Heat Transfer Engineering, 26(3):20–27, 2005
C Taylor & Francis Inc.
Copyright ISSN: 0145-7632 print / 1521-0537 online
DOI: 10.1080/01457630590907167
Experimental Study of Flow Patterns,
Pressure Drop, and Flow Instabilities
in Parallel Rectangular Minichannels
PRABHU BALASUBRAMANIAN and SATISH G. KANDLIKAR
Thermal Analysis and Microfluidics Laboratory, Mechanical Engineering Department, Rochester Institute
of Technology, Rochester, New York
Flow boiling heat transfer in parallel minichannels and microchannels is one of the solutions proposed for cooling high
heat flux systems. The associated increase in the pressure drop poses a problem that needs to be studied in detail before the
small diameter channels can be implemented in practical systems. The pressure drop fluctuations and the flow instability in
a network of parallel channels connected by a common header also need to be addressed for the stable operation of flow
boiling systems. The current work focuses on studying the flow patterns, pressure drop fluctuations, and flow instabilities in
a set of six parallel rectangular minichannels, each with 333 µm in hydraulic diameter. Deionized and degassed water was
used for all the experiments. The pressure fluctuations are recorded and signal analysis is performed to find the dominant
frequencies and their amplitudes. These pressure fluctuations are then mapped to their corresponding flow patterns observed
using a high speed camera. The results help us to relate pressure fluctuations to different flow characteristics and their effect
on flow instability.
Flow boiling heat transfer consists of nucleate and convective
boiling components. It has been shown by a number of investigators [1–3] that the nucleate boiling mode becomes dominant
during flow boiling in micro- and minichannels. These findings are supported by visual observations of nucleating bubbles
[3]. These bubbles have an explosive growth rate following nucleation, rapidly developing into slugs that fill the entire channel. Brutin et al. [4] devised transparent minichannels with a
hydraulic diameter of 889 µm to observe the two-phase flow
patterns and oscillations. Under steady-state boiling conditions,
they observed that smaller bubbles flowing near the channel sides
flow faster than the large bubbles in the middle of the channel.
Back flow or reverse flow extending all the way into the inlet manifold was also reported in their work. Under unsteady
boiling conditions, they observed higher amplitudes of pressure
oscillation at the inlet as compared to those at the outlet.
Peles [5] studied two-phase flow boiling instabilities in multiple channels with hydraulic diameters ranging from 50–200 µm.
He reported rapid bubble growth, similar to the back flow or reverse flow, to be the most significant boiling regime observed
under his experimental conditions. Kasza et al. [6] performed a
detailed study of flow visualization of nucleate boiling in small
channels. They observed that the dynamics of bubble nucleation, growth, and coalescence to be so intense that the flow
locally undergoes periodic reversal and intense mixing over the
entire channel cross-section. A recent paper by Campbell and
Kandlikar [7] explains the time-dependent pressure drop fluctuations during two-phase flow in a minichannel. They observed
the peak-to-peak variation of pressure drop to be constant for
low surface temperatures, indicating the presence of nucleate
boiling.
The objective of the present work is to study the pressure
fluctuations in minichannels and relate them to the observed
flow patterns.
EXPERIMENTAL SETUP
Figure 1 shows a schematic of the visualization test section.
The test section is a compact heat exchanger consisting of a set
of six parallel, rectangular minichannels machined on top of a
copper block. The channels are made using a conventional circular milling cutter. The finished channels measure 990 µm in
width by 207 µm in depth, with a total length of 63.5 mm. The
Address correspondence to Satish G. Kandlikar, Rochester Institute of
Technology, Department of Mechanical Engineering, James E. Gleason
Building, 76 Lomb Memorial Drive, Rochester, NY 14623-5604. E-mail:
sgkeme@rit.edu
20
EXPERIMENTAL STUDY OF FLOW PATTERNS, PRESSURE DROP, AND FLOW INSTABILITIES
21
EXPERIMENTAL PROCEDURE
Figure 1 Schematic of test section.
copper block is sandwiched between an optically clear Lexan
top cover and a phenolic bottom cover. The copper block is an
electrolytic tough pitched alloy C1100, with a thermal conductivity of 399 W/m-K at 20◦ C. The Lexan cover is a polycarbonate
material with a low thermal conductivity of 0.19 W/m-K. The
Lexan cover helps in flow visualization and also serves as a
header for the supply of water to the minichannels. The inlet
and exit plenums are machined in the Lexan cover to minimize
any preheating of water before it enters the minichannels. A
circular cartridge heater is inserted in the copper block to provide a uniform heat flux along its length. A slight distortion
in the constant heat flux boundary condition is expected at the
microchannel surfaces due to the high thermal conductivity of
copper.
The sealing surface of the copper block in contact with the
Lexan cover is lapped to provide a good seal without any gasket.
The bottom face of the test section is insulated with a phenolic
piece. All three layers together form the test section, which is
held together by a set of ten mounting screws.
The test setup is shown in Figure 2. It consists of (1) the test
section, (2) the water supply loop, (3) the data acquisition system, and (4) a high-speed camera system. The data acquisition
system and the high-speed camera system are not shown in the
schematic of the test loop.
Figure 2 Schematic of the test loop.
heat transfer engineering
The experimental procedures for preparing degassed water
and collecting experimental data are described next. An 8 M
deionized and degassed water source is used in all the experiments. The details of the degassing procedure are given by [8]
and [9].
A commercially available pressure cooker equipped with a
deadweight to attain a pressure of 100 kPa (15 psi) is used for
preparing degassed water. The pressure cooker is filled with deionized water and pressurized by heating on a hot plate. Once the
cooker attains the design pressure of 100 kPa, it is depressurized
by removing the deadweight. This process leads to vigorous
boiling, forcing the dissolved gases in the water along with steam
out of the chamber.
The deadweight is reapplied and the chamber is again pressurized by supplying heat. Steam continuously escapes from the
chamber while a constant pressure is maintained. The resulting
water is degassed to a corresponding saturation temperature of
121◦ C. The remaining dissolved air present in the water will
not precipitate as long as the temperature of the surfaces contacting water stays below 121◦ C. Thus, by using the pressure
cooker, the water is degassed and delivered to the test section
while maintaining a constant pressure.
The test section is assembled and insulated with fiberglass
felt, and the heat loss experiments are then conducted. The calibration chart for heat loss is plotted as a function of the supplied
power to the test section in Watts and the difference in temperature between the test section surface and the ambient. For
example, for one of the test section assemblies, a temperature
difference of 50◦ C between the surface of the test section and the
ambient had a corresponding heat loss of 3.26 W. The heat loss
data are then used in calculating the actual heat carried away by
water flowing through the test section.
The experiments are performed with degassed water. The water is drawn from the bottom of the pressure cooker and is passed
through a flat plate heat exchanger that provides the desired water inlet temperature to the test section. A flow meter is used to
measure the flow rate. The accuracy of the flow meter is 3% of
the full scale, or 0.25 cc/min. LabVIEW is used to monitor the
thermocouples used for measuring temperature. A silicon diaphragm differential pressure transducer is used for measuring
the pressure drop with a response time of 1 ms. The cartridge
heater is powered and the flow is started in the channels. The
mass flux is held constant while the input power to the cartridge
heater is varied through the desired range. Flow rate, test section
inlet and outlet temperatures, temperatures in the copper block
along the flow length, and the differential pressure are recorded.
The differential pressure is measured across the channel inlet
and the exit manifolds.
The visual images are acquired using a high-speed camera
after the system has reached steady state. It was observed that
steady state is reached when the time-dependent surface temperatures reach their respective constant values without any fluctuations for a time period of fifteen minutes, as read from the
vol. 26 no. 3 2005
22
P. BALASUBRAMANIAN AND S. G. KANDLIKAR
LabView System. An AF Micro Nikkor lens is used with the
high-speed camera to gather detailed images of specific features
and events. Most of the images were recorded with a recording
speed between 4,000 and 15,000 fps.
The experimental accuracy for measured quantities is as follows: T = 0.1◦ C, DP = 1.03 kPa, Volts = 0.05 V, and I =
0.005 Amp.
RESULTS
Experiments are conducted by keeping the mass flux constant
and increasing the heat flux to find the influence of the increase
in surface temperature on flow instability and maldistribution.
Subcooled water enters the channel and becomes quickly saturated; the length required to reach saturation (the length of
the subcooled region) is calculated from energy balance using
the applied wall heat flux (after correcting for the heat loss).
For the conditions tested, Table 1 gives the values of the calculated length required to reach saturation with the corresponding
average surface temperature and applied heat flux. As can be
seen from the table, the length required for saturation is small
compared to the overall channel length (63.5 mm). Most of the
reported flow visualization is along the saturated length of the
microchannel.
Figure 3 Bubble/slug growth and thin film nucleation, G = 112 kg/m2 s,
Ts = 109.5◦ C.
close to the lower wall of the channel, residing under the thin
film at the beginning of its coalescence.
The presence of nucleation in the thin film provides an important clue to the flow boiling heat transfer mechanism. It clearly
establishes the presence of nucleate boiling under annular flow
conditions. This fact needs to be incorporated in developing a
mechanistic model for flow boiling in microchannels.
Nucleation in Thin Liquid Film
Interface Velocity
The occurrence of thin film nucleate boiling was previously
reported by Kasza et al. [6]. Figure 3 shows bubble nucleation
and its subsequent growth into a slug in a channel. The bulk flow
in the channel is from left to right. The interesting phenomenon
of thin film nucleate boiling was observed and is highlighted in
the figure.
The video was taken at a 15,000 frames/s, and Frames (a–e)
are separated by an equal time interval of 0.067 ms. Frame (a)
shows the entire channel filled with water and a bubble beginning
to nucleate at the upper wall of the channel. It also shows some
small bubbles floating in the bulk fluid. Frames (b) and (c) show
that the bubble that nucleated at the upper wall of the channel fills
up the entire width of the channel, leaving a thin film of liquid
on the sides of the wall. Frame (d) shows the complete channel
filled with the slug, while another bubble starts to nucleate in
the thin film at the same location. Frame (e) shows a new bubble
The duration of the bubble nucleating in the thin film as
seen in Frame (e) to grow and coalesce with the slug was about
1.33 ms. The thin film nucleation was observed to be quite periodic. The cross-cursor in the camera software is calibrated to
measure the bubble diameter and film thickness. The measured
film thickness for this particular case after the slug fills the entire
channel length is approximately 50 µm.
Figure 4 shows the velocity of the liquid–vapor interface associated with bubble growth. This growth rate corresponds to
Figure 3, in which the sequence of bubble growth and slug formation is shown at a time interval of 0.067 ms.
Table 1 Length required for saturation (L sat ), G = 120 kg/m2 s
Heat flux q kW/m2
Average Ts in ◦ C
L sat in mm
208
230
251
273
296
316
104.4
105.8
107.1
108.3
109.2
110.2
26.35
23.89
21.86
20.68
18.63
17.31
heat transfer engineering
Figure 4 Bubble/slug growth rate, G = 112 kg/m2 s, Ts = 109.5◦ C.
vol. 26 no. 3 2005
EXPERIMENTAL STUDY OF FLOW PATTERNS, PRESSURE DROP, AND FLOW INSTABILITIES
23
Figure 5 Pressure drop fluctuations, G = 120 kg/m2 s, Ts = 104.4◦ C.
The velocity of the interface calculated from the image sequences is observed to be extremely high, on the order of 3.5 m/s.
As the bubble nucleates and is confined to the wall, it experiences a sudden reduction in its growth rate. This point can
be noted in Figure 4, when the bubble/slug length is close to
1 mm. After the bubble becomes confined to the channel wall,
it starts to grow toward the sides at a very high velocity. An
increase in velocity at this stage is due to an increase in the
evaporation rate of the bubble as it gets closer to the channel
walls.
Pressure Drop Fluctuations
Figures 5 and 6 show the pressure drop fluctuations as a
function of time. The data are recorded every 1 ms for a period of 10 s. The resulting frequency is 1 kHz (1/0.001s). Highspeed data acquisition is necessary, as it was shown in Figure 3
that some bubble nucleation occurs at a very high frequency.
The pressure drop data are obtained at a given mass flux and
for increasing heat fluxes. Adiabatic experiments were initially
performed, and results indicated a constant pressure drop
across the channel. This was done to check the influence of
noise level in the pressure drop data obtained for two-phase
flow.
It is noted from Figures 5 and 6 that the magnitude of the
pressure drop increases with an increase in surface temperature
for a given mass flux. It should also be noted that the peak-topeak variation of pressure drop remains almost constant for small
changes in the surface temperature. (This can also be seen later
Figure 6 Pressure drop fluctuations, G = 120 kg/m2 s, Ts =110.2◦ C.
heat transfer engineering
Figure 7 Flow maldistribution in parallel minichannels, G = 120 kg/m2 s,
Ts = 110.2◦ C. The bulk flow direction in channels is from left to right.
from Figure 13, which shows a plot of peak-to-peak variation
of pressure drop for higher temperatures.) The flow patterns
observed for this condition are described next.
Flow Maldistribution—Reverse Flow
Figure 7 shows the flow maldistribution and reverse flow condition in the parallel minichannels. All channels are seen to be
active, primarily with slug flow. The third channel from the top
is seen to be darker, as it was oxidized when tested previously
under single channel experiments. Slug flow in certain cases
also extends into the inlet header, causing severe flow maldistribution. As seen in Figure 7, the direction of the flow in some
channels is opposite to the bulk flow direction, which is from
left to right. This chaotic nature is predominately seen during
flow boiling in small diameter parallel channels.
Figure 8 shows a schematic representation of the two-phase
flow pattern seen in Figure 7. The bulk flow is from left to right.
The direction of the reverse flow in the channel is indicated by
dotted arrows, and the bulk flow is indicated by solid arrows. In
this case, there is always a thin film of water left between the
slug and the walls of the channel. The first two channels from
the top show the slug flowing in opposite directions to the bulk
flow in the channel. The third and fourth channels have slug flow
in the same direction as the bulk flow. The fifth channel has the
Figure 8 Schematic of flow maldistribution shown in Figure 7. The bulk flow
direction is from left to right.
vol. 26 no. 3 2005
24
P. BALASUBRAMANIAN AND S. G. KANDLIKAR
flow opposite to the bulk flow, which is similar to the flow in
channels one and two. The sixth channel has flow in the same
direction as the bulk flow.
Another important behavior that is noted in Figure 7 is the
change in the contact angles between the slug and the walls of
the channels. When the vapor slug flows in the reverse direction,
the head of the slug has a receding contact angle with the walls
of the channel. When the vapor slug flows in direction of the
bulk flow, the head of the slug has an advancing contact angle
with the walls of the channel. This change in contact angle is
due to the contact angle hysteresis.
Pressure Fluctuations and FFT Analysis
From the flow patterns described earlier, it can be seen that in
a network of parallel channels, each channel may have a different flow direction from the other, and this flow maldistribution
leads to severe pressure fluctuations. The corresponding plot
of pressure fluctuations with time is shown in Figure 6. The
pressure fluctuations do not follow a uniform pattern in parallel
minichannels. It is extremely complex to map the flow patterns
with the time-dependent pressure drop fluctuations because the
effect of pressure drop caused by each channel gets superimposed on the other, and hence, the pressure fluctuations become
somewhat chaotic in nature.
The use of signal analysis helps us to study this chaotic nature of pressure drop fluctuation in minichannels [7]. The MATLAB software [10] is used for performing the signal analysis
of the pressure fluctuations. Fourier analysis is performed using
discrete Fourier transform (DFT) as it breaks down the signal
into constituent sinusoids of different frequencies. The algorithm
used for computing the DFT is Fast Fourier Transform (FFT).
FFT transforms the time-dependent pressure domain data into
frequency domain data. The standard form of transformation for
engineering applications is given by the following equation:
N −1
1 − j2 kn
ck =
x(n) exp
(1)
N n=0
N
Figure 10 Dominant frequency as a function of surface temperature.
indicates a complex number; and ck is the complex value of the
Fourier coefficients. Frequencies above the Nyquist frequency
are ignored because they are due to the periodicity of the FFT and
not due to the higher frequency components in the signals. The
Nyquist frequency is defined as half of the sampling frequency
(1/2T) [11]. The following section gives the result of signal
analysis performed on pressure drop data obtained from this
network of parallel channels, and the trends are explained in
conjunction with the observed flow patterns.
The frequency of the measured pressure fluctuations was plotted against the magnitude of the Fourier coefficient to obtain the
power spectrum for all the test conditions. Figure 9 shows the
power spectrum for a surface temperature of Ts = 105.8◦ C.
These results are used to obtain information regarding the dominant frequency. For example, in the present case, the dominant
frequency is 1.6 Hz.
Figure 10 shows a plot of the dominant frequency as a function of the surface temperature. The dominant frequencies are
identified by the magnitude of the Fourier coefficient of the
pressure signals. There is an increase in the dominant frequency
with an increase in the surface temperature. A decrease in frequency is also seen for surface temperatures higher than 109◦ C.
This confirms the experimental observation of [7] for a single
where N represents the number of data points; n is the time
domain incrementing index (n = 0, 1, 2, . . . N − 1); x(n) is
the value of the data point in the time domain; k represents the
exponential signals and has the value k = 0, 1, 2, . . . ; N − 1, j
Figure 9 Fourier co-efficient of pressure signal, Ts = 105.8◦ C.
heat transfer engineering
Figure 11 Bubble nucleation at Ts = 104.5◦ C (a–e) and slug flow at
Ts =110.2◦ C (f–j) at the exit of the minichannel. The bulk flow direction is
from left to right.
vol. 26 no. 3 2005
EXPERIMENTAL STUDY OF FLOW PATTERNS, PRESSURE DROP, AND FLOW INSTABILITIES
25
Figure 12 Schematic of bubble nucleation at Ts = 104.5◦ C and slug formation at Ts = 110.2◦ C
minichannel, that the decreasing trend is due to the development
of slug flow, which has low frequency compared to the nucleating bubbles. This is further supported by our observation of
the flow patterns, which shows increased activity of slug flow at
high temperatures compared to the nucleating bubbles at lower
temperatures.
Figure 14 Peak-to-peak variation of differential pressure as a function of
surface temperature.
Bubble Nucleation and Slug Formation
Figure 11 shows Frames (a–j) separated by equal time steps
of 0.25 ms each. Frames (a–e) are taken at a surface temperature of Ts = 104.5◦ C, and Frames (f–j) are taken at a surface temperature of Ts = 110.2◦ C for the same mass flux of
G = 120 kg/m2 s. These frames are taken at the same channel
location. From Frames (a–e) a nucleating site is seen to be active
on the upper wall of the channel, close to the indicated vertical
line (k). This nucleation site gives out a train of bubbles nucleating at a very high frequency. This nucleation site is found
to be very active as long as it is covered with water. The same
location in Frames (f–j) at a higher temperature shows that this
nucleating site is not more active; instead, we see the occurrence
of slug flow. The slug flow takes over these nucleating sites, and
we see “thin film nucleation,” similar to the observations made
in Figure 3.
Effect of Surface Temperature on Fourier Coefficient
Figure 12 shows a schematic of the flow pattern shown in
Figure 11. Frames (a–e) and (f–j) in Figure 11 correspond to
Frames (1) and (2) in Figure 12, respectively. From the visual
observations, the calculated frequency of a complete slug formation is found to be lower than the bubble nucleation frequency.
From this, it can be concluded that slug flow corresponds to
the low frequency signature in the pressure drop fluctuations,
as compared to the nucleating bubbles, which occur at a much
higher frequency.
Figure 13 shows the magnitude of the Fourier coefficient at
the dominant frequency. The slope is seen to be almost constant
for lower surface temperatures. As slug flow becomes dominant
at higher temperatures, an increasing trend in the magnitude is
expected.
The corresponding plot of the magnitude of the dominant
frequency in Figure 13 shows that there is always an increase
in magnitude of the Fourier coefficient with an increase in the
surface temperature. It can be concluded that the bubbles nucleating at low surface temperatures have a high frequency but a
lower magnitude, and the slug flow occurring at a higher surface temperature has a lower frequency but a higher magnitude.
When temperature is lowered from 110.2◦ C to 104.5◦ C, these
nucleating sites become active again.
Figure 14 shows the peak-to-peak variation of channel differential pressure as a function of the surface temperature. The
magnitude of this variation is seen to be almost constant with
an increase in temperature. A similar observation was made by
Campbell and Kandlikar [7] in their low temperature data. Depending on the surface temperature, the slugs are formed either
at the inlet or toward the exit of the minichannels. This behavior is observed at all surface temperatures investigated in this
work. A wider range of operating conditions is required to further map the flow patterns with the pressure fluctuations and
their corresponding dominant frequencies.
CONCLUSIONS
The following conclusions were reached:
Figure 13 Magnitude of Fourier coefficient at dominant frequency as a function of surface temperature.
heat transfer engineering
1. An experimental study of pressure drop fluctuation and associated flow patterns in a set of parallel channels with a
hydraulic diameter of 333 µm is performed to identify the
dominant frequency and the corresponding flow patterns.
2. Bubble growth and its transition to slug formation is clearly
observed with a thin liquid film covering the walls of the
channel. Nucleation in the thin film was observed at a higher
vol. 26 no. 3 2005
26
3.
4.
5.
6.
P. BALASUBRAMANIAN AND S. G. KANDLIKAR
frequency compared to a single nucleating bubble in the
channel.
The velocity of the bubble/slug growth in a single channel
from a set of parallel minichannels is calculated from the
experimental visualization. The observed highest velocity is
3.5 m/s.
The dominant frequency of the pressure drop fluctuations increases with an increase in the surface temperature, indicating an increase in the bubble nucleation frequency. It shows a
decreasing trend for surface temperatures higher than 109◦ C
due to the slug formation, which has a lower frequency than
the bubble nucleation frequency.
Nucleate boiling is observed in the bulk liquid flow as well
as in the thin liquid film surrounding the vapor slug in the
channel. This confirms the dominance of nucleate boiling
heat transfer as reported by earlier investigators. The presence
of nucleate boiling in the thin film region of the slug flow (or
annular flow) provides an important insight for developing
accurate mechanistic models for flow boiling in minichannels
and microchannels.
Slug flow was observed to be dominant for higher surface
temperatures, and in some cases the resulting back flow extended into the inlet manifold, causing severe flow maldistribution.
NOMENCLATURE
ck
G
j
k
L sat
N
n
q T
Ts
Ts−amb
x(n)
complex value of the Fourier coefficients
mass flux, kg/m2 s
complex number
exponential signals, k = 0, 1, 2, . . . , N − 1
length required for saturation, mm
number of data points
time domain incrementing index, n = 0, 1, 2, . . .
N −1
heat flux, W/m2
sample time in seconds
surface temperature, ◦ C
difference between test section surface temperature
and the ambient temperature
value of the data point in the time domain
[3] Steinke, M. E., and Kandlikar, S. G., Flow Boiling and Pressure
Drop in Parallel Flow Microchannels, First International Conference on Microchannels and Minichannels, Rochester, New York,
ICMM2003-1070, April 24–25, 2003.
[4] Brutin, D., Topin, F., and Tadrist, L., Experimental Study of Unsteady Convective Boiling in Heated Minichannels, International
Journal of Heat and Mass Transfer, vol. 46, pp. 2957–2965, 2003.
[5] Peles, Y., Two-Phase Flow in Microchannels—Instabilities Issues and Flow Regime Mapping, First International Conference on Microchannels and Minichannels, Rochester, New York,
ICMM2003-1069, 2003.
[6] Kasza, K. E., Didascalou, T., and Wambsganss, M. W., Microscale
Flow Visualization of Nucleate Boiling in Small Channels: Mechanisms in Influencing Heat Transfer, Proc. International Conference on Compact Heat Exchanges for the Process Industries,
pp. 343–352, Begell House, Inc., New York, 1997.
[7] Campbell, L. A., and Kandlikar, S. G., Experimental Study of Heat
Transfer, Pressure Drop, and Dryout for Flow Boiling of Watger
in an Oil Heated Minichannel, paper being presented at the Second International Conference on Microchannels and Minichannels, Rochester, New York, 2004.
[8] Steinke, M. E., and Kandlikar, S. G., Control and Effect of Dissolved Air in Water during Flow Boiling in Microchannels, International Journal of Heat and Mass Transfer, vol. 47, nos. 8–9,
pages 1925–1935, 2004.
[9] Kandlikar, S. G., Steinke, M. E., and Balasubramanian, P., SinglePhase Flow Characteristics and Effect of Dissolved Gases on Heat
Transfer near Saturation Conditions in Microchannels, Paper No.
IMECE2002-32382, IMECE 2002, New Orleans, Nov. 17–21,
2002.
[10] Etter, D. M., Engineering Problem Solving with MATLAB, 2nd
ed., Prentice Hall, Upper Saddle River, NJ, 1996.
Prabhu Balasubramanian completed his graduate
studies from the Mechanical Engineering Department
at the Rochester Institute of Technology, Rochester,
New York. He received his B. S. in Mechanical Engineering from the University of Madras in 2000. His
M.S. thesis was on the study of heat transfer and
two-phase flow mechanism during flow boiling in
minichannels and microchannels, with applications
to electronics cooling and fuel cells. His other areas
of research interest include fuel cells, compact heat
exchangers, and CFD analysis. He has been a member
of ASME since 2000. Currently he is working with
MTI MicroFuel Cells Inc., located in Albany, New York.
REFERENCES
[1] Kew, P. A., and Cornwell, K., Correlations for the Prediction of Boiling Heat Transfer in Small-Diameter Channels, Applied Thermal Engineering, vol. 17, nos. 8–10, pp. 705–715,
1997.
[2] Yen, T. H., Kasagi, N., and Suzuki, Y., Forced Convective Boiling
Heat Transfer in Microtubes at Low Mass and Heat Fluxes, Symposium on Compact Heat Exchangers, A Festschrift on the 60th
Birthday of Ramesh K. Shah, Grenoble, France, pp. 401–406, 24
August 2002. Edizoni ETS, Pisa, Italy, 2004.
heat transfer engineering
Satish Kandlikar is the Gleason Professor of Mechanical Engineering at RIT. He received his Ph.D.
from the Indian Institute of Technology in Bombay
in 1975 and has been a faculty member there before
coming to RIT in 1980. His research is mainly focused in the area of flow boiling. After investigating
the flow boiling phenomena from an empirical standpoint, which resulted in widely accepted correlations
for different geometries, he started to look at the problem from a fundamental perspective. Using the highspeed photography techniques, he demonstrated that
vol. 26 no. 3 2005
EXPERIMENTAL STUDY OF FLOW PATTERNS, PRESSURE DROP, AND FLOW INSTABILITIES
small bubbles are released at a high frequency under flow conditions. His current
work involves stabilizing flow boiling in microchannels, interface mechanics
during rapid evaporation, and advanced chip cooling with single-phase liquid
flow. He has published over 100 journal and conference papers. He is the Heat in
History Editor of Heat Transfer Engineering and a fellow member of ASME, and
heat transfer engineering
27
has been the organizer of the two international conferences on Microchannels
and Minichannels sponsored by ASME. He is the founder of the ASME Heat
Transfer chapter in Rochester and founder and first Chairman of the E-cubed
fair—science and engineering fair for middle school students in celebration of
the Engineers Week.
vol. 26 no. 3 2005