Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 17.2.2 - Pressure Drop and Heat Transfer of Single-Phase Flow and Two-Phase Boiling Flow in Thin-Rectangular Channels
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00416
 Material Information
Title: 17.2.2 - Pressure Drop and Heat Transfer of Single-Phase Flow and Two-Phase Boiling Flow in Thin-Rectangular Channels Micro and Nano-Scale Multiphase Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Koizumi, Y.
Ohtake, H.
Sato, K.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: thin rectangular channel
two-phase flow
flow patterns
pressure drop
boiling heat transfer
 Notes
Abstract: Heat transfer and pressure drop of boiling flow in the horizontal-thin-rectangular channels have been examined. The channel width was 10 mm and the channel height was varied from 1.104 mm through 0.184 mm. Test fluid was distilled water. The sub-cooling of fluid at the inlet of the test section was from 5 °C to 20 °C . Bubbly flow, slug flow, semi annular flow and annular flow were observed. The flow pattern transition agreed well with the Baker flow pattern map for the usual sized flow path. The Martinelli and Nelson method for the pressure drop of boiling two-phase flow predicted present experimental results well. In the low quality region, measured pressure drop was larger than the predicted value because of sub-cooled boiling. When the channel height was higher than 0.4 mm, a flow state was bubbly flow until the CHF condition and flow boiling heat transfer was well expressed with the Rohsenow pool boiling correlation. When the channel height was narrower than 0.4 mm, a film flow state came out at an early stage after boiling initiation. Tiny bubbles were noted in the film on the heat transfer surface. The heat transfer coefficient became larger than that of the Rohsenow pool boiling correlation because of the effective heat transfer of the evaporation of the film. The CHF was lower than the value of the usual sized flow channel. The Koizumi and Ueda method that was developed by considering the mechanism of the formation of a dry area on the heat transfer surface predicted well the trend of the critical heat flux of the present experiments. The best fit curve of the CHF was correlated for the present experimental conditions.
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: VID00416
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1722-Koizumi-ICMF2010.pdf

Full Text

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


Pressure Drop and Heat Transfer of Single-Phase Flow and Two-Phase Boiling Flow in
Thin-Rectangular Channels

Y. Koizumi*, H Ohtaket and K. Satot


Prof, Department of Functional Machinery and Mechanics, Shinshu University
3-15-1, Tokida, Ueda, Nagano 386-8567, JAPAN
koizumiy @shinshu-u.ac.jp
tProf., Department of Mechanical Engineering, Kogakuin University
2665-1, Nakanomachi, Hachoji, Tokyo 192-0015, JAPAN
atl0988(@ns.kogakuin.ac.jp
tDepartment of Mechanical Engineering, Kogakuin University
1-24-2, Nishishinjuku, Shinjuku, Tokyo 163-8677, JAPAN
am08027ins kogakuin ac jp



Keywords: thin rectangular channel, two-phase flow, flow patterns, pressure drop, boiling heat transfer




Abstract

Heat transfer and pressure drop of boiling flow in the horizontal-thin-rectangular channels have been examined. The
channel width was 10 mm and the channel height was varied from 1.104 mm through 0.184 mm. Test fluid was distilled
water. The sub-cooling of fluid at the inlet of the test section was from 5 OC to 20 OC Bubbly flow, slug flow, semi annular
flow and annular flow were observed. The flow pattern transition agreed well with the Baker flow pattern map for the usual
sized flow path. The Martinelli and Nelson method for the pressure drop of boiling two-phase flow predicted present
experimental results well. In the low quality region, measured pressure drop was larger than the predicted value because of
sub-cooled boiling. When the channel height was higher than 0.4 mm, a flow state was bubbly flow until the CHF condition
and flow boiling heat transfer was well expressed with the Rohsenow pool boiling correlation. When the channel height was
narrower than 0.4 mm, a film flow state came out at an early stage after boiling initiation. Tiny bubbles were noted in the film
on the heat transfer surface. The heat transfer coefficient became larger than that of the Rohsenow pool boiling correlation
because of the effective heat transfer of the evaporation of the film. The CHF was lower than the value of the usual sized flow
channel. The Koizumi and Ueda method that was developed by considering the mechanism of the formation of a dry area on
the heat transfer surface predicted well the trend of the critical heat flux of the present experiments. The best fit curve of the
CHF was correlated for the present experimental conditions.


INTRODUCTION
Flow and heat transfer are quite fundamental phenomena
in usual sized machines. Thus, huge researches have been
performed for these. As a result of it, the results have been
built up into the fluid engineering and the heat transfer
engineering. Thus, design parameters such as the pressure
drop and the heat transfer coefficient for the usual size are
easily obtained.
If the flow area is scaled down, that effect is usually
classified by using the Knudsen number Kn of the ratio of
the mean free path of molecules to the characteristic length
of the flow path (Kandlikar and Grande, 2002). The
boundary between the continuum flow and the slip flow is
roughly 100 tpm for gas and 1 tpm for liquid.
When the flow size is scaled down from the usual size
toward the slip flow region, the Reynolds number
decreases and the flow becomes laminar in most cases. The
pressure drop is proportional to the square of the flow size


(hydraulic diameter) and the heat transfer coefficient is
inversely proportional to flow path size. It means that
when the flow size path is scaled down, heat transfer is
augmented in exchange for a large increase in the pressure
drop. The large pressure drop and the large heat transfer
rate in conjunction with an increase in the ratio of surface
area to volume cause the problems of the effect of
compressibility and the variation of physical properties
along the flow path due to temperature change. The effect
of surface roughness also becomes relatively important.
When the flow path size is decreased to the boundary to
slip flow, it has been proved that (Wu and Little, 1983 and
1984, Phahler et al., 1991, Choi et al., 1991 and Wang and
Peng, 1994) the friction factor has the same trend as that
for the usual size although it is a little bit lower than the
value of the usual sized flow path and that the Nusselt
number in the laminar flow region shows dependency on
the Reynolds number on the contrary to the usual sized









flow path.
These are for single-phase flow. In the case of two-phase
flow, the situation becomes more complicated. As stated
above, the boundary between the continuum flow and the
slip flow in the liquid single-phase flow is one hundredth
than that for the gas single phase flow. In two-phase flow,
the interaction between gas and liquid is important. The
interaction is closely related to interface shape. The
interface shape becomes dependent on the flow path size as
it becomes small. A surface tension effect may also
increase. Thus, as the flow path size of the two-phase flow
is decreased down toward the boundary between the
continuum flow and the slip flow, the effect of the
down-sizing of the flow path size may become prominent
well short of the boundary.
Serizawa et al. (2002) have reported from their
experiments for 20 ~ 100 pm tubes that although they
observed a new flow pattern of ring flow, flow patterns
were similar to those for the usual sized flow path and
were roughly correlated by the Mandhane flow pattern
map (Mandhane et al., 1974) for usual-sized pipes. Triplett
et al. (1999) conducted experiments for 1.09 ~ 1.49 mm
tubes. They reported the difference of the flow patterns
between the mini flow channels and the usual sized flow
path. Chung et al. (2003) also observed the difference of
the flow patterns, however the flow patterns that they
observed were a little bit different from what Triplett et al.
have report
As for the pressure drop, Chung et al. stated that the
separated model of Lockhart-Martinelli (1949) correlated
well with their data. Mishima and HIbiki (1996) also
pointed out that their data for tubes of 1 ~ 4 mm diameter
were well correlated by the Chisholm and Laird correlation
(1958) with some modification. Pehlvian et al. (2006) have
reported from their experiments for tubes with diameter of
0.8, 1.0 and 3.0 mm that the major sources of the
discrepancies between experimental results and existing
flow pattern maps were the difficulty of the definition of
the flow patterns and the complex nature of the gas-liquid
interface. They also pointed out about the pressure drop
that the homogeneous model lost the accuracy as the flow
channel size reaches the micro size and the Chisholm
model under predicted the pressure drop. As stated above,
it seems that even though the flow pattern and the pressure
drop are the most fundamental items of the two-phase flow,
clear results for these have not been obtained so far.
Kandlikar (2"_114 has classified the flow channel by
using the hydraulic diameter Dh such as conventional
channels (Dh > 3 mm), minichannels (3 m > Dh > 200 tpm),
microchannels (200 ptm > Dh > 10 pm, transition channels
(10 pm > Dh > 0.1 tpm), transition microchannels (10 pm >
Dh > 1 pm), transition nanochannels (1 pm > Dh > 0.1 pm)
and molecular nanochannels I(0.1 pm > Dh). He has
pointed out that nucleate boiling seems to be dominant in
flow boiling in the microchannels. He concluded that
further experimental research was necessary to develop
more accurate models and predictive techniques for a flow
boiling heat transfer coefficient and a critical heat flux
(CHF) in the microchannles. Liu and Garimella (2005)
reported that the modified Chen model (1963) provided
good prediction for their flow boiling experimental results
for parallel microchannels of 0.275 mm x 0.636 mm
rectangular cross-section


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

It is clear that reliable results for predicting the pressure
drop, the heat transfer coefficient and the CHF of boiling
flow in mini- and micro-channels are still needed before
going down to nano-channels. Koizumi et al (2008)
examined the pressure drop, the heat transfer coefficient
and the CHF for the rectangular flow channels of 10 mm
wide and 0.2 mm 0.62 mm high. They reported that the
friction factor and the heat transfer coefficient for the
single-phase flow began to deviate from the values for the
usual sized flow path around the gap height of 0.35 mm.
The Baker flow pattern map seemed a good measure to
identify flow patterns. Boiling was dominant during the
forced flow boiling. The critical heat flux was lower than
the value of the usual sized flow channel. In the present
paper, further experimental results will be presented
mainly focused on the results of the boiling two-phase flow.
The experiments were performed for the gap height of
0.184 1.104 mm by using distilled water.


Fig. 1 Experimental Apparatus


EXPERIMENTAL
PROCEDURES


APPARATUS


AND


Experimental Apparatus

The experimental apparatus used in the present study is
schematically shown in Fig. 1. The apparatus is composed
of a water storage tank, water circulation pumps, a test
section and a condenser. Distilled water is used in
experiments. When a flow rate is high or flow area is small,
thus high pressure drop in the test section is expected, two
or three pumps connected in series are used to get high
delivery pressure. The water storage tank is a closed-type
to protect it from dust although it is open to atmospheric
pressure. The tank has electric heaters to control water
temperature. Water pumped out from the tank by the
circulation pump flows into the test section through a
rotameter. Test fluid flowing out from the test section
outlet returns to the tank. Water temperature is measured in
the water tank and at the inlet and the outlet of the test
section with chromel-alumel thermocouples.









Heating Surface


240mm
E 9 95mm mm




Gap


Cartridge heaters


Fig. 2 Details of Test Section

Details of the test sections are presented in Fig. 2. The
top cover of the test section is made of a transparent
poly-carbonate plate which is one of acrylic resin for high
temperature use. The width, the thickness and the length of
the test section are 65, 20 and 300 mm, respectively. A
groove of 10 mm wide and 250 mm long is formed on the
bottom surface of the top cover by mechanical machining.
The depth of the groove 6; the test flow channel height
which will be mentioned later more precisely, is 0.184 ~
1.014 mm in the present experiments. The top surface of
the groove is polished with abrasive compound. The
bottom plate of the test section is made of a glass-epoxy
plate. The width, the thickness and the length are 65, 24
and 300 mm, respectively. At the center portion of the plate,
a rectangular window of the width of 10 mm and the length
of 50 mm is formed. A heating test section is made of
copper and the length is 50 mm. The cross-section of it is
reversed T-shaped. The width and the height of the leg of T
are 10 mm and 24 mm, respectively. The thickness and the
length of the top bar of T are 22 mm and 65 mm,
respectively. The heating test section is fit upward into the
rectangular window of 50 mm long and 10 mm wide
formed in the epoxy plate from the bottom surface of the
epoxy plate so that the top surface of the heating test
section is flush with the top surface of the epoxy plate.
Then, the top cover of the test section and the bottom plate
are tightly fixed. The rectangular cross-section groove
formed on the bottom surface of the top cover becomes the
test flow channel. Therefore, the groove of the width of 10
mm, the depth of depth 6 and the length of 250 mm
becomes the test flow channel. By changing the depth of
the groove, the test flow channel height is varied. The inlet
side of the test heating section is located at 95 mm from
the inlet of the test flow channel. This test section is
horizontally placed.
Three electric heaters of 500 W each are fixed at the
bottom of the test heating section for heating. Electric
power supplied for the heaters is controlled by a voltage
regulator. Nine chromel-alumel thermocouples are fixed as
shown in Fig. 2. A surface heat flux is derived by using the


Pressure tap
Test flow channel
Thermo-couples


u


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

temperature gradient which is calculated from
temperatures at the three elevations arranged vertically in a
straight line and the thermal conductivity of cupper. A
surface temperature is obtained by extrapolating the
temperatures at the three elevations using the temperature
gradient.
A pressure in the test flow channel is measured at the
inlet side of the heating test section with pressure
transducers as shown in Fig. 2. A pressure drop in the test
flow channel is measured between the inlet and the outlet
of the heating test section with differential pressure
transducers.
The apparatus was well thermally insulated.

Experimental Procedure

Rotameters, thermocouples, pressure transducers and
differential pressure transducers were calibrated before
experiments. The surface of the heating test section made
of copper was polished with abrasive compound before
each experiment. The height of the rectangular flow
channel was measured with a laser displacement meter
before each experiment at 15 locations of the heating test
section. The average of the nine values was determined to
be the test flow channel height 6 in that experiment. After
the experiment, the test flow channel height 6 was
confirmed again with the same measuring method.
A water tank was filled with distilled water. Then, using
a bypass line between a pump and the tank, water was
circulated. There was a filter of 5 pm mesh in the bypass
line. The tank was cleaned up by this bypass circulation.
Then, water was circulated in the test flow channel. Guide
tubes for pressure and differential pressure transducers
were filled with water.
Water temperature subcooling at the test section inlet
was set at 20 5 C by adjusting electric power supply for
heaters in the water tank. A flow rate to the test section was
fixed at a pre-scribed value. Then, electric power supply
for heaters of the heating test section was increased
stepwise. After it was confirmed that the flow condition
was fully stabilized at each step, the flow rate, the pressure
at the heating test section inlet, the pressure drop between
the inlet and the outlet of the heating test section, the water
temperatures at the inlet and the outlet of the test section,
the temperatures of the heating test section at nine
locations were recorded. A flow state was recorded from
the top through the transparent top cover of the test section
by a high speed video camera at 500 frames/s. This
procedure was iterated until the surface temperature of the
heating test section showed an unstable state such as it
fluctuated or kept going up. The flow channel height 6
tested in the present experiments were in a range from
0.184 mm to 1.014 mm; the hydraulic diameter Dhy from
0.361 mm to 1.84 mm.
It was sometimes observed in preparatory experiments
that bubbles generated in the test flow channel moved
oscillatory back and forth. Large flow resistance was
added between the pump outlet and the test flow channel to
avoid it. After the large resistance was introduced, the
bubbles never went back and a float in a rotameter did not
show fluctuation. Thus, a constant flow rate at the inlet of
the test section was maintained in the boiling flow heat
transfer experiments.



























a Semi-Annular Flow Annular Flow


Fig. 3 Flow Patterns



Amular-D isper sed Flow


105 Anndiar Fl c



S104
-b
o V


Stratifed Fbw


Bubbk Flow



Slig Fbw


Plug Flow


10-1 10 101 102 103 104


Gap= 0 567mm, u= 0 49m/s, Re= 1093 1644
Gap= 0 478mm, u 1 04m/s, Re 2531 32)0
Gap 0 367mm, u 0 78m/s, Re 1370 1762
Gap= 0 205mm, u 1 53m/s, Re 2511 2611



Fig. 4 Flow Pattern Map


EXPERIMENTAL RESULTS AND DISCUSSIONS

Flow Pattern Correlation

Typical flow patterns observed in the present
experiments are presented in Fig. 3. These are bubbly flow,
slug flow, semi annular flow and annular flow. Since the
experiments are heating experiments, the flow pattern
varied along the flow channel. The flow patterns observed
at the exit side thermocouple location are correlated on the
Baker flow pattern map (Ueda, 1981). The result is
illustrated in Fig. 4. The present results are correlated well
with the Baker flow pattern map.

Boiling Two-Phase Flow Pressure Drop

Martinelli and Nelson (1948) presented the method to
calculate the pressure drop during the forced-circulation
boiling of water. Present results are compared with that
prediction. If the inlet condition of the flow channel is


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

saturated, the ratio of the two-phase flow pressure drop per
unit length; APF, in the boiling channel to the single-phase
flow pressure drop of saturated-water per unit length in the
case that the flow rate is equal to the total mass flow rate:
APLO, is expressed as


AP F- I dl / dPl
APLo L C -dzlF Cdz Lo

1 2-n
= xe (1-x) ttdx (1)
Xe O

where L is the channel length, P is the pressure, x is the
quality, Xe is the exit quality, z is the flow directional
coordinate and Ltt is the two-phase flow pressure drop
multiplier. The exponent n is 0.2 or 0.25 for turbulent flow.


0.04 0.06
Quality


Fig. 5 Pressure Drop of Boiling Flow


The inlet condition in the present experiments was
sub-cooled. Thus, the saturation point was estimated from
the heat flux and the flow rate and Eq. (1) was applied to
the saturated region. The pressure drop between the inlet of
the flow channel and the saturation point was derived from
the length ratio of the inlet-to-saturation point to the
heating test channel length and the measured single phase
pressure drop just before the condition that the outlet of the
heating test section became saturated. This pressure drop
in the single-phase region was subtracted from the
measured pressure drop between the inlet and the out let of
the heating test section. This value was set as the
two-phase pressure drop in the experiment. Then, it was


Gap = 0.567 mm, u = 1.010 m/s, Re = 2593 ~ 3456
- Gap = 0.567 mm, u = 0.492 m/s, Re = 1093 ~ 1629
- Gap 0.510 mm, u 0.763 m/s, Re 1897 2317
--- Gap 0.478 mm, u 1.109 m/s, Re = 2531 3200
Gap = 0.465 mm, u = 1.660 m/s, Re = 3899 4656
--Gap = 0.459 mm, u = 0.874 m/s, Re = 1849 ~ 2356
Gap = 0.459 mm, u = 1.682 m/s, Re = 3999 ~ 4761
- Gap = 0.395 mm, u = 2.002 m/s, Re = 4044 ~ 4612
-- Gap 0.368 mm, u 0.752 m/s, Re 1148- 1651
Gap = 0.367 mm, u = 0.780 m/s, Re = 1370 ~ 1761
- Gap 0.356 mm, u 0.799m/s, Re 1485 ~ 1726
-- Gap = 0.305 mm, u 1.245 m/s, Re 1920 ~ 2270
--- Gap = 0.285 mm, u = 0.570 m/s, Re = 829 ~ 989
- Gap = 0.287 mm, u = 1.760 m/s, Re = 2546 ~ 3067
-- Gap = 0.260 mm, u = 1.530 m/s, Re = 2511 ~ 2661






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


transformed to the measured two-phase flow pressure drop
of boiling flow per unit length; APF. The measured single
phase pressure drop just before the condition that the outlet
of the heating test section became saturated was
transformed to the single-phase flow pressure drop per unit
length: APLO.
The predicted and the measured two-phase flow pressure
drop of boiling flow are compared in Fig. 5. Except for the
low quality region, the Martinelli and Nelson method of Eq.
(1) provides good prediction for the experimental results.
The measured pressure drop is higher in the low quality
region. The inlet condition in the present experiments was
sub-cooled. Boling bubbles were observed in the
sub-cooled region. It is considered that the sub-cooled
boiling increased the pressure drop of the boiling flow.
This effect is more prominent when the heat flux is low
and the quality is low.


single phase flow heat transfer coefficient
ONB Bergles-Rohsenow
- FDB-Rohsenow(Cs,0.01)


10' 102
Wall superheat ATt K


Fig. 6 Boling Curve (6 = 0.567 mm)


----- single phase flow heat transfer coefficient
ONB Bergles-Rohsenow
- FDB-Rohsenow(C,0.011)


101 10"
Wall superheat AT,, K

Fig. 7 Boling Curve (6 = 0.479 mm)


single phase flow heat transfer coefficient
ONB Bergles-Rohsenow
-- FDB-Rohsenow(Cfs0.013)


Wall superheat ATt K


Fig. 8 Boling Curve (6 = 0.367 mm)

Boiling Heat transfer

Results of heat transfer experiments are presented in
Figs. 6 8 in the form of a boiling curve. In the figures,
the relation of the water single-phase heat transfer by the
Dittus and Boelter correlation, the onset condition of
nucleate boiling by the Bergles and Rohsenow correlation
(1964), the nucleate boiling relation by the Rohsenow
correlation (1962), the critical heat flux of saturated pool
boiling by the Kutateladze correlation (Ueda, 1981) are
illustrated with a dotted line, a solid line, a one-dot chain
line and a dashed line, respectively for comparison. The
arrow of ONB in the figures designates the onset condition
of nucleate boiling that is visually determined from
pictures recorded by a high speed video camera. The arrow
of CHF expresses the critical heat flux (CHF) condition in
the present experiments. Following the CHF point, further
increase in the heat flux resulted in surface temperature
fluctuation or continuous surface temperature increase.
Difference between the measured and the predicted
onset condition of boiling is not so large in Figs. 6 8. The
Dittus-Boeltr correlation predicts well the experimental
results before the boiling initiation. After boiling initiation,
data points start to increase and come close to the relation
of the Rohsenow pool boiling correlation in Figs. 6 and 7.
The inclinations of data plots during the fully developed
boiling period are close to that of the Rohsenow
correlation in these figures. On the contrary, the heat fluxes
after boiling initiation in Fig. 8 are higher than those in
Figs. 6 and 7 and also the values of the Rohsenow pool
boiling correlation. The CHFs are close to the values of the
Kutateladze correlation. The weak trend that the CHF
increases with a decrease in the gap height is observed in
Figs. 6 8.
Figure 9 11 are the flow states that were recorded with
a high speed camera. In Figs. 9 and 10, many large boiling
bubbles are noted. On the other hand, the flow state is film
flow and tiny bubbles are observed in the liquid film in Fig.
11. The flow states in the former cases and the latter case








were bubbly flow and annular flow until the CHF
condition, respectively. In the narrow gap case of the gap
height = 0.367 mm, the evaporation of the liquid film
became dominant in the boiling region and the heat
transfer was augmented, which resulted in the higher heat
flux than that of the Rohsenow correlation as shown in Fig.
8.


o 0.8

S0.6
o
> 0.4
Cz
Qo
S 0.2


Fig. 9 Boiling State (6 = 0.567mm)


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


-- Gap = 0.305 mm, u = 1.24 m/s (measured value)
A Gap = 0.567 mm, u = 1.01 m/s (measured value)
-*- Gap = 0.305 mm, u = 1.24 m/s (calculated value)
Gap = 0.567 mm, u = 1.01 m/s (calculated value)


0.4 0.6 0.8 1 1.2
Heat flux q, W/m2 [x106]


Fig. 10 Boiling State (6 = 0.479mm)


Fig. 11 Boiling State (6 = 0.367mm)


The area occupied by bubbles on the recorded picture
was measured and it was converted to a void fraction. The
void fractions are presented in Fig. 12. The void fractions
that were calculated by assuming homogeneous flow are
also included in the figure for comparison. The void
fraction in the narrow gap height case is much larger than
that in the wide gap height case. The void fraction in the
narrow gap height case exceeds 0.8 at low heat flux. It
implies that a flow state turns to the annular flow at an
early stage after boiling initiation. Thus, when the gap
height becomes narrow, the boiling heat transfer tends to
be augmented. It is supposed from the present
experimental results that the tendency becomes prominent
when the gap height become narrower than 0.4 mm.


Fig. 12 Void Fraction in Flow Channel


Critical Heat Flux

In the present experimental conditions, film flow on the
heat transfer surface seems important. Figure 13 is the
example of a flow state just before the CHF condition. The
heat transfer surface is partly covered with water film.
Large dried areas are observed on the heat transfer surface.
The dried areas spread out and finally the CHF condition
comes out.


Fig 13 Flow State just before CHF Condition

Koizumi, et al. (1998) have developed the way to
calculate the CHF for the film flow such as the annular
flow. Small bubbles were formed in the film on the heat
transfer surface. The bubbles burst and the heat transfer
surface was covered again by the film. At the point close to
the CHF condition, a dried area was left on the surface
after the bubble bursts. If the dried area is rewetted again
by the film flow, the state returns to the film flow state. If
the dried area is not rewetted by the film flow, the dried
area begins to grow and the surface temperature initiates to
rise. Then, several dried areas merges to form the
substantial large dried area, which eventually leads to the
CHF condition. It depends on the inertia force of the film
flow and the surface tension at the boundary between the
film and the dried area whether the dried area is rewetted
or not. Following this idea, they proposed the CHF


f









correlation focusing on the bubble burst in the film and
rewetting the dry area after the bubble burst as follows:


For upward film flow
0 0 2008
qCHF/Hfg = 0.028 x L0.0

P1Umcr P1


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

The best fit curve of the CHF for the present
experimental results is


qcHF/Hfg

PlUmcr


We-0.33


For downward film flow
K o0.08
CHF/ 0.0085 x xWe-33 (3)
PlUmcr P1i

In equations, We is the Weber number that is defined as

2
We = P mcrL (4)


Here, Hfg is the latent heat, Ls is the boiling length, qCHF
is the critical heat flux and umcr is the mean film velocity at
the critical point. Symbols p and a are the density and the
surface tension, respectively. Subscripts g and 1 denote gas
and liquid, respectively.
The film of the upward film flow is more disturbed than
the film of the downward flow. The larger disturbance
intermittently brings the larger inertia force in the dried
area than in the case of the downward film flow. The larger
inertia force tends to rewet the dried area to suppresses the
appearance of the dried area that lead to the CHF condition.
Thus, the CHF in the upward film flow is larger than that
in the downward flow.
Experimental results are compared with the predictions
by Eqs. (3) and (4) in Fig. 14. The trend of the CHF on the
parameter is well expressed, however, the CHFs in the
present experiments are lower than those of the correlation
for the downward film flow. When the flow channel size is
decreased down to the micro channel, the wall effect due to
the viscosity that restricts the growth of the film flow
disturbance may become prominent. Thus, as a result of
the less disturbed film condition, the CHF becomes smaller
than that for the downward film flow of the usual sized
flow path.


o x 20.08
0.0053 x L- xWe-0.33
Pg-13


CONCLUSIONS


Heat transfer and pressure drop of boiling flow in the
horizontal-thin-rectangular channels have been examined.
The channel width was 10 mm and the channel height was
varied from 1.104 mm through 0.184 mm. Test fluid was
distilled water. The sub-cooling of fluid at the inlet of the
test section was from 5 C to 20 C Following
conclusions were obtained.
1. Bubbly flow, slug flow, semi annular flow and
annular flow were observed. The flow pattern transition
agreed well with the Baker flow pattern map for the
usual sized flow path.
2. The Martinelli and Nelson method for the pressure
drop of boiling two-phase flow predicted present
experimental results well. In the low quality region,
measured pressure drop was larger than the predicted
value because of sub-cooled boiling.
3. When the channel height was higher than 0.4 mm, a
flow state was bubbly flow until the CHF condition and
flow boiling heat transfer was well expressed with the
Rohsenow pool boiling correlation. When the channel
height was narrower than 0.4 mm, a film flow state came
out at an early stage after boiling initiation. Tiny bubbles
were noted in the film on the heat transfer surface. The
heat transfer coefficient became larger than that of the
Rohsenow pool boiling correlation because of the
effective heat transfer of the evaporation of the film.
4. The CHF was lower than the value of the usual
sized flow channel. The Koizumi and Ueda method that
was developed by considering the mechanism of the
formation of a dry area on the heat transfer surface
predicted well the trend of the critical heat flux of the
present experiments. The best fit curve of the CHF was
correlated for the present experimental conditions.

REFERENCES


Kolzumi Ueda Correlation
(qcr(pg/P "i, -=C'We-033
C = 0 028 Down Forward Flow
-- C = 0 0085 Up Forward Flow
C = 0 0053 Horizontal Forward Flow
In2 -


* Gap =0 260 n u= 1 53 ims (2010)
* Gap= 0285n u= 0 57 i's (2010)
V Gap 0 287 n u= 1 76 ims (2009)
* Gap 0 305 nm u= 1 22 rms (2009)
' Gap = 0 318 n u= 1 00 rs (2009)
A Gap= 0356 n u= 0 79 s (2009)
* Gap = 0 367 n u 0 78 ils (2009)
* Gap = 0 368 u= 0 75 ils (2009)
A Gap= 0441 nmu= 076 ms (2007)
O Gap =0 459 m, u 0 90 ms (2008)
V Gap =0 459 nm u= 1 68 is (2009)
* Gap 0 462 n u 0 72 ims (2007)
* Gap 0 465 nm u 1 79 mis (2008)
A Gap 0 478 n u= 1 04 ims (2008)
A Gap 0 479 nm u= 1 61ms (2009)
V Gap = 0 510 n u 0 76 imls (2009)
o Gap 0 567 m u =049 s (2009)
m Gap0 567nr ul 01 ls (2009)
Gap =0 620 nrm u 0 27 is (2007)


We

Fig. 14 Correlation of CHF


Bergles, A. E. and Rohsenow, W. M., 1964. The
Determination of Forced-Convection Surface-Boiling Heat
Transfer, Trans. ASME, Ser. C, 86, 365 372.

Chen, J. C., 1963. A Correlation for Boiling Heat
Transfer to Saturated Fluids in Convective Flow, ASME
Paper 63-HT-34.

Chisholm, D. and Laird, A. D. K., 1958. Two-Phase
Flow in Rough Tubes, Trans. ASME Vol. 80, No. 2,
276-286.

Chung, P. M. Y., Kawaji, M., Kawahara, A. and Shibata,
Y., 2003. Two-Phase Flow through Square and Circular
Microchannels -Effect of Channel Geometry, Proc. the 6th
ASME-JSME Thermal Engineering Joint Conference,


10' 101






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

CD-ROM FED-AJ03.

Liu, D. and Garimella, S. V, 2005. Flow Boiling in
Microchannel Heat Sink, 2005 ASME IMECE, CD-ROM
IMECE2005-79555.

Kandlihkar, S. G. and Grande, W. K., 2002. Evaluation
of Microchannel Flow Passages Thermohydraulic
Performance and Fabrication Technology, ASME IMECE,
CD-ROM, IMECE2002-32043.

Kandlihkar, S. G, 2004. Heat Transfer Mechanisim
during Flow Boiling in Microchannels, Trans. ASME, J. of
Heat Transfer Vol. 126, No. 1, 8 16.

Koizumi, Y, Matsuo, T., Miyota, Y. and Ueda, T,
1998. Dry-out Heat Fluxes of Falling Film and Low-Mass
Flux Upward Flow in Heated Tubes, Trans. JSME, Ser. B,
Vol. 64, No. 624, 212 219.

Koizumi, Y, Yamada, T and Ohtake, H., 2008. Flow
Boiling Heat Transferand Two-Phase Flow Pressure Drop
in Thin-Rectangular Channels, ASME ICNMM2008,
CD-ROM ICNMM2008-62026.

Lockhart, R. W. and Maritinelli, R. C., 1949. Proposed
Correlation of Data for Isothermal Two-Phase
Two-Component Flow in Pipes, Chem Eng. Progress, 45,
39-48.

Mandhane, J. M., Gregory, G A. and Azik, K., 1974. A
flow pattern Map for Gas-Liquid Flow in Horizontal Pipes,
Int. J. of Multiphase Flow, Vol. 1, No. 4, 537-553.

Martinelli, R. C. and Nelson, D. B., 1948. Prediction of
Pressure Drop during Forced-Circulation Boiling of Water,
Trans. ASME, Vol. 70, 695-702.

Mishima, K. and Hibiki, T., 1996. Some Characteristics
of Air-Water Two-Phase Flow in Small Diameter Vertical
Tubes, Int. J. Multiphase Flow, Vol. 22, No. 4, 703-712.

Pehlivan, K., Hassan, I. and Vaillancourt, M., 2006.
Experimental Study on Two-Phase Flow and Pressure Drop
in Millimeter-size Channels, Applied Thermal Engineering,
Vol. 26, 1506-1514.

Rohsenow, W. M., 1952. A Method of Correlating Heat
Transfer Data for Surface Boiling Liquid, Trans. ASME,
74, 969-976.

Serizawa, A., FEN, Z. and Kawahara, Z., 2002.
Two-Phase Flow in Micro Channel, Proc. JSMF Annual
Meeting 2002, 29 31.

Triplett, K. A., Ghiaasiaan, S. M., Abdel-Khalik, S. I.
and Sadowski, D. L., 1999. Gas-liquid Two-Phase Flow in
Microchannels, Part I: Two-Phase Flow Patterns, Int. J. of
Multiphase Flow, Vol. 25, 377-394.

Ueda, T, 1981. Two-Phase Flow -Flow and Heat
Transfer-, Yokendo Co., 11-12, 256-265.




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