Measurement and prediction of heat transfer coefficient on ammonia flow boiling in a microfin plate evaporator

Accepted Manuscript Measurement and prediction of heat transfer coefficient on ammonia flow boiling in a microfin plate evaporator Kohei Koyama , Hirotaka Chiyoda , Hirofumi Arima , Akio Okamoto , Yasuyuki Ikegami PII:

S0140-7007(14)00117-0

DOI:

10.1016/j.ijrefrig.2014.05.005

Reference:

JIJR 2781

To appear in:

International Journal of Refrigeration

Received Date: 7 February 2014 Revised Date:

8 May 2014

Accepted Date: 10 May 2014

Please cite this article as: Koyama, K., Chiyoda, H., Arima, H., Okamoto, A., Ikegami, Y., Measurement and prediction of heat transfer coefficient on ammonia flow boiling in a microfin plate evaporator, International Journal of Refrigeration (2014), doi: 10.1016/j.ijrefrig.2014.05.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Measurement and prediction of heat transfer coefficient on ammonia flow boiling in a

RI PT

microfin plate evaporator

Kohei Koyamaa*, Hirotaka Chiyodab, Hirofumi Arimaa, Akio Okamotoc and Yasuyuki Ikegamia Institute of Ocean Energy, Saga University, 1-48 Hirao, Kubara-aza, Yamashiro-cho, Imari,

SC

a

b

M AN U

Saga 849-4256, Japan

Department of Mechanical Engineering, Saga University, 1 Honjo-machi, Saga, Saga,

840-8502 Japan

Titanium Technology Department, Kobe Steel, Ltd., Shinagawa-ku, Tokyo 141-8688, Japan

* Corresponding author

TE D

c

Tel: +81-955-20-2190, Fax: +81-955-20-2191

AC C

EP

E-mail: [email protected]

1

ACCEPTED MANUSCRIPT

Abstract Thermal characteristics of ammonia flow boiling in a microfin plate evaporator are

RI PT

experimentally investigated. Titanium microfin heat transfer surface is manufactured to enhance boiling heat transfer. Longitudinally- and laterally-microfined surfaces are used and those

SC

performances are compared. Heat transfer coefficient of microfin plate evaporator is also

M AN U

compared with that of plain-surface plate evaporator. The effects of mass flux, heat flux, channel height, and saturation pressure on heat transfer coefficient are presented and discussed. The experiments are conducted for the range of mass flux (5 and 7.5 kg m-2 s-1), heat flux (10, 15, and 20 kW m-2), channel height (1, 2, and 5 mm), and saturation pressure (0.7 and 0.9 MPa).

TE D

Heat transfer coefficient is compared with that predicted by available empirical correlations proposed by other researchers. Modified correlations using Lockhart-Martinelli parameter to

EP

predict heat transfer coefficient are developed and they cover more than 87% of the

AC C

experimental data.

Keywords

Evaporation, Heat transfer coefficient, Heat transfer enhancement, Microfin, Ammonia, Titanium

2

ACCEPTED MANUSCRIPT

1. Introduction Electric power generation systems using renewable energy are necessary to overcome

RI PT

environmental issues, to address depletion of fossil fuel, and to develop sustainable society. Power generation system based on small temperature difference is an attractive option as an

SC

environmentally-friendly system. Small temperature difference exists in the ocean, wasted heat

M AN U

from industry, and hot spring. Power generation system using these temperature differences can supply electricity steadily compared with other renewable energy source such as wind or solar power generations. This kind of system generally consists of pump, evaporator, turbine, and condenser. Working fluid is supplied to evaporator in which the liquid working fluid is turned

TE D

into vapor. The vapor is fed into turbine at which generator is connected. The vapor leaving from turbine is supplied to condenser in which vapor is returned into liquid. Development of

EP

each component of the power generation system is necessary to improve its thermal efficiency.

AC C

This study focuses on thermal performance of an evaporator. As discussed in Uehara and Nakaoka (1985), minimizing heat transfer area is required to develop efficient ocean thermal energy conversion plant, which uses temperature difference between shallow and deep seawater. Heat source for electric power generation system using small temperature difference is limited. Hence, improvement of thermal performance of an evaporator is required to develop efficient generation system.

3

ACCEPTED MANUSCRIPT

Selection of refrigerant as working fluid is an important factor for power generation systems as well. Working fluid experiences evaporation and condensation in the power cycle.

RI PT

Not only thermophysical properties but also ozone depletion potential (ODP) and global warming potential (GWP) should be considered for selecting working fluid. Basic potentials of

SC

refrigerants were reviewed by Thome et al. (2008) and therefore ammonia is selected as a

M AN U

working fluid in this study.

The current investigation is a subsequent work of Koyama et al. (2013). Flow boiling of ammonia in a plate evaporator was explored to clarify its basic thermal characteristics. In the current study the heat transfer surface of the evaporator is microfinned to enhance heat transfer.

TE D

Microfin structure on a heat transfer surface has been attracted by many researchers to enhance boiling heat transfer. Fujii et al. (1995) revealed heat transfer and pressure drop of

EP

HCFC 22 in a grooved copper tube. They found that heat transfer coefficients for wavy-annular and annular flow in grooved tube were about two to four times as high as those in smooth tube.

AC C

#1-2

Chamra et al. (1996) compared pressure drop and heat transfer coefficient of R22 evaporation in four commercial microfin tubes. They reported that the highest performance was achieved on a cross-grooved tube with 20° helix angle. Yu et al. (2002) investigated R134a flow boiling in horizontal smooth and microfin tubes to present flow pattern map and heat transfer coefficient. They observed wavy, intermittent, semi-annular, and annular flows in their experimental range

4

ACCEPTED MANUSCRIPT

and they produced flow pattern map including these flow patterns. Kim and Shin (2005) measured and compared heat transfer coefficient of R22 and R410A flow boiling in smooth and

RI PT

microfin tubes. They reported that higher heat transfer performance was obtained by using R410A due to higher thermal conductivity. Bandarra Filho et al. (2004) measured pressure drop

SC

of convective boiling of R134a in horizontal smooth and microfin tubes and they developed a

M AN U

correlation using Martinelli parameter. Zhiping et al. (2001) investigated saturated boiling heat transfer in tubes which have plain, axially-grooved, and sintered porous surfaces with ethanol and distilled water. They showed that plain and grooved tube can enhance heat transfer by reducing gap size. They presented empirical correlations as a function of bond number to

TE D

calculate heat transfer coefficient of plain tube. Thome (1996), Del Col et al. (2002), and Zhang (2011) summarized mechanisms of heat transfer enhancement of microfin tubes as follows:

EP

(1) Increase in wetted surface area per unit length.

AC C

(2) Increase in liquid-phase convective heat transfer due to the effect of flow disturbance and turbulence induced by fin geometry. (3) Increase in wetting region around the circumference of a tube due to capillary forces. (4) Increase in nucleation heat transfer and nucleation sites between fins. (5) Increase in turbulence and forces entrained liquid droplets to the wall due to the effect of swirl on mist flow at high vapor quality.

5

ACCEPTED MANUSCRIPT

This study investigates thermal characteristics of ammonia flow boiling in a microfin plate evaporator. A few investigations of ammonia flow boiling in microfin passage have been

RI PT

conducted so far. Momoki et al. (2009) reported heat transfer of ammonia flow boiling in a horizontal electrically-heated spirally-grooved tube. They found insignificant effect of heat flux

SC

on heat transfer coefficient at small mass flux. Kelly et al. (2002a, 2002b) measured heat

M AN U

transfer coefficient and pressure drop in annular tube counter-flow heat exchanger in which ammonia was flowing in inner tube and liquid R134a is flowing in outer tube. They reported that microfin greatly increases heat transfer coefficient at low flow rate. They noted that available correlations to predict heat transfer coefficient were insufficient. More comprehensive

TE D

study and modeling are necessary to explore thermal characteristics of ammonia flow boiling in a microfin device. In this study, the effect of microfin orientation on heat transfer coefficient is

EP

firstly investigated and experimental results are compared with those of previous work Koyama

AC C

et al. (2013). Then the effects of mass flux, heat flux, channel height, and saturation pressure on heat transfer coefficient are discussed. Finally the measured heat transfer coefficient is compared with that predicted by available empirical correlations. Modified empirical correlations to predict heat transfer coefficient of the current study are developed at the end of discussion.

6

#1-2

ACCEPTED MANUSCRIPT

2. Experimental 2.1 Experimental setup

RI PT

Figure 1 depicts a schematic diagram of experimental setup, which consists of working fluid, hot water, and two cold water loops. The working fluid loop is constructed by stainless

SC

steel tube. The hot and cold water loops are constructed by plastic tube and reinforced rubber

M AN U

hose. The working fluid loop includes an evaporator (test section), condenser, pre-heater, and sub-cooler as well as pump and tank. Brazed heat exchangers (Tokyo Braze, HES-Ti-05H) are used as the pre-heater and sub-cooler. Hot water loop is connected to the pre-heater at which a temperature of the working fluid is controlled. An electric heater (Hakko Electric, BWA1120) is

TE D

used to produce hot water in this loop. The primary cold water loop is connected to the condenser and working fluid tank to cool working fluid. The secondary cold water loop is

EP

connected to sub-cooler. Cold water is produced by using chillers (Orion, RKE1500B-V). The

AC C

working fluid is heated in the evaporator. Electric heaters (Kyusyu Nissho, 100V-150W) are used as heat source and heat flux is controlled by adjusting voltage. Measuring equipment is listed in Table 2. Measured data are accumulated by a multimeter (Keithley, Model 2701) and are stored in a personal computer with LabVIEW. Figure 2 shows schematic illustrations of the plate evaporator used in this study. A stainless steel cover, Teflon packing (not depicted here), Titanium base plate, heater block, and

7

#1-3

ACCEPTED MANUSCRIPT

thermal insulator are stacked. The gap between the cover and base plate is assigned as passage of the working fluid. For the sake of simplicity, holes for bolts and nuts to stack all components

RI PT

are not illustrated in Fig. 2.

Figure 3 shows heat transfer surface of the plate evaporator. Photoresist is patterned to

#1-4

SC

manufacture grooves on 0.4-mm-thick titanium sheet, and then the plate is etched. The rest of

M AN U

titanium sheet between grooves is designated as microfin. Dimension of the microfin is depicted in Fig. 3. Ten-point mean surface roughness of top of the microfin is 0.6 µm and that of bottom of the fin is 5 µm. The grooved sheet is soldered on the titanium base plate. Thickness of the soldering portion is 0.07 mm and no defect is confirmed by ultrasonic testing. This study

TE D

measures local heat transfer coefficient as a function of the vapor quality for two heat transfer surfaces called longitudinally- and laterally-finned. Directions of working fluid flow for each #1-5

EP

surface are schematically illustrated in Fig. 3.

AC C

The heat transfer plate is composite structure as shown in Fig. 3. The effective thermal conductivity of the heat transfer plate keff is obtained as k eff =

l1 + l2 l1 + l2 − ls ls + k1 ks

(1)

where l1 is distance between two thermocouples, l2 is distance between thermocouple and heat transfer surface, and ls is thickness of brazing portion. k1 and ks are thermal conductivity of titanium (k1 = 22 W m-1 k-1) and brazing material (ks = 338 W m-1 K-1), respectively. Substituting 8

#2-1

ACCEPTED MANUSCRIPT

these values into Eq. (1) keff = 22.03 W m-1 K-1 is obtained. 2.2 Data reduction

RI PT

Local heat transfer coefficient along the direction of working fluid flow and thermal equilibrium vapor quality are obtained in this study. Figure 4 shows detail of the heat transfer

one-dimensional heat conduction derives Eq. (2) T1 − T2 l1

M AN U

q=k

SC

plate. A pair of thermocouples is used for calculation of local heat flux. Assumption of

(2)

where T1 and T2 are temperatures in the heat transfer plate and k is thermal conductivity of the heat transfer plate. Electric heaters equipped on back of the plate evaporator can heat uniformly

TE D

heat transfer plate. Average and maximum heat losses from the evaporator to atmosphere are confirmed as 6.28% and 8.32% of heating power, respectively. This means that the most of the

EP

heating power is used for heating the working fluid. Temperatures in the heat transfer plate are

AC C

measured by thermocouples at 10 sites as shown in Fig. 2. The thermocouples are grouped into two clusters: 5 of 10 thermocouples are close to heater block and other 5 are adjacent to working fluid channel. Temperature variation in each cluster was less than 7.66% for all experimental data. Temperature distribution on heat transfer surface can be reasonably negligible and therefore surface temperature of the heat transfer plate is regarded as isothermal. This means that heat from the heater block is transferred to working fluid by one-dimensional

9

#1-6

ACCEPTED MANUSCRIPT

heat conduction. One-dimensional heat conduction in the heat transfer plate is a reasonable approximation due to small heat loss and temperature distribution on heat transfer surface.

Twall = T2 −

RI PT

Temperature of the heat transfer surface Twall is calculated by extrapolation as ql2 k eff

(3)

SC

where l2 is distance between heat transfer surface and thermocouple T2. Local heat transfer coefficient h is obtained by: q Twall − Tsat

M AN U

h=

(4)

where Tsat is a saturated temperature based on an assumption that an inlet of the evaporator is a

pressure.

TE D

saturated pressure. P-PROPATH (PROPATH group, 2002) is used for calculation of saturation

Local thermal equilibrium quality x is obtained by: i − i sat ,l i fg

(5)

EP

x=

AC C

where i is local specific enthalpy, ifg is latent heat of evaporation, and isat,l is specific enthalpy of saturated liquid.

Specific enthalpy ij at five measuring point shown in Fig. 2 is obtained by j

Qk k =1 m

i j = iin + ∑

(6)

where iin is specific enthalpy at inlet of the evaporator, Qk is local heat transfer rate based on heat flux and heat transfer area, and m is mass flow rate of the working fluid.

10

ACCEPTED MANUSCRIPT

The working fluid at inlet of the evaporator is saturated condition. This means that specific enthalpy at inlet of the evaporator cannot be calculated directly. Then the specific

iin = i pre,in +

RI PT

enthalpy at inlet of the evaporator iin is obtained by Q pre

(7)

m

SC

where ipre,in is specific enthalpy based on inlet temperature and pressure of the pre-heater, Qpre is

M AN U

heat transfer rate of the pre-heater.

Experimental range of mass flux, heat flux, channel height, and saturation pressure are listed in Table 1. 2.3 Experimental procedure

TE D

The experiments are carried out by the following procedure: (1) Prepare hot and cold water for per-heater, condenser, and sub-cooler. Circulate cold water

EP

before heating working fluid to ensure experimental safety.

AC C

(2) Circulate the working fluid.

(3) Adjust voltage of the electric heaters for the evaporator to heat working fluid. (4) Collect data after steady state at specific experimental condition is confirmed. The data are accumulated every five seconds for three minutes after steady state is achieved. The data during this time period are averaged and used for data reduction. In this study, the experimental situation is regarded as steady state when fluctuation of mass flux and

11

ACCEPTED MANUSCRIPT

pressure of the working fluid are within 0.2 kg m-2 s-1 and 2 kPa, respectively. The maximum error induced by this fluctuation of mass flux is 4% of operating condition and that of pressure

RI PT

is 0.3%. The vapor quality defined by Eq. (5) at inlet of the plate evaporator is controlled in accordance with Eq. (7) to obtain heat transfer characteristics for wide range of the quality. The

SC

experiments are repeated 3 times for a single experimental condition. All repeated data are

M AN U

presented in results and discussion. 2.4 Uncertainty analysis

Uncertainties based on experimental condition of current study are listed in Table 3. Temperature, pressure, and mass flow rate are measured to obtain heat flux, wall temperature on

TE D

heat transfer surface, heat transfer coefficient, specific enthalpy, and quality in this study.

EP

Accuracies of measuring equipment are listed in Table 2.

AC C

3. Results and discussion

Heat transfer coefficient of microfin plate evaporator is measured. Effect of microfin and

its orientation on heat transfer coefficient is firstly discussed. Then effects of mass flux, heat flux, channel height, and saturation pressure are discussed. Finally the heat transfer coefficients are compared with that calculated by available empirical correlations. 3.1 Effect of microfin orientation

12

#2-3

ACCEPTED MANUSCRIPT

Figure 5 shows the local heat transfer coefficient as a function of the quality for plain, longitudinally- and laterally-finned heat transfer surfaces. Two fin orientations and direction of

RI PT

working fluid flow are schematically illustrated in Fig. 3. FC-72 flow direction 1 in Fig. 3 is assigned for longitudinally-finned heat transfer surface while FC-72 flow direction 2 is assigned #1-5

SC

for laterally-finned heat transfer surface. The experimental condition in Fig. 5 is G = 5 kg m-2 s-1,

M AN U

δ = 2 mm, q = 20 kW m-2 and Psat = 0.7 MPa.

The heat transfer coefficient rapidly increases with increasing quality for x < 0.4. Onset of flow boiling causes x > 0 and increases heat transfer coefficient significantly. Detouched bubbles from heat transfer surface are swept by liquid working fluid and the bubbles disrupt

TE D

thermal boundary layer. Decrease in liquid film thickness on heat transfer surface coincides with #1-7

vapor growth. Vapor quality and heat transfer coefficient is subsequently increased. Vapor

EP

bubbles experience coalescence and collision. The heat transfer coefficient gradually increases

AC C

with increasing quality for x > 0.4. Behavior of bubbles such as vapor collision and coalescence should dominate heat transfer process in this quality range. Insignificant discrepancy is observed on trend of the heat transfer coefficients among the heat transfer surfaces. This means that variation trend of heat transfer coefficient on quality is insensitive to microfin orientation.

#2-5

The discrepancy of heat transfer coefficient between longitudinally-finned and plain surfaces is within uncertainty. This means that the longitudinal microfin does not remarkably

13

#1-8

ACCEPTED MANUSCRIPT

contribute to heat transfer enhancement of the evaporator. In contrast the heat transfer coefficient of laterally-finned surface is larger than other two heat transfer surfaces. The

RI PT

mechanisms of heat transfer enhancement can be explained by increases in wetted surface area, #1-9

liquid-phase convective heat transfer, number of nucleation sites, and the effect of turbulence.

SC

Wetted surface area of the evaporator is increased by adding microfin on heat transfer surface. Wetted surface area of the plain evaporator is approximately 25000 mm2 while that of

M AN U

microfin evaporator is approximately 30000 mm2 regardless of microfin orientation. This extended wetted surface area intensifies boiling heat transfer. However, heat transfer enhancement in the longitudinally-finned evaporator is not clear compared with the plain

TE D

evaporator for x > 0.4 despite increment in surface area. This means that increase in wetted surface area by adding microfin does not effectively contribute to heat transfer process. This can

EP

be explained by situation of working fluid on the heat transfer surface.

AC C

Microfin orientations and direction of working fluid flow is illustrated in Fig. 3. Flow disturbance and swirl induced by fin geometry in laterally-finned evaporator are obviously larger than that in longitudinally-finned evaporator. This means that heat transfer for laterally-finned evaporator is enhanced by flow disturbance and swirl compared with longitudinally-finned one. Wetted surface area of two orientations is almost same; however, fin orientation

14

#1-10

ACCEPTED MANUSCRIPT

significantly affects behavior of liquid working fluid on the heat transfer surface because the evaporator of this study is vertically placed. This means that gravitational force is a significant

#1-10

RI PT

factor for fluid flow and heat transfer process. Liquid working fluid may not remain on heat transfer surface of the longitudinally-finned evaporator due to gravitational force. In contrast

#2-6

SC

lateral fin can retain liquid working fluid between microfins. Wetting region of laterally-finned

M AN U

evaporator is larger than that of longitudinally-finned one. This means that number of nucleation site of laterally-finned evaporator is larger than that of longitudinally-finned one and therefore heat transfer coefficient of laterally-finned evaporator is larger as well. The difference of heat transfer coefficient of two orientations at x > 0.4 can be explained by this mechanism. In this

TE D

quality region amount of liquid working fluid is small. The longitudinally-finned evaporator cannot vaporize the working fluid effectively whereas the laterally-finned one can vaporize

EP

small amount of working fluid effectively since the lateral fin can retain liquid as descried

AC C

above. This mechanism of heat transfer enhancement suggests that microfin geometry which retains liquid working fluid can enhance effectively heat transfer coefficient of vertically-placed evaporator. Visualization of flow boiling in the evaporator would be help to reveal behavior of bubbles and liquid working fluid. Unfortunately visualization is left as a future task because it #1-10 cannot be conducted by using the current experimental setup. Note that the laterally-finned evaporator is used in the following discussions since

15

ACCEPTED MANUSCRIPT

thermal performance of this evaporator is favorable. 3.2 Effect of mass flux

RI PT

Figure 6 depicts representative experimental result, which shows the local heat transfer coefficient as a function of quality for two mass fluxes G = 5 and 7.5 kg m-2 s-1. Experimental

SC

conditions of channel height, heat flux, and saturation pressure are 5 mm, 15 kW m-2, and 0.7 MPa, respectively.

M AN U

The heat transfer coefficient of microfin evaporator at the same experimental conditions is up to 13% larger than that obtained for a plain evaporator in the previous work Koyama et al. (2013). Note that data of heat transfer coefficient in the previous

#2-7

TE D

work is not presented in Fig. 6 for the sake of clarity. The heat transfer coefficient is

independent from the mass flux of ammonia for x < 0.3 whereas that is dependent from the mass

EP

flux for x > 0.3. Increment of mass flux slightly increases the heat transfer coefficient beyond

AC C

uncertainty. This can be explained by dominant factor on heat transfer process. For x < 0.3 onset of boiling and bubble nucleation rather than convection are significant on heat transfer process. The microfin on heat transfer surface increases numbers of nucleation site and intensifies bubble nucleation. In contrast, bubble behavior induced by convection of the working fluid dominates heat transfer characteristics for x > 0.3. Bubbles detached from heat transfer surface experience collision and coalescence along direction of working fluid flow. These chaotic behaviors result

16

#1-8

ACCEPTED MANUSCRIPT

in heat transfer enhancement due to destruction of thermal boundary layer. Yu et al. (2002) and Hatamipour and Akhavan-Behabadi (2010) presented that heat

RI PT

transfer coefficient of microfin tubes using R-134a increased with increasing mass flux of working fluid. Zhang (2011) showed that heat transfer enhancement on internally-grooved

SC

horizontal tube was achieved by increasing mass flux of R417 whereas heat transfer was

M AN U

deteriorated by increasing mass flux of R22. Momoki et al. (2009) showed that heat transfer coefficient of grooved tube using ammonia was increased with increasing mass flux from 40 to 60 kg m-2 s-1 whereas that was not affected by increasing mass flux from 60 to 80 kg m-2 s-1.

coefficient. 3.3 Effect of heat flux

TE D

More data including other refrigerants is necessary to clarify effect of mass flux on heat transfer

EP

Figure 7 shows the local heat transfer coefficient as a function of quality for different heat

AC C

fluxes 10, 15, and 20 kW m-2. Experimental conditions except the heat flux is G = 7.5 kg m-2 s-1, δ = 5 mm, Psat = 0.7 MPa.

The heat transfer coefficient of microfin evaporator at the same experimental conditions is

up to 12% larger than that obtained for a plain evaporator in the previous work Koyama et al. (2013). Note that value of heat transfer coefficient in the previous work is not presented in Fig. #2-7 7 to avoid complexity of the graph. One-step increase in heat flux slightly affects heat transfer

17

ACCEPTED MANUSCRIPT

coefficient and the heat transfer coefficient clearly increases with increasing the heat flux from 10 to 20 kW m-2. Increment of heat transfer coefficient can be explained by number of

RI PT

nucleation sites, bubble frequency and flow disturbance. Higher heat flux increases number of nucleation sites and bubble frequency on heat transfer surface. Strong flow disturbance is

SC

accompanied by these phenomena in the channel of evaporator. Consequently the heat transfer

M AN U

coefficient increases as shown in Fig. 7.

Dependency of heat flux on heat transfer coefficient is complicated in former works. Padovan et al. (2011) showed that heat transfer coefficient of microfin tube using R134a and R410A increased with increasing heat flux due to effect of nucleate boiling. Momoki et al.

TE D

(2009) reported insignificant effect of heat flux on heat transfer coefficient. Seo and Kim (2000) presented the result that heat transfer coefficient was slightly decreased by increasing heat flux

EP

for R22 evaporation in microfin tubes. Increase in nucleation site is expected by increasing heat

AC C

flux; however response to heat transfer enhancement is complicated. The dependency of heat transfer coefficient on heat flux at high vapor quality in current study indicates that convective boiling is not achieved in the current experimental condition. This dependency of heat transfer #2-8

coefficient can be observed in experiments with small mass flux. This can explain discrepancies of effect of heat flux found in different studies. 3.4 Effect of channel height

18

ACCEPTED MANUSCRIPT

Figure 8 shows the local heat transfer coefficient as a function of quality for different channel heights 1, 2, and 5 mm. Experimental conditions except the channel height are G = 5 kg

RI PT

m-2 s-1, q = 10 kW m-2, and Psat = 0.7 MPa.

The effect of channel height on heat transfer coefficient is classified into two cases in this

SC

study. The heat transfer coefficient is not influenced by the channel height reducing from 5 mm

M AN U

to 2 mm. In contrast the heat transfer coefficient dramatically increases with reducing the channel height from 2 mm to 1 mm. The increment in this case is approximately 40%. This can be explained by thin liquid film on heat transfer surface. Thickness of liquid film which contributes to heat transfer enhancement on heat transfer surface is thinned because bubbles in

TE D

passage of evaporator are elongated and confined. Heat transfer coefficients for δ = 1 are scattered. This can be explained by bubble behavior in narrow channel. Microfin in narrow #2-13

EP

channel δ = 1 occupies large volume compared with deep channel δ = 2 and 5 mm. The microfin

AC C

produces flow disturbance and chaotic bubble behavior, which influences response of heat transfer coefficient.

The effect of flow confinement on heat transfer coefficient has been discussed in several

literatures. Saturated flow boiling in compact heat exchangers was reviewed by Watel (2003). #1-11 The effect of gap size of flow passage on heat transfer coefficient was compiled associated with two-phase flow regions of isolated bubble, confined bubble, and dryout. Heat transfer

19

ACCEPTED MANUSCRIPT

coefficient increases with decreasing gap size in confined bubble region with low heat flux. This variation trend corresponds to the experimental results of current study. Transition criteria of

RI PT

flow confinement has been discussed by using non-dimensional numbers. Barber et al. (2010) used confinement number for FC-72 flow boiling in a rectangular microchannel. Confinement number Cof is defined by

SC

σ [g (ρ l − ρ g )] Dh

(8)

M AN U

Co f =

where σ is surface tension, g is gravitational acceleration, ρ is density, and Dh is hydraulic diameter. Transition criterion of flow confinement is indicated as Cof ≥ 0.5. Confinement number of current study is listed in Table 4, which shows flow of δ = 1 mm is confined and that

TE D

of δ = 2 mm is on the criterion. Note that confinement number defined by Eq. (8) is calculated by fluid properties and channel dimensions while mass flux is not included. Harirchian and

EP

Garimella (2010) used bond and Reynolds numbers for transition criterion of confined flow. The transition criterion is expressed as 1  ρl − ρ g g µ l  σ

AC C

Bo 0.5 Re =

 GDh2 < 160  

(9)

Where µ is viscosity. The results obtained by Eq. (9) is presented in Table 5, which explains flow of δ = 1 mm is confined. Few investigations of flow boiling in rectangular passage with microfin surface have been carried out so far. Effect of microfin tube diameter on heat transfer coefficient was reported by

20

ACCEPTED MANUSCRIPT

Seo and Kim (2000), who found that heat transfer coefficient of large diameter microfin tube was higher than that of small diameter tube.

RI PT

3.5 Effect of saturation pressure

Figure 9 shows the local heat transfer coefficient as a function of quality for different

M AN U

= 5 mm, G = 7.5 kg m-2 s-1, and q = 10 kW m-2.

SC

saturation pressures 0.7 and 0.9 MPa. Experimental conditions except saturation pressure are δ

The heat transfer coefficient is independent on saturation pressure for x < 0.3 whereas the effect of saturation pressure can be seen for x > 0.3. The heat transfer coefficient tends to increase with saturation pressure.

TE D

This can be explained by number of nucleation sites on heat transfer surface of the evaporator. The earlier work Koyama et al. (2013) which used plain plate evaporator showed

EP

that the effect of saturation pressure on heat transfer coefficient was insignificant. In contrast, Maqbool et al. (2012) showed that heat transfer coefficient increased with saturation pressure,

AC C

#2-9

because nucleation was increased with saturation pressure. Note that surface roughness in flow channel of Maqbool et al. (2012) is larger than that of Koyama et al. (2013). This means that number of nucleation sites in Maqbool et al. (2012) is essentially larger. In the current study, number of nucleation sites is increased by adding microfin on heat transfer surface. This leads increase in heat transfer coefficient by increasing saturation pressure.

21

ACCEPTED MANUSCRIPT

3.6 Empirical correlation Heat transfer coefficient obtained in this study is compared with that estimated by

RI PT

empirical correlations proposed by other researchers. Khan et al. (2009) have compiled empirical correlations to predict heat transfer coefficient of plate heat exchanger using

SC

refrigerants including ammonia. Some possible correlations are used to predict heat transfer

M AN U

coefficient of this study. 3.6.1 Ayub (2003) correlation

Ayub (2003) has derived a correlation of two-phase heat transfer coefficient for plate evaporators expressed as  Re l2 i fg   L 

   

0.4124

 p    p   cr 

0.12

 65    β 

0.35

TE D

 k hTP = C  l  Dh

(10)

EP

Note that the variables in Eq. (10) are in U.S. units in Ayub (2003). Figure 10 (a) shows the relation between heat transfer coefficients obtained by experiment

AC C

and Eq. (10) for different channel heights. Ayub (2003) correlation gives a single heat transfer coefficient for a certain mass flux. Experimentally-measured heat transfer coefficient in present study is a function of quality as presented in previous section whereas Ayub (2003) correlation is independent from quality. Equation (10) is not represent trend of experimental result of present study even though most of the experimental data for δ = 2 mm are within ±30%. 3.6.2 Kandlikar and Balasubramanian (2004) correlation 22

ACCEPTED MANUSCRIPT

Kandlikar and Balasubramanian (2004) have proposed flow boiling correlation for minichannels and microchannels. Two-phase heat transfer coefficient is calculated by

hTP ,CBD = 1.136Cov

−0.2

−0.9

(1 − x )0.8 hliq + 1058.0 Bo 0.7 (1 − x )0.8 FFl hliq

(11)

RI PT

hTP , NBD = 0.6683Cov

(1 − x )0.8 hliq + 667.2Bo 0.7 (1 − x )0.8 FFl hliq

(12)

SC

The lager of hTP,NBD and hTP,CBD is designated as two-phase heat transfer coefficient hTP. Convection number Cov and boiling number Bo is defined as

Bo =

0. 5

 1− x       x 

0. 8

M AN U

 ρg Cov =   ρl

q Gi fg

(13) (14)

Fluid-surface parameter FFl is 0.7, which is recommended by Zamfirescu (2002). Single-phase

hliq =

Nukl Dh

TE D

all-liquid laminar heat transfer coefficient hliq is obtained by (15)

EP

According to Shah and London (1978) with assumption of fully developed flow, Nusselt

AC C

number in Eq. (15) is 5.385 for the evaporator of present study. Figure 10 (b) shows the relation between heat transfer coefficients obtained by

experiment and Eq. (11-15) for different channel heights. For δ = 2mm the correlation of Kandlikar and Balasubramanian (2004) can predict heat transfer coefficient within 30% error. However, the correlation cannot be employed for δ = 1 and 5 mm. 3.6.3 Mortada et al. (2012) correlation

23

ACCEPTED MANUSCRIPT

Mortada et al. (2012) have proposed a correlation of Nusselt number for flow boiling in minichannels as

(

)

b

d Nu = a ⋅ Bo 2Wel ⋅ Coc ⋅ Reliq

RI PT

(16)

Weber number We, Reynolds number Re, and constant Coc are defined by G 2 Dh

SC

Reliq =

(17)

ρ lσ

GDh

µl

1− x  Coc =    x 

0.7

 ρg   ρl

  

0. 5

M AN U

Wel =

(18) (19)

Constants in Eq. (16) are as follows: a = 8678.5, b = 0.2415, c = -0.75, d = -0.115. Figure 10 (c) shows the relation between heat transfer coefficients obtained by experiment

TE D

and Eq. (16-19) for different channel heights. Most of predicted results are out of 30% range. 3.6.4 Djordjevic and Kabelac (2008) correlation

EP

Djordjevic and Kabelac (2008) have proposed a correlation of heat transfer coefficient

AC C

based on that of Danilova (1981). Djordjevic and Kabelac (2008) recommended that Danilova (1981) correlation was scaled up by a factor of 1.15. Danilova correlation is expressed as h = 4.2 Re0g.3 Bo0.33 Re0S.2

kl Dh

(20)

where Re g = Bo =

xGDh

(21)

ρ gν g

gρl Dh2

(22)

σ

24

ACCEPTED MANUSCRIPT

Re S =

qDh i fg µl

(23)

Figure 10 (d) shows the relation between heat transfer coefficients obtained by

RI PT

experiment and Eq. (20-23) for different channel heights. The predicted heat transfer coefficient for δ = 2 and 5 mm is close or within 30% range; however, applicability of the prediction is

SC

limited.

M AN U

3.6.5 Modified correlation

Authors proposed empirical correlations using Lockhart-Martinelli parameter for a plate evaporator which has a plane channel in Koyama et al. (2013). The current study attempts to develop empirical correlation by using same manner.

TE D

Reynolds number in the current experimental range is up to 443. This means that flow of the working fluid is laminar. Lockhart-Martinelli parameter of laminar flow Χvv is defined as 0.5

 ρg  ρ  l

   

0. 5

 µl   µg 

   

0.5

EP

Χ vv

1− x  =   x 

(24)

AC C

Heat transfer coefficient of the evaporator is non-dimensionalized by using liquid-phase heat transfer coefficient as  1 h = A hliq  Χ vv

  

n

(25)

Where A is constant and hliq is calculated by Dittus-Boelter equation as

25

ACCEPTED MANUSCRIPT

hliq

  G (1 − x )Dh     µl  

 k = 0.023 l  Dh

0. 8

Prl0.4

(26)

RI PT

Empirical correlations are separated with respect to channel height because the effect of channel height on heat transfer coefficient is significant as shown in Fig. 8. Figure 11 (a) shows the relation defined by Eq. (25) for δ = 1 mm and Fig. 11 (b) shows that for δ = 2 and 5 mm.

SC

The correlation for δ = 2 and 5 mm is classified into two equations because the slope of the

M AN U

experimental result is changed at Χvv = 1 For δ = 1 mm:  1 h = 48.0 hliq  Χ vv

   

0.95

 1 h = 41.8 hliq  Χ vv

   

 1 h = 47.1 hliq  Χ vv

   

0.96

(1 / Χvv ≥ 1)

(28)

(1 / Χvv ≤ 1)

(29)

EP

0.51

TE D

For δ = 2 and 5 mm:

(27)

The broken line in Fig. 11 shows the range of ± 30% of the empirical correlations.

AC C

Equation (27) includes 92% and Eqs. (28-29) includes 87% of the experimental data within ± 30% range.

4. Conclusion This study investigates thermal characteristics of ammonia flow boiling in a microfin plate evaporator. Heat transfer coefficient is measured and compared with available empirical 26

ACCEPTED MANUSCRIPT

correlations. Modified empirical correlations based on experimental results are developed. The conclusions of this study are summarized as:

RI PT

(1) Microfin structure enhances heat transfer of the evaporator compared with plain heat transfer surface. Laterally-finned orientation is superior to longitudinally-finned one with respect to heat

M AN U

nucleation accompanied by increase in wetting region.

SC

transfer enhancement, which is induced by the effect of flow disturbance and increase in

(2) Heat transfer coefficient increases with increasing mass flux, heat flux, and saturation pressure, and with decreasing channel height of the evaporator. Additional experiments including investigations for various working fluids and channel geometry are necessary to

TE D

clarify thermal performance of microfin plate evaporator. (3) Empirical correlations using Lockhart-Martinelli parameter to predict heat transfer

EP

coefficient of the laterally-microfinned plate evaporator are developed. The correlations can

AC C

predict 92% of the experimental data for channel height δ = 1 mm 87% of the experimental data for δ = 2 and 5 mm within ± 30% range.

27

ACCEPTED MANUSCRIPT

Nomenclature constant

Bo

boiling number

Cof

confinement number

Cov

convection number

Dh

hydraulic diameter, m

FFl

fluid-surface parameter

g

gravitational acceleration, m s-2

G

mass flux, kg m-2 s-1

h

heat transfer coefficient, W m-2 K-1

i

specific enthalpy, J kg-1

ifg

latent heat of evaporation, J kg-1

k

thermal conductivity, W m-1 K-1

m

SC M AN U

TE D

EP

AC C

l

RI PT

A

distance between thermocouples, m

mass flow rate, kg s-1

n

exponent

Nu

Nusselt number

P

pressure, Pa

28

ACCEPTED MANUSCRIPT

critical pressure, Pa

Pr

Prandtl number

Qk

local heat transfer rate based on heat flux and heat transfer area, W

q

heat flux, W m-2

Re

Reynolds number

t1

thickness of titanium plate, m

t2

thickness of brazing portion, m

t3

thickness of titanium base plate, m

T

temperature, °C

We

Weber number

x

thermal equilibrium quality

β

chevron angle, degree

δ

channel height, m

ρ

EP

TE D

M AN U

SC

#2-2

AC C

µ

RI PT

Pcr

viscosity, Pa s

density, kg m-3

σ

surface tension, N m-1

Χvv

Lockhart-Martinelli parameter (laminar-laminar flow)

Subscript

29

eff

effective value

g

gas phase

l

liquid phase

liq

only liquid phase flows

NBD

nucleate boiling dominant

s

brazing portion

sat

saturated state

TP

two-phase

wall

wall of the plate evaporator

SC

convection boiling dominant

AC C

EP

TE D

M AN U

CBD

RI PT

ACCEPTED MANUSCRIPT

30

ACCEPTED MANUSCRIPT

Acknowledgements The authors appreciate KOBE STEEL Ltd. This study has been conducted by joint research with

AC C

EP

TE D

M AN U

SC

RI PT

KOBE STEEL Ltd.

31

ACCEPTED MANUSCRIPT

References Ayub, Z.H., 2003, Plate heat exchanger literature survey and new heat transfer and pressure

RI PT

drop correlations for refrigerant evaporators, Heat Transfer Engineering 24, 3-16.

Bandarra Filho, E.P., Saiz Jabardo, J.M., Barbieri, P.E.L., 2004, Convective boiling pressure

SC

drop of refrigerant R-134a in horizontal smooth and microfin tubes, Int. J. Refrigeration 27,

M AN U

895-903.

Chamra, L.M., Webb, R.L., Randlett, M.R., 1996, Advanced micro-fin tubes for evaporation, Int. J. Heat and Mass Transf 39, 1827-1838.

Danilova, D.N., 1981, Heat transfer in different plate geometries, Kholod. Tek. 4, 25–31.

TE D

Del Col, D., Webb, R.L., Narayanamurthy, R., 2002, Heat transfer mechanisms for condensation and vaporization inside a microfin tube, Journal of Enhanced Heat Transfer 9, 25-37.

EP

Djordjevic E., Kabelac, S., 2008, Flow boiling of R134a and ammonia in a plate heat exchanger,

AC C

Int. J. Heat Mass Transf 51, 6235-6242. Fujii, T., Koyama, S., Inoue, N., Kuwahara, K., Hirakuni, S., 1995, An Experimental study of evaporation heat transfer of refrigerant HCFC22 inside an internally grooved horizontal tube, JSME International Journal, Series B: Fluids and Thermal Engineering 38, 618-627. Harirchian, T., Garimella, S.V., 2010, A comprehensive flow regime map for microchannel flow boiling with quantitative transition criteria, Int. J. Heat Mass Transf. 53, 2694-2702.

32

ACCEPTED MANUSCRIPT

Kandlikar, S.G., Balasubramanian, P., 2004, An Extension of the Flow Boiling Correlation to Transition, Laminar, and Deep Laminar Flows and Microchannels, Heat Transfer Engineering 25, 86-93.

RI PT

#2-12

Kelly, J.E., Eckels, S.J., Fenton, D.L., 2002a, An experimental investigation of in-tube

SC

evaporation of pure ammonia in a smooth and a microfin tube, Part I - Heat transfer (RP-866),

M AN U

HVAC and R Research 8, 239-256.

Kelly, J.E., Eckels, S.J., Fenton, D.L., 2002b, An experimental investigation of in-tube evaporation of pure ammonia in a smooth and a microfin tube, Part II - Pressure drop (RP-866), HVAC and R Research 8, 257-276.

TE D

Khan, T.S., Khan, M.S., Chyu, M.-C., Ayub, Z.H., Chattha, J.A., 2009, Review of heat transfer and pressure drop correlations for evaporation of fluid flow in plate heat exchangers (RP-1352),

EP

HVAC and R Research 15, 169-188.

AC C

Kim, M.H., Shin, J.S., 2005, Evaporating heat transfer of R22 and R410A in horizontal smooth and microfin tubes, Int. J. Refrigeration 28, 940-948. Koyama, K., Chiyoda, H., Arima, H., Ikegami, Y., 2013, Experimental study on thermal characteristics of ammonia flow boiling in a plate evaporator at low mass flux, Int. J. Refrigeration, in press. Maqbool, M.H., Palm, B., Khodabandeh, R., 2012, Boiling heat transfer of ammonia in vertical

33

ACCEPTED MANUSCRIPT

smooth mini channels: Experimental results and predictions, Int. J. Therm. Sci. 54, 13-21. Momoki, S., Arima, H., Yamaguchi, T., Shigechi, T., 2009, Experimental study on evaporation

RI PT

heat transfer of ammonia flowing inside a horizontal internally spirally grooved tube, Transaction of the JSRAE 26, 305-314 (in Japanese).

SC

Mortada, S., Zoughaib, A., Arzano-Daurelle, C., Clodic, D., 2012, Boiling heat transfer and

M AN U

pressure drop of R-134a and R-1234yf in minichannels for low mass fluxes, Int. J. Refrigeration 35, 962-973.

Padovan, A., Del Col, D., Rossetto, L., 2011, Experimental study on flow boiling of R134a and R410A in a horizontal microfin tube at high saturation temperatures, Applied Thermal

TE D

Engineering 31, 3814-3826.

PROPATH Group, http://www2.mech.nagasaki-u.ac.jp/PROPATH/p-propath.html

EP

Seo, K., Kim, Y., 2000, Evaporation heat transfer and pressure drop of R-22 in 7 and 9.52 mm

AC C

smooth/micro-fin tubes, Int. J. Heat Mass Transf 43, 2869-2882. Shah, R.K., London, A.L., 1978, Laminar flow forced convection in ducts: a source book for compact heat exchanger analytical data, Academic Press. Thom, J.R., Cheng, L., Ribatski, G., Vales, L.F., 2008, Flow boiling of ammonia and hydrocarbons: A state-of-the-art review, Int. J. Refrigeration 31, 603-620. Thome, J.R., 1996, Boiling of new refrigerants: A state-of-the-art review, Int. J. Refrigeration 19,

34

ACCEPTED MANUSCRIPT

435-457. Thome, J.R., Cheng, L., Ribatski, G., Vales, L.F., 2008, Flow boiling of ammonia and

RI PT

hydrocarbons: A state-of-the-art review, Int. J. Refrigeration 31, 603-620.

Uehara, H. Nakaoka, T., 1985, OTEC Plant Consisting of Plate type Heat Exchangers, Heat

SC

Transfer-Japanese Research 14, 19-35.

Int. J. Therm. Sci. 42, 107-140.

M AN U

Watel, B., 2003, Review of saturated flow boiling in small passages of compact heat-exchangers,

Yu, M.-H., Lin, T.-K., Tseng, C.-C., 2002, Heat transfer and flow pattern during two-phase flow boiling of R-134a in horizontal smooth and microfin tubes, Int. J. Refrigeration 25, 789-798.

TE D

Zamfirescu, C., Chiriac, F., 2002. Heat transfer measurements on ammonia forced convection boiling in vertical tubes. Exp. Therm. Fluid Sci 25 (7), 529–534.

EP

Zhang, X., 2011, Heat transfer and enhancement analyses of flow boiling for R417A and R22,

AC C

Exp. Therm. Fluid Sci 35, 1334-1342. Zhiping, L., Tongze, M., Zhengfang, Z., Lichun, Z., 2001, Enhanced boiling heat transfer from different surfaces in narrow space, Journal of Enhanced Heat Transfer 8, 373-381.

35

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 1 Schematic diagram of experimental setup.

36

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 2 Configuration of the plate evaporator.

37

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 3 Orientation of microfin heat transfer surface and FC-72 flow direction.

38

#1-5

AC C

EP

TE D

M AN U

Figure 4 Configuration of a pair of thermocouples.

SC

RI PT

ACCEPTED MANUSCRIPT

39

#2-1

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 5 Effect of microfin orientation on heat transfer coefficient. Experimental conditions: G

AC C

EP

TE D

= 5 kg m-2 s-1, δ = 2 mm, q = 20 kW m-2, Psat = 0.7 MPa.

40

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 6 Effect of mass flux on heat transfer coefficient. Experimental conditions: δ = 5 mm, q

AC C

EP

TE D

= 15 kW m-2, Psat = 0.7 MPa.

41

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Figure 7 Effect of heat flux on heat transfer coefficient. (a) δ = 5 mm, G = 7.5 kg m-2 s-1, Psat =

AC C

EP

TE D

0.7 MPa.

42

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 8 Effect of channel height on heat transfer coefficient. Experimental conditions: G = 5 kg

AC C

EP

TE D

m-2 s-1, q = 10 kW m-2, Psat = 0.7 MPa.

43

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 9 Effect of saturation pressure on heat transfer coefficient. Experimental conditions: δ =

AC C

EP

TE D

5 mm, G = 7.5 kg m-2 s-1, q = 10 kW m-2.

44

SC

RI PT

ACCEPTED MANUSCRIPT

(b)

TE D

M AN U

(a)

(d)

EP

(c)

AC C

Figure 10 Prediction of heat transfer coefficient by using (a) Ayub (2003) (b) Kandlikar and Balasubramanian (2004) (c) Mortada et al. (2012) (d) Djordjevic and Kabelac (2008) at δ = 1 to 5 mm, G = 5 kg m-2 s-1, q = 20 kW m-2, and Psat = 0.7 MPa.

45

EP

TE D

M AN U

(a)

SC

RI PT

ACCEPTED MANUSCRIPT

(b)

AC C

Figure 11 New empirical correlations for heat transfer coefficient using Lockhart-Martinelli parameter. (a) δ = 1 mm. (b) δ = 2 and 5 mm.

46

ACCEPTED MANUSCRIPT

Figure captions

Figure 2 Configuration of the plate evaporator.

RI PT

Figure 1 Schematic diagram of experimental setup.

Figure 4 Configuration of a pair of thermocouples.

SC

Figure 3 Orientation of microfin heat transfer surface and FC-72 flow direction.

Figure 5 Effect of microfin orientation on heat transfer coefficient. Experimental conditions: G

M AN U

= 5 kg m-2 s-1, δ = 2 mm, q = 20 kW m-2, Psat = 0.7 MPa.

Figure 6 Effect of mass flux on heat transfer coefficient. Experimental conditions: δ = 5 mm, q = 15 kW m-2, Psat = 0.7 MPa.

0.7 MPa.

TE D

Figure 7 Effect of heat flux on heat transfer coefficient. (a) δ = 5 mm, G = 7.5 kg m-2 s-1, Psat =

EP

Figure 8 Effect of channel height on heat transfer coefficient. Experimental conditions: G = 5 kg

AC C

m-2 s-1, q = 10 kW m-2, Psat = 0.7 MPa. Figure 9 Effect of saturation pressure on heat transfer coefficient. Experimental conditions: δ = 5 mm, G = 7.5 kg m-2 s-1, q = 10 kW m-2. Figure 10 Prediction of heat transfer coefficient by using (a) Ayub (2003) (b) Kandlikar and Balasubramanian (2004) (c) Mortada et al. (2012) (d) Djordjevic and Kabelac (2008) at δ = 1 to 5 mm, G = 5 kg m-2 s-1, q = 20 kW m-2, and Psat = 0.7 MPa.

47

ACCEPTED MANUSCRIPT

Figure 11 New empirical correlations for heat transfer coefficient using Lockhart-Martinelli

AC C

EP

TE D

M AN U

SC

RI PT

parameter. (a) δ = 1 mm. (b) δ = 2 and 5 mm.

48

ACCEPTED MANUSCRIPT

Table 1 Experimental range of this study. Range

Mass flux G (kg m-2s-1)

5, 7.5

Heat flux q (kW m-2)

10, 15, 20

Channel height δ (mm)

1, 2, 5

Saturation pressure Psat (MPa)

0.7, 0.9

AC C

EP

TE D

M AN U

SC

RI PT

Parameter

49

ACCEPTED MANUSCRIPT

Manufacture

Model

Coriolis flowmeter

Endress+Hauser

PROMASS 83A

Magnetic flowmeter

Keyence

FD-M, FD-P

K-type sheathed thermocouple

Hayashi Denko

Pressure transducer

Yokogawa Electric

SR6 Class1

FP101 0-2MPa

M AN U TE D EP AC C 50

Accuracy

RI PT

Equipment

SC

Table 2 Measuring equipment.

±0.1% of reading

±1.6% full scale

±1.5°C of reading

±0.25% full scale

ACCEPTED MANUSCRIPT

Table 3 Results of uncertainty analysis. Maximum

Parameter

±0.5%

Wall tempetature Twall

±0.7%

Heat transfer coefficient h

±9.8%

Specific enthalpy i

±3.5%

SC

Heat flux q

RI PT

uncertainty

±9.6%

AC C

EP

TE D

M AN U

quality x

51

ACCEPTED MANUSCRIPT

δ mm

Cof

1

0.993

2

0.497

5

0.199

RI PT

Table 4 Transition criterion of confined flow using confinement number.

AC C

EP

TE D

M AN U

SC

#1-11

52

ACCEPTED MANUSCRIPT

Table 5 Transition criterion of confined flow using bond and Reynolds numbers. Bo0.5Re for G = 5 kg m-2 s-1

Bo0.5Re for G = 7.5 kg m-2 s-1

1

62

93

2

247

371

5

1544

RI PT

δ mm

2316

AC C

EP

TE D

M AN U

SC

#1-11

53

ACCEPTED MANUSCRIPT

Highlights

RI PT

- Thermal characteristics of microfin plate evaporators are investigated.

- Longitudinally- and laterally-finned heat transfer surfaces are used and compared.

SC

- Mechanisms of heat transfer enhancement by using microfin are discussed.

M AN U

- Effects of mass flux, heat flux, channel height, and saturation pressure are discussed.

AC C

EP

TE D

- Modified empirical correlations to predict heat transfer coefficient are developed.