two-phase flow in
a small diameter tube. Experimental Thermal and Fluid
Science 32, pp. 892-904, 2008

Demori M., Ferrari V., Strazza D., A sensor system for
oil fraction estimation in a two-phase oil-water flow.
Procedia Chemistry, Vol. 1, pp. 1247-1250, 2009

dos Reis E., Carajal Florez F.A., Mendez de Moura L.F.,
Numerical modeling of a capacitive probe for measuring
the water layer thickness in the annular heavy oil-water
flow. 11Ith Intemnational Conference on Multiphase Flow
In Industrial Plant, Palermo, Italy, 2008

Elkow K.J., Rezkallah K.S., Void Fraction Measure-
ments in gas-liquid flows under 1-g and p-g conditions
using capacitance sensors. International Journal of Mul-
tiphase Flow 23, pp. 815-829, 1997

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

Geraets J.J., Borst J.C., A capacitance sensor for two-
phase void fraction measurement and flow pattern iden-
tification. International Joumnal of Multiphase Flow 14,
pp. 305-320, 1988

Huang S.M., Stott A.L., Green R.G., Beck M.S., Elec-
tronic transducers for industrial measurement of low ca-
pacitance value. Journal of Physics E: Scientific Instru-
ments 21, 242-250, 1988


point of view the homogeneous model can be consid-
ered valid only for fine dispersions Brauner (2004) -
while in our experiments we find that for Ho over 5 %
we reach bubbly flow.
The aim of this work is to develop a relative simple
sensor for sensing two-phase dispersion when the con-
tinuous phase is a conductive fluid and from Fig. 8 and
Fig. 9 it is possible to affirm that the proposed sensor
system is capable to detect small amounts of the dis-
persed phase with appropriate sensitivity and a mono-
tonic response with the increase of the dispersed phase.
Once tuned the calibration curve with reference hold-up
values, it is possible to have time continuous measure-
ments of the hold-up.
QCV data are in substantial agreement with the ho-
mogeneous model prediction, Fig. 10, at least for fine
oil in water dispersion, however data repeatability is not
always satisfactory. The QCV values can be used as ref-
erence values for a qualitative calibration of the probe,
but for its validation other methods are needed.


Ho model (-)


Figure 10l: Comparison between oil hold-up Ho QCV
values and predicted ones. The solid line
represent perfect agreement.



Conclusions

In this work we present a capacitance sensor system for
sensing liquid-liquid and gas-liquid dispersions. The
probe is developed to work when the continuous phase
is a conductive fluid. Problems connected with con-
ductivity are solved working on the sensitivity in the
cross-section and also reducing the losses along the pipe
with a proper length of the guarding electrodes: the
probe shows optimal dynamic response and provide dif-







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


Kumara W.A.S., Halvorsen B.M., Melaaen M.C.,
Single-beam gamma densitometry measurements of
oil-water flow in horizontal and slightly inclined
pipes. International Joumnal of Multiphase Flow, doi:
10.1016/j .ijmultiphaseflow.2010.02.003, 2010

Maxwell J.C., A treatise on electricity and magnetism,
Vol. 1, 3rd edition. Clarendon Press, Oxford, 1881

Oddie G., Shi H., Durlofsky L.J., Aziz K., Pfeffer
B., Holmes J.A., Experimental study of two and three
phase flows in large diameter inclined pipes. Intema-
tional Joumnal of Multiphase Flow, Vol. 29, pp. 527-558,
2003

Simmons M.J.H., Azzopardi B.J., Drop size distribution
in dispersed liquid-liquid pipe flow. International Jour-
nal of Multiphase Flow, Vol. 27, pp. 843-859, 2001

Van der Meulen G.P., Zangana M., Zhao D., Azzopardi
B.J., Phase distribution measurements by conductance
probes and pressure drop in gas liquid flows. 7th World
Conference on Experimental Heat Transfer, Fluid Me-
chanics and Thermodynamics, Krakow, Poland, 2009




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