Ankit Kalani
Department of Mechanical
Engineering & Microsystems Engineering,
Rochester Institute of Technology,
Rochester, NY 14623
e-mail: [email protected]
Satish G. Kandlikar1
Department of Mechanical
Engineering & Microsystems Engineering,
Rochester Institute of Technology,
Rochester, NY 14623
e-mail: [email protected]
Evaluation of Pressure Drop
Performance During Enhanced
Flow Boiling in Open
Microchannels With
Tapered Manifolds
Boiling can provide several orders of magnitude higher performance than a traditional
air cooled system in electronics cooling applications. It can dissipate large quantities of
heat while maintaining a low surface temperature difference. Flow boiling with microchannels has shown a great potential with its high surface area to volume ratio and latent
heat removal. However, flow instabilities and low critical heat flux (CHF) have prevented
its successful implementation. A novel flow boiling design is experimentally investigated
to overcome the above-mentioned disadvantages while presenting a very low pressure
drop. The design uses open microchannels with a tapered manifold (OMM) to provide
stable and efficient operation. The effect of tapered manifold block with varied dimension
is investigated with distilled, degassed water at atmospheric pressure. Heat transfer coefficient and pressure drop results for uniform and tapered manifolds with plain and microchannel chips are presented. The OMM configuration yielded a CHF of over 500 W/cm2
in our earlier work. In the current work, a heat transfer coefficient of 277.8 kW/m2 C
was obtained using an OMM design with an inlet gap of 127 lm and an exit gap of
727 lm over a 10 mm flow length. The OMM geometry also resulted in a dramatic reduction in pressure drop from 158.4 kPa for a plain chip and 62.1 kPa for a microchannel
chip with a uniform manifold, to less than 10 kPa with the tapered OMM design. A
tapered manifold (inlet and exit manifold heights of 127 and 727 lm, respectively) with
microchannel provided the lowest pressure drop of 3.3 kPa.
[DOI: 10.1115/1.4026306]
Keywords: open microchannels, OMM, flow boiling, electronics cooling, high heat flux,
uniform, and tapered manifolds
1
Introduction
Single phase cooling has been the dominant mode of heat transfer in electronics cooling for the last several decades due to its
low cost and reliable operation. Miniaturization and increase in
chip power densities have generated the recent need for high heat
flux dissipation. Boiling has the ability to dissipate large quantities
of heat due to its latent heat potential. Cooling with microchannels
has shown a great potential since the pioneering work of Tuckerman and Pease [1]. Colgan et al. [2] used enhanced microchannels
in a single phase study to dissipate heat fluxes of over 1 kW/cm2.
They were limited by the high chip temperature and large pumping power requirement. The state of research on flow boiling in
microchannels has been well reviewed by many researchers [3–5].
A brief literature review is presented with a focus on the heat
transfer and pressure drop performance of flow boiling in
microchannels.
For two-phase flow, Kandlikar [6] had pointed out that flow
instability [7], low heat transfer coefficient [8], and low critical
heat flux [9] were some of the critical issues responsible for the
poor performance of microchannel flow boiling systems. Different
techniques were proposed by a few researchers to prevent high
1
Corresponding author.
Contributed by the Heat Transfer Division of ASME for publication in the
JOURNAL OF HEAT TRANSFER. Manuscript received May 16, 2013; final manuscript
received December 10, 2013; published online February 26, 2014. Assoc. Editor:
Sujoy Kumar Saha.
Journal of Heat Transfer
pressure drop fluctuations and flow boiling instabilities in these
systems. Kandlikar et al. [10] used artificial nucleation sites and
inlet restrictors to provide stable flow boiling. Wang et al. [11]
studied flow boiling instability using three different inlet/outlet
configurations. Reduced pressure drop fluctuations were observed
with inlet restrictions, while no restrictions were needed on
the exit side. Wu and Cheng [12] and Lee et al. [13] used microchannels with trapezoidal cross sections in their flow boiling
study. Lu and Pan [14] achieved stability with artificial nucleation
sites using diverging, parallel microchannels as proposed by
Mukherjee and Kandlikar [15]. A design with evenly distributed
cavities along the channel showed the best performance. Hetsroni
et al. [16] used parallel triangular microchannels with varied experimental parameters. Zhang et al. [17] extensively studied the
Ledinegg instability in microchannels. They concluded that the
presence of inlet restrictors, increase in the system pressure and
the channel diameter, and reduction in the number of channels
and the channel length lead to a more stable flow in the
microchannels.
Balasubramanian et al. [18] used straight and expanding microchannels in their flow boiling study. The authors observed lower
pressure drops and wall temperature fluctuations with their
expanding microchannel geometry. Cho et al. [19] and Megahed
[20] used cross-linked microchannels in their experimental work.
Sitar et al. [21] used square parallel microchannels of 25 25 lm
and 50 50 lm cross sections with FC-72 and water. The authors
used a combination of inlet/outlet restrictors, inlet/outlet
C 2014 by ASME
Copyright V
MAY 2014, Vol. 136 / 051502-1
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Fig. 1 Schematic of the tapered manifold [27]
manifolds, and fabricated cavities to limit the instabilities. They
observed a reduction in the onset of nucleate boiling temperature
and an even flow distribution. Recently, Alam et al. [22] used
microgap on a printed circuit board to obtain low pressure drop in
the system.
High heat flux testing was also been undertaken by various
researchers. Qu and Mudawar [23] tested a microchannel heat
sink with 21 parallel channels and obtained a maximum heat flux
of 130 W/cm2. Kuo and Peles [24] used 200 lm 253 lm parallel
microchannel with structured reentrant cavities. Mass flux was
varied from 83 kg/m2s to 303 kg/m2 s. The authors concluded that
lower boiling incipience and increased CHF were observed
with structured reentrant cavities. Heat fluxes of up to 643 W/cm2
at 80 C wall superheat were recorded with a mass flux of
303 kg/m2s. Liu and Garimella [25] experimentally investigated
flow boiling in microchannels with inlet water temperatures of
67–95 C, and mass fluxes of 221–1283 kg/m2 s. The authors
obtained a heat flux of 129 W/cm2 at an exit quality of 0.2. Balasubramanian et al. [26] conducted experiments using their stepped
fin microchannel geometry. They obtained a maximum heat
dissipation of 400 W/cm2 at a wall superheat of around 20 C
with 664 kg/m2 s mass flux. Recently, Kandlikar et al. [27] and
Kandlikar [28] used OMM configuration as shown in Fig. 1 to
simultaneously increase the CHF and the heat transfer coefficient.
They obtained a heat flux of 506 W/cm2 at a wall superheat of
26.2 C without reaching CHF. Preliminary work with tapered
manifolds showed low pressure drop and high heat transfer
performance.
In the current work, the tapered microchannel configuration is
further investigated and its performance, especially related to the
pressure drop, is studied. Plain and microchannel chips are used
with distilled and degassed water as the working fluid at atmospheric pressure (at the exit). Three tapered manifolds having the
same inlet manifold height (127 lm) and gradually increasing exit
manifold heights (327, 527, and 727 lm, respectively) are used
and their effects on pressure drop are studied. Results are compared with the uniform manifold for both plain and microchannel
chips.
2
Experimental Setup
Figure 2 shows the flow boiling test setup used in the current
study. It is similar to that used in an earlier study by Kandlikar
et al. [27]. The test setup consisted of a redesigned heating block,
base plate, and a manifold block. The heater consisted of a copper
block with eight 200 W cartridge heaters. The top half of the
heater had three equally spaced holes for thermocouple probes
and the tip was a 10 mm 10 mm square section contacting the
test chips. The manifold block consisted of inlet and outlet openings for the fluid flow. The test section was placed on the ceramic
base plate and a gasket was used to seal the system. A 127 lm
thick rigid gasket was used for all the test runs. This presents a
fixed manifold height in the uniform manifold case, and the inlet
manifold height in the case of tapered manifolds.
Fig. 2 Schematic of the flow boiling test setup showing the
uniform and tapered manifolds (not to scale)
A Micropump pump was used to provide the desired flow rate.
A flow rate of 80 ml/min was used for the current work. Distilled
water from a supply tank was degassed initially before entering
the test section. The supply tubing from the tank to the system
was wrapped in fiberglass insulation to minimize heat losses. An
inlet subcooling of 10 C was employed in all tests.
The tapered manifold created a tapered gap above the test section, while the uniform manifold had no recess in the manifold
block as shown in Fig. 2. The tapered manifold was designed to
allow a larger flow cross-sectional area on the exit side of the fluid
path. Three different tapered manifolds were used, each with
the same inlet height (127 lm) and with gradually increasing
manifold heights of 327 lm, 527 lm, and 727 lm on the exit side.
Polysulfone was used for the manifold block.
Heat flux was calculated using the one-dimensional heat
conduction equation with the three temperatures measured in the
copper block
q00eff ¼ kCu
dT
dx
(1)
Three-point backward Taylor’s series approximation was used to
calculate the temperature gradient dT/dx
dT 3T1 4T2 þ T3
¼
dx
2Dx
(2)
R
A differential OmegaV pressure sensor was used to measure
the pressure drop. An NI cDaq-9172 data acquisition system with
NI-9213 temperature module and NI-9205 pressure module
was used to record the temperature and pressure, respectively. A
V
LABVIEW virtual instrument was used to display and record temperature, pressure and heat flux.
R
3
Test Section
A copper chip with overall dimensions of 20 mm 20 mm
3 mm thickness was used as the test chip. Only the central
10 mm 10 mm area was exposed to the fluid through an opening
in the gasket. A square 2 mm 2 mm groove, as shown in Fig. 3,
Fig. 3
Schematic of the microchannel copper chip
051502-2 / Vol. 136, MAY 2014
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Transactions of the ASME
Table 1 Manifold configuration and mass fluxes at inlet and outlet for microchannel chip
Plain
Open microchannel
Manifold
Taper height (lm)
Inlet height (lm)
Exit height (lm)
Ginlet (kg/m2 s)
Goutlet (kg/m2 s)
Ginlet (kg/m2 s)
Goutlet (kg/m2 s)
Uniform
Taper A
Taper B
Taper C
0
200
400
600
127
127
127
127
127
327
527
727
1050
1050
1050
1050
1050
408
253
183
372
372
372
372
372
238
175
138
was machined on the underside of the chip to reduce the heat
spreading effect. A thermocouple hole was provided in the test
section to measure the actual chip temperature. A microchannel
chip of 450 lm depth and 181 lm wide channels and 195 lm wide
fins was also tested. A CNC machine was used to make the microchannels on the copper chip.
The wall temperature at the top of the chip surface was estimated from the measured chip temperature and the temperature
drop across the distance x1, which is the distance between the thermocouple location and the top surface. The maximum heat flux
tested was limited to 300 W/cm2 due to the heater capacity. CHF
was not reached in any of the testing. The work was mainly
focused on the heat transfer coefficient and pressure drop in this
range of testing.
x1
(3)
Twall ¼ Tc q00eff
kCu
The heat losses from the test section to the working fluid were
reduced in the current setup since the exposed area of the test chip
in contact with the fluid was restricted to 10 mm 10 mm. The
remaining area was surrounded by a high temperature ceramic
insulating sleeve and the base ceramic block. Furthermore, numerical simulations were conducted to obtain the heat losses (less
than 3%) in the above geometry. Appropriate corrections were
applied in the heat flux calculations. A detailed uncertainty analysis for this is reported by Kandlikar et al. [27]. The flow uncertainty at the highest flow rate (333 ml/min) was 3%, and at the
tested flow rate it was 5%. The thermocouples have an accuracy
of 0.1 C. The estimated uncertainty in heat flux and surface temperature at high heat fluxes were 6% and 0.24 C, respectively.
The heat fluxes are calculated using the 10 mm 10 mm projected
area of the boiling surface. The uncertainty bars are similar to the
symbol sizes, and hence are not shown in the heat transfer performance plots. The heat fluxes are expressed in W/cm2 in the text
since it is customary in the electronics cooling application. However, all equations are based on the SI units.
4
G¼
m_
Ac
(4)
where the cross-sectional area Ac is the actual flow area at a given
section calculated as the sum of the total microchannel area and
the manifold gap area. The cross-sectional area at the inlet was
the same for all manifolds (3.39 106 m2), while the crosssectional area at the exit for the tapered manifold varied depending on the taper height. The maximum exit quality observed in the
current testing was below 0.1. The exit quality was calculated taking into the account the subcooling and was given by
00 1
q A
Cp DT
(5)
x¼
hfg
m_
where q00 is the heat flux, A is the projected area, m_ is the mass
flow rate in kg/s, Cp is the specific heat of water, DT is the degree
of subcooling, and hfg is the latent heat of vaporization.
4.1 Uniform Manifold Testing. The uniform manifold was
tested with plain and microchannel chips first so as to establish
the baseline results. The uniform manifold had a constant height
provided by the gasket over the chip. Both the inlet and exit manifolds had a height of 127 lm in this case.
Figure 4 shows the boiling performance of the two chips with
the uniform manifold. The plain chip showed a slight boiling
overshoot at a wall superheat of 17.7 C. A maximum heat flux of
227.1 W/cm2 at 22.1 C wall superheat was recorded for the plain
chip. The microchannel chip showed a similar overshoot but performed significantly better than the plain chip. A maximum heat
flux of 283.2 W/cm2 at a wall superheat of 12.9 C was obtained.
Testing was not continued at higher heat fluxes.
Figure 5 shows the heat flux and its corresponding pressure
drop for the uniform manifold. The plain chip showed a pressure
drop from 100 kPa at low heat fluxes to 158.4 kPa at high heat
fluxes. The introduction of the microchannel chip does show a
reduction in the pressure drop. At low heat fluxes, a pressure drop
Results
Both uniform and tapered manifolds were tested with the plain
and microchannel chips and the results for heat transfer and pressure drop are presented in this section. Heat flux was calculated
using Eq. (1) for a heater area of 100 mm2. The pressure data
were obtained using the differential pressure sensor. For all the
test runs, the flow rate was kept constant at 80 ml/min, and a gasket of thickness 127 lm was used to provide a fixed height at the
inlet manifold. The results showed good repeatability with variation within 3–5% in heat flux. The performances of a uniform
height manifold and three tapered manifold are presented.
The configurations for the uniform and the tapered manifolds
used in the current system are listed in Table 1. The inlet height
remains constant for all test runs, while the exit height changed
depending on the type of manifold used. Both the inlet and exit
heights were referenced from the top plane of the microchannel to
the top of the manifold. The mass fluxes at the inlet and outlet sections are calculated using the following equation:
Journal of Heat Transfer
Fig. 4 Boiling performance showing heat flux versus wall
superheat for plain and microchannel chip with uniform
manifold
MAY 2014, Vol. 136 / 051502-3
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Fig. 5 Pressure drop versus heat flux for plain and microchannel chip with uniform manifold
Fig. 7 Pressure drop performance for plain and microchannel
chips with tapered manifold C
Fig. 8 Boiling performance of plain chip with tapered and uniform manifold
Fig. 6 Boiling performance of plain and microchannel chips
with the tapered manifold C
of 40 kPa is observed. For high fluxes, a maximum pressure drop
of 62.1 kPa is seen at a heat flux of 250 W/cm2. The overall pressure drop fluctuation for the microchannel chip increased with the
increase in heat flux.
4.2 Tapered Manifold Testing. Tapered manifold was
designed to provide additional flow area in the manifold along the
flow direction to accommodate the vapor flow and reduce the
pressure drop.
Figure 6 compares the boiling performance of the uniform and
tapered manifold C for the microchannel and the plain chips.
Expectedly, the microchannel chip performed better than the plain
chip. The highest heat flux tested for the microchannel chip was
281.2 W/cm2 at 10.1 C wall superheat. For the plain chip, a maximum heat flux of 208.3 W/cm2 at a wall superheat of 15.6 C was
recorded. Boiling overshoot was not observed for both plain and
microchannel chips.
Figure 7 compares the pressure drops for the tapered manifold
C for the microchannel and the plain chips. The effect of the
tapered manifold is significant in terms of the pressure drop reduction over the uniform manifold for both chips. The plain chip
shows a maximum pressure drop of 6.3 kPa at a heat flux of
208.3 W/cm2, while the microchannel chip showed a pressure
drop of around 2 kPa for a similar heat flux. The solid points
shown on the graph are the average values over the pressure
range. At intermediate heat fluxes, small negative values of pressure drop are seen. However, it is quite infrequent and insignificant and is not affecting the heat transfer performance adversely.
Furthermore, no back flow was observed during these tests. The
plain chip shows an increasing trend of pressure drop with the
heat flux, while the microchannel chip shows only a slight
increase with the increasing heat flux.
4.3 Heat Transfer and Pressure Drop Performance With
the Plain Chip. The effect of both types of manifolds on plain
chip performance is discussed in this section. Figure 8 shows the
boiling performance of tapered and uniform manifolds. The
tapered manifold showed an improved performance compared
with the uniform manifold. The uniform manifold showed a boiling overshoot and recorded a heat flux of 227.1 W/cm2 at 22.1 C
wall superheat. The three tapered manifolds showed similar performances to one another. Hence, the effect of taper height itself
was not significant in this case as seen from Fig. 8. Boiling overshoot was not observed with any of the tapered manifolds. The
tapered manifold B recorded a heat flux of 255.1 W/cm2 at a wall
superheat of 17.7 C.
The pressure drop performance of the plain chip with the uniform and the three different tapered manifolds is shown in Fig. 9.
In most cases, the error bars were either similar size as the symbol
size or smaller. The tapered manifolds show a significant
pressure drop reduction compared with the uniform manifold. The
uniform manifold showed the highest pressure drop values at both
low and high heat fluxes. For the tapered manifold, the values
were below 20 kPa over the entire range. The highest pressure
drop was observed with the tapered manifold B at a heat flux of
255.1 W/cm2 of 19.6 kPa. Tapered manifold C showed the lowest
pressure drop over the entire heat flux range. A maximum pressure drop of 6.3 kPa at 208.3 W/cm2 heat flux was recorded for the
tapered manifold C. The introduction of a tapered manifold was
thus seen to drastically reduce the pressure drop from 158.4 kPa
(uniform) to 6.3 kPa (tapered manifold C) for similar heat flux
values.
4.4 Heat Transfer and Pressure Drop Performance With
the Microchannel Chip. Results of the microchannel chip with
the uniform and the tapered manifolds are discussed in this
051502-4 / Vol. 136, MAY 2014
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Transactions of the ASME
Fig. 9 Pressure drop performance of plain chip with uniform
and tapered manifolds
Fig. 11 Pressure drop performance of microchannel chip with
uniform and tapered manifolds
Fig. 10 Boiling performance of the microchannel chip with uniform and tapered manifolds
Fig. 12 Boiling performance comparison for tapered manifold
with microchannel chip and uniform manifold with both plain
and microchannel chips
section. Figure 10 shows the heat flux versus wall superheat plots
for the two types of manifolds with the microchannels. Unlike the
plain chip performance, the effect of varying the taper was
observed to affect the heat transfer performance. The tapered
manifold C recorded a heat flux of 281.2 W/cm2 at a wall superheat of 10.1 C. The uniform manifold however showed a better
performance than the tapered manifold A, although the testing
was not continued to higher heat fluxes. Earlier results from Kandlikar et al. [27] showed the tapered manifold performed better at
higher heat fluxes. CHF was not reached for any of the tests,
hence showing potential for greater heat dissipation. The tapered
manifold B dissipated a heat flux 239.1 W/cm2 at a wall superheat
of 8.6 C. The slope of the tapered manifold A curve suggests that
at higher heat fluxes, it might perform better than the uniform
manifold.
Three tapered manifolds were tested with the microchannel
chip and their pressure drop performance is shown in Fig. 11. The
maximum pressure drop observed was 11 kPa for tapered manifold A at a heat flux of approximately 170 W/cm2. A maximum
pressure fluctuation of around 7 kPa was observed for all three
tapered manifolds. The tapered manifold C showed the lowest
pressure drop of 3.3 kPa at a heat flux of 281.2 W/cm2 in comparison with the other two tapered manifolds. A maximum pressure
drop of 10 kPa was observed with the tapered manifold B, and it
showed lower pressure fluctuations compared with the tapered
manifold A.
5
Discussion
In this section, the flow boiling performances of the uniform
and the three tapered manifolds for both plain and microchannel
Journal of Heat Transfer
chips are compared. The results of heat transfer and pressure drop
are further analyzed.
5.1 Comparison Between the Microchannel and the Plain
Chip With Uniform and Tapered Manifold. Figure 12 shows
the heat transfer performance for tapered manifold C with the
microchannel chip and the uniform manifold with both chips.
Expectedly, the microchannel chip showed significant performance improvement compared with the plain chip for both manifolds. The introduction of the tapered manifold yields similar
performance at the midrange heat fluxes (between the microchannel chip with different manifolds), but the heat flux is seen to rise
for a given wall superheat at higher heat fluxes. The maximum
heat flux obtained with the tapered manifold was greater than
uniform manifold with the microchannel chip. Both plain and
microchannel chips showed a small temperature overshoot with
the uniform manifold. No temperature overshoot was observed
with the tapered manifold.
Figure 13 shows pressure drop versus the corresponding heat
flux with the uniform manifold and the tapered manifold C. The
highest pressure drop was observed with the uniform manifold
with the plain chip. At high heat fluxes (225 W/cm2), a pressure
drop of 158.4 kPa was recorded with the plain chip. At a similar
heat flux, the microchannel chip with a uniform manifold recorded
a pressure drop of 50 kPa. The reduction in the pressure drop was
mainly due to the increase in the flow cross-sectional area provided by the microchannels. The tapered manifold C showed the
lowest pressure drop of 2 kPa at a heat flux of 225 W/cm2. The
combination of tapered manifold with the microchannel chip
clearly showed a significant pressure drop reduction over the
entire range of heat flux. The expanding cross-sectional area along
MAY 2014, Vol. 136 / 051502-5
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Table 2 Summary of all test runs for plain and microchannel
chips, including maximum heat flux, wall superheat, heat transfer coefficient, and pressure drop
Fig. 13 Pressure drop performance comparison with tapered
manifold C with microchannel chip and uniform manifold with
both chips
Fig. 14 Comparison of heat transfer coefficient for plain and
microchannel chip with tapered manifolds B and C
the flow direction was able to accommodate the increased vapor
flow and resulted in an extremely low pressure drop. The overall
increase in the pressure fluctuation with increasing heat flux is
also limited for the tapered manifold, hence showing a more stable
flow with the tapered manifold in comparison with the uniform
manifold, while simultaneously offering better heat transfer performance in terms of higher heat flux at a given wall superheat.
5.2 Comparison Between the Microchannel Chip and the
Plain Chip With the Tapered Manifold. Figure 14 shows the
heat transfer coefficient versus heat flux for the microchannel and
the plain chips with the tapered manifolds B and C. The results
for the tapered manifold A were not included in the figures so as
to avoid overcrowding of data points. Heat transfer coefficient is
an important parameter in comparing the thermal performance of
different surfaces. The microchannel chip performed significantly
better than the plain chip for both tapered manifolds. At higher
heat fluxes, a maximum heat transfer coefficient of 277 kW/m2 C
for both tapers was recorded. A maximum heat transfer coefficient
of 144.3 kW/m2 C at 250 W/cm2 heat flux was observed for the
plain chip with taper B. For taper A, similar results were obtained.
Table 2 shows the maximum values for heat flux, wall superheat, heat transfer coefficient, and corresponding pressure drop for
plain and microchannel chips. Taper C with the microchannel
chip showed the best performance in terms of pressure drop and
heat transfer. For the plain chip, similar heat transfer performance
was obtained for all three tapers. The tapered manifolds showed
significant pressure drop reductions compared with the uniform
manifold with the plain chip.
The tapered manifolds with microchannel chips yielded a dramatic enhancement in heat transfer performance, while providing
Chip
Manifold
q00max ,
W/cm2
DTsat,
C
h,
kW/m2 C
DP,
kPa
Plain
Uniform
Taper A
Taper B
Taper C
227.1
228.6
255.1
208.3
22.1
15.8
17.7
15.6
102.4
144.4
144.3
133.5
158.4
12.6
19.6
6.3
Microchannel
Uniform
Taper A
Taper B
Taper C
283.2
263.8
239.1
281.2
12.9
14.1
8.6
10.1
217.9
186.7
277.6
277.8
62.1
7.5
6.2
3.3
an extremely low pressure drop. This feature makes them particularly suited to cooling the high performance IC chips. The
low pressure drop feature provides a very high coefficient of performance (ratio of heat removed to pumping power) that is particularly attractive for the 3D IC chip cooling architecture. For the
tapered manifold C with the microchannel chip, a heat flux of
281.2 W/cm2 is dissipated at a wall superheat of 10.1 C with a
heat transfer coefficient of 277.8 kW/m2 C. The corresponding
pressure drop was only 3.3 kPa. Further performance enhancements are expected with optimizing the microchannel geometry
and the taper configuration.
The main mechanism responsible for reducing the pressure
drop with the tapered manifolds is the gradual increase in the flow
cross-sectional area as the vapor is generated along the flow direction. As seen from Table 2, the cross-sectional area increases for
tapered manifolds, and the pressure drop is corresponding lower.
However, as reported by Kandlikar et al. [27], increasing taper
further does not continue this trend. The results for plain chips are
affected due to the presence of backflow under relatively low heat
flux conditions. The liquid flows through the microchannels promoting nucleation and is responsible for delaying the CHF. Further work on establishing the CHF limits for these configurations
is suggested by redesigning the heater unit to deliver higher heat
fluxes. This work is continuing in the authors’ lab.
6
Conclusions
The current work involves an experimental investigation of
flow boiling performance with a plain chip and a microchannel
chip using four different manifolds: one uniform and three
tapered. The testing was limited to heat fluxes below about
300 W/cm2 due to the heater limitations. Distilled and degassed
water at atmospheric pressure at a flow rate of 80 ml/min was
used for all test runs with an inlet subcooling of 10 C.
Three tapered manifolds with a gradual increase in the gap toward the exit with an inlet gap of 127 lm and the exit gaps of
327 lm, 527 lm, and 727 lm and a uniform manifold of a 127 lm
gap were tested with the microchannel and the plain chips in the
current setup:
(1) A heat flux of 227.1 W/cm2 at a wall superheat of 22.1 C
was recorded for the uniform manifold with a plain chip,
while a heat flux of 283.2 W/cm2 at 12.9 C wall superheat
was recorded for the microchannel chip with the same
manifold. CHF was not reached in any of the tests.
(2) The combination of the microchannel chip and the tapered
manifold significantly reduced the pressure drop in the system. Taper C (with inlet and exit manifold heights of 127
and 727 lm above the top surface of the chip) with a microchannel chip (450 lm depth, 181 lm wide channels, and
195 lm wide fins) showed the best performance with the
lowest pressure drop of 3.3 kPa compared with the
158.4 kPa pressure drop with the plain chip and the uniform
manifold.
051502-6 / Vol. 136, MAY 2014
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms
Transactions of the ASME
(3) A heat flux of 281.2 W/cm2 at 10.1 C wall superheat with
taper C was recorded with the microchannel chip. The
microchannel chip with the tapered manifolds showed significant performance improvement compared with the plain
chip with the uniform manifold.
(4) Similar improved performance in heat transfer coefficient
for the microchannel chip with tapered manifold was
observed in comparison with the plain chip with tapered
manifold. A maximum heat transfer coefficient of
277.8 kW/m2 C was recorded with microchannel chip and
taper C.
(5) The testing was not conducted to the CHF limit, which was
reported to be higher than 500 W/cm2 in an earlier publication [26]. The comparison presented here shows that the
microchannel chip with the taper C has the best heat transfer performance among the chips and manifolds tested.
(6) The main mechanism for the dramatic reduction in pressure
drop is due to the increased flow cross-sectional area to
accommodate the vapor generated along the flow direction.
This combines the inherent benefits of microchannels in
providing a superior heat transfer performance, with the
flow stability and low pressure drop due to the tapered
manifold.
(7) The open microchannels with tapered manifolds (OMM)
configuration were able to provide significant heat transfer
coefficient and heat transfer enhancements at dramatically
reduced pressure drops. This configuration holds promise in
overcoming the limitations posed by the microchannels
during flow boiling as pointed out by Kandlikar [6]. Additional testing to cover a higher exit quality, a broader range
of operating parameters, and additional manifold/microchannel geometry combinations up to the CHF limits of the
OMM configuration are proposed.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
Acknowledgment
The work was conducted in the Thermal Analysis, Microfluidics and Fuel Cell Laboratory at the Rochester Institute of Technology in Rochester, NY. This material is based upon the work
supported by the National Science Foundation under Award No.
CBET-1236062.
Nomenclature
Ac ¼
Cp ¼
G¼
hfg ¼
h¼
kCu ¼
m_ ¼
q00 ¼
Tc ¼
Tsat ¼
Twall ¼
DTsat ¼
x¼
cross-sectional area, m2
specific heat capacity, J/kg K
mass flux, kg/m2 s
latent heat of vaporization, kJ/kg
heat transfer coefficient, W/m2 C
thermal conductivity of copper, W/m C
mass flow rate, kg/s
heat flux, W/m2
chip temperature, C
saturation temperature, C
wall temperature, C
wall superheat, C
distance, m
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
References
[1] Tuckerman, D. B., and Pease, R. F. W., 1981, “High-Performance Heat Sinking
for VLSI,” IEEE Electron Device Lett., 2(5), pp. 126–129.
[2] Colgan, E. G., Furman, B., Gaynes, M., Graham, W. S., LaBianca, N. C.,
Magerlein, J. H., Polastre, R. J., Rothwell, M. B., Bezama, R. J., Choudhary,
R., Marston, K. C., Toy, H., Wakil, J., Zitz, J. A., and Schmidt, R. R., 2007, “A
Journal of Heat Transfer
[27]
[28]
Practical Implementation of Silicon Microchannel Coolers for High Power
Chips,” IEEE Trans. Compon. Packag. Technol., 30(2), pp. 218–225.
Thome, J. R., 2004, “Boiling in Microchannels: A Review of Experiment and
Theory,” Int. J. Heat Fluid Flow, 25(2), pp. 128–139.
Bertsch, S., Groll, E., and Garimella, S., 2008, “Review and Comparative Analysis of Studies on Saturated Flow Boiling in Small Channels,” Nanoscale
Microscale Thermophys. Eng., 12(3), pp. 187–227.
Tibiriça, C. B., and Ribatski, G., 2013, “Flow Boiling in Micro-Scale
Channels—Synthesized Literature Review,” Int. J. Refrig., 36(2), pp.
301–324.
Kandlikar, S. G., 2002, “Fundamental Issues Related to Flow Boiling in
Minichannels and Microchannels,” Exp. Therm. Fluid Sci., 26(2–4), pp.
389–407.
Kandlikar, S. G., 2004, “Heat Transfer Mechanisms During Flow Boiling in
Microchannels,” ASME J. Heat Transfer, 126(1), pp. 8–16.
Steinke, M. E., and Kandlikar, S. G., 2004, “An Experimental Investigation of
Flow Boiling Characteristics of Water in Parallel Microchannels,” ASME J.
Heat Transfer, 126(4), pp. 518–526.
Bergles, A. E., and Kandlikar, S. G., 2005, “On the Nature of Critical Heat Flux
in Microchannels,” ASME J. Heat Transfer, 127(1), pp. 101–107.
Kandlikar, S. G., Kuan, W. K., Willistein, D. A., and Borrelli, J., 2006,
“Stabilization of Flow Boiling in Microchannels Using Pressure Drop
Elements and Fabricated Nucleation Sites,” ASME J. Heat Transfer, 128(4), pp.
389–396.
Wang, G., Cheng, P., and Bergles, A. E., 2008, “Effects of Inlet/Outlet Configurations on Flow Boiling Instability in Parallel Microchannels,” Int. J. Heat Mass
Transfer, 51(9–10), pp. 2267–2281.
Wu, H. Y., and Cheng, P., 2004, “Boiling Instability in Parallel Silicon Microchannels at Different Heat Flux,” Int. J. Heat Mass Transfer, 47(17–18), pp.
3631–3641.
Lee, P. C., Tseng, F. G., and Pan, C., 2004, “Bubble Dynamics in Microchannels. Part I: Single Microchannel,” Int. J. Heat Mass Transfer, 47(25), pp.
5575–5589.
Lu, C. T., and Pan, C., 2009, “A Highly Stable Microchannel Heat Sink for
Convective Boiling,” J. Micromech. Microeng., 19(5), p. 055013.
Mukherjee, A., and Kandlikar, S. G., 2005, “Numerical Study of the Effect
of Inlet Constriction on Flow Boiling Stability in Microchannels,” Proceedings
of the 3rd International Conference on Nanochannels, Microchannels and
Minichannels, Toronto, Canada, June 13–15, Paper No. ICNMM2005-75143.
Hetsroni, G., Mosyak, A., Pogrebnyak, E., and Segal, Z., 2005, “Explosive
Boiling of Water in Parallel Micro-Channels,” Int. J. Multiphase Flow, 31(4),
pp. 371–392.
Zhang, T., Tong, T., Chang, J.-Y., Peles, Y., Prasher, R., Jensen, M. K., Wen,
J. T., and Phelan, P., 2009, “Ledinegg Instability in Microchannels,” Int. J. Heat
Mass Transfer, 52(25–26), pp. 5661–5674.
Balasubramanian, K., Lee, P. S., Jin, L. W., Chou, S. K., Teo, C. J., and Gao,
S., 2011, “Experimental Investigations of Flow Boiling Heat Transfer and Pressure Drop in Straight and Expanding Microchannels—A Comparative Study,”
Int. J. Therm. Sci., 50(12), pp. 2413–2421.
Cho, E. S., Koo, J.-M., Jiang, L., Prasher, R. S., Kim, M. S., Santiago, J. G.,
Kenny, T. W., and Goodson, K. E., 2003, “Experimental Study on Two-Phase
Heat Transfer in Microchannel Heat Sinks With Hotspots,” Nineteenth Annual
IEEE Semiconductor Thermal Measurement and Management Symposium, pp.
242–246.
Megahed, A., 2011, “Experimental Investigation of Flow Boiling Characteristics in a Cross-Linked Microchannel Heat Sink,” Int. J. Multiphase Flow, 37(4),
pp. 380–393.
Sitar, A., Sedmak, I., and Golobic, I., 2012, “Boiling of Water and FC-72 in
Microchannels Enhanced With Novel Features,” Int. J. Heat Mass Transfer,
55(23–24), pp. 6446–6457.
Alam, T., Lee, P. S., Yap, C. R., Jin, L., and Balasubramanian, K., 2012,
“Experimental Investigation and Flow Visualization to Determine the Optimum
Dimension Range of Microgap Heat Sinks,” Int. J. Heat Mass Transfer,
55(25–26), pp. 7623–7634.
Qu, W., and Mudawar, I., 2003, “Measurement and Prediction of Pressure Drop
in Two-Phase Micro-Channel Heat Sinks,” Int. J. Heat Mass Transfer, 46(15),
pp. 2737–2753.
Kuo, C.-J., and Peles, Y., 2007, “Local Measurement of Flow Boiling in Structured Surface Microchannels,” Int. J. Heat Mass Transfer, 50(23–24), pp.
4513–4526.
Liu, D., and Garimella, S. V., 2007, “Flow Boiling Heat Transfer in Microchannels,” ASME J. Heat Transfer, 129(10), pp. 1321–1332.
Balasubramanian, K., Lee, P. S., Teo, C. J., and Chou, S. K., 2013, “Flow Boiling Heat Transfer and Pressure Drop in Stepped Fin Microchannels,” Int. J.
Heat Mass Transfer, 67, pp. 234–252.
Kandlikar, S. G., Widger, T., Kalani, A., and Mejia, V., 2013, “Enhanced Flow
Boiling Over Open Microchannels With Uniform and Tapered Gap Manifolds,”
ASME J. Heat Transfer, 135(6), p. 061401.
Kandlikar, S. G., 2013, “Heat Transfer Enhancement Through Flow Field
Incorporating a Taper,” patent pending.
MAY 2014, Vol. 136 / 051502-7
Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 04/15/2014 Terms of Use: http://asme.org/terms