- Email: [email protected]
Condensation heat transfer characteristics of CO2 in a horizontal smooth- and microfin-tube at high saturation temperatures
Applied Thermal Engineering 36 (2012) 51e62
Contents lists available at SciVerse ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Condensation heat transfer characteristics of CO2 in a horizontal smooth- and microfin-tube at high saturation temperatures Chang-Hyo Son*, Hoo-Kyu Oh Department of Refrigeration and Air-Conditioning Engineering, College of Engineering, Pukyong National University, San 100, Yongdang-dong, Nam-gu, Pusan 608-739, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 May 2011 Accepted 9 December 2011 Available online 16 December 2011
The condensation heat transfer characteristics for CO2 at high saturation temperatures in a horizontal smooth- and microfin-tube were investigated by experiment with respect to condensation temperature and mass flux. The test sections consist of 2400 mm length with a horizontal copper tube of 4.6 mm (smooth) and 4.95 mm (microfin) inner diameter. The experiments were conducted at refrigerant mass flux of 400e800 kg/(m2s), and saturation temperature of 20e30 C. The main experimental results showed that annular flow almost dominated the major of condensation flow in the horizontal smoothand microfin-tube. The condensation heat transfer coefficients for the smooth- and microfin-tube increase with the decreasing saturation temperature and increasing mass flux. The average heat transfer enhancement factor (EF) for the microfin tube is approximately from 1.006 to 1.48, and penalty factor (PF) varies from 1.14 to 1.23. The experimental data in the smooth- and microfin-tube were compared against previous heat transfer correlations. Most of correlations failed to predict the experimental data. However, the correlations by Kondou and Hrnjak [1] and Cavallini et al. [4] showed relatively good agreement with experimental data in the smooth- and microfin-tube, respectively. But it is necessary to develop accurate and reliable correlation to predict condensation heat transfer coefficient at high saturation temperatures in the horizontal smooth- and microfin- tube. Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Keywords: Condensation heat transfer coefficient Carbon dioxide (CO2) Microfin tube Smooth tube Subcritical cycle High saturation temperature
1. Introduction During summer season, commercial refrigeration systems using CO2 spend most operating hours in a transcritical cycle, while these systems at autumn and spring season are operated as a subcritical cycle. [1] The operation conditions of the subcritical cycle are quite different from those of the transcritical one. In the summer season, the heat rejection temperatures of the transcritical system of CO2 are usually above the critical temperature of CO2 (31.1 C), the systems using CO2 will have to operate in the transcritical cycle. As such, the heat rejection takes place above the critical pressure in a so-called gas cooler (corresponding to the condenser in the subcritical systems), while the heat rejection process at autumn and spring season remains below the subcritical region. Consequently, the CO2 heat transfer characteristics of the heat rejection process in the subcritical cycle are considerably different from those in the transcritical cycle. Thus, it is necessary to understand the characteristics of the subcritical cycle to ensure
* Corresponding author. Tel.: þ82 51 629 6802; fax: þ82 51 611 6368. E-mail addresses: [email protected], [email protected] (C.-H. Son).
reliable system performance and to evaluate the annual performance. Heat transfer enhancement has been a substantial factor to obtain high energy efficiency in refrigeration and air-conditioning applications. Microfin tubes represent a technology that has been able to beneficially enhance condensation heat transfer without causing similar increases in pressure drop and refrigerant charge, both in single-phase and two-phase application. Condensation heat transfer in microfin tubes is increased of (i) the larger surface area (A), (ii) the thinning of the condensate film (d) by a redistribution of the liquid due to spiraling and surface tension forces, and (iii) the presence of film disturbances caused by the presence of the fins, all of which results in an increase of the thermal capacity performance of the heat exchanger [2,3]. For conventional refrigerants, flow condensation in smooth and microfin tubes was investigated by several researchers [4e14]. But, as shown in Table 1, the characteristics of CO2 flow condensation heat transfer in horizontal smooth and microfin tubes are not well known. However, in recent years, some researchers [15e18] have studied for CO2 condensation heat transfer inside smooth and microfin tubes. As shown in Table 1, most of the previous studies about flow condensation in tubes were performed at low temperatures below
1359-4311/$ e see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.applthermaleng.2011.12.017
52
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Nomenclature A cp d e EF dp/dz G g ifg j kw m N n P DP PF Pr Q Re T VCR X x dz
heat transfer area, m2 specific heat at constant pressure, kJ/(kgK) diameter, m fin height, mm enhancement factor frictional pressure gradient along tube axis, kPa/m mass flux, kg/(m2s) gravitational acceleration, m/s2 latent heat, kJ/kg superficial velocity, m/s thermal conductivity, kW/(mK) mass flow rate, kg/s number of data number of microfins pressure, kPa pressure drop, kPa penalty factor Prandtl number, mcp/k heat capacity, kW Reynold number, rvdi/m temperature, K volumetric refrigeration capacity, kJ/m3 Martinelli parameter vapor quality length of test section, m
0 C. Also, a few of CO2 condensation heat transfer coefficient data in smooth and microfin tubes have been reported in any publication [15,16] known to the authors. More experimental studies are necessary to develop CO2 condensation heat transfer database and correlations in smooth and microfin tubes. Accordingly, the purpose of this study is to present new experimental data and search the suitable existing predictions which describe the present data and also analyze the experimental data to find condensation heat transfer characteristics in the horizontal smooth- and microfin-tube. 2. Experimental apparatus and procedures 2.1. Test facility The experimental facility as schematically shown in Fig. 1 is designed to investigate the condensation heat transfer coefficient of CO2 in the smooth- and microfin-tube. Basically, the main loops of the system are a CO2 loop and a cooling water loop. Detailed descriptions of the two independent loops of the test facility are provided below. As shown in Fig. 1, the CO2 loop is composed of a magnetic gear pump, a mass flow meter, two preheaters, a condenser (test section for condensation heat transfer experiment), a subcooler and a receiver etc. The refrigerant then passes in series through the magnetic gear pump, the refrigerant flow meter, the preheater, and
Greek symbols b microfin apex angle, g microfin helix angle, m dynamic viscosity, Pa s r density, kg/m3 d film thickness, m s surface tension, mN/m, deviation n specific volume, m3/kg Subscripts abs mean avg average cal calculated cr critical point cs source water exp experimental i inner, inside in inlet l liquid out outlet pre preheater pred predicted re refrigerant sat saturation sub subsection v vapor w wall
enters the test section. The subcooled refrigerant is charged in the receiver where it is further cooled to increase its density. The refrigerant liquid in the receiver is circulated by a magnetic gear pump which can be regulated by means of an inverter, and then flows through a flow meter. The liquid of refrigerant enters the preheater, where it is heated up to the desired inlet quality and temperature before entering the test section. Electrically insulated heating wires are wrapped around the surface of copper tubes in the preheater. The preheater is insulated with glass fibers and rubber. The amount of heat loss from the preheater is calibrated through pretests with water; this is correlated to the voltage input. In passing through the test section, the two-phase refrigerant is completely condensed and subcooled by cooling water. Afterwards, the CO2 is subcooled by the subcooler and goes into the liquid receiver. Finally, the subcooled refrigerant is recirculated through the refrigerant loop. The subcooler is a counterflow heat exchanger with refrigerant flowing in the inner tube and the water flowing in the annulus. It is used to condense the refrigerant leaving the test section. The mass flow rate of refrigerant into the test section is adjusted by changing the speed of the gear pump. The refrigerant flow rate is measured by the mass flow meter. The pressure of the cycle is controlled by the charged amount of refrigerant. The cooling water loop is composed of a centrifugal pump, an inline electric heater and a heat exchanger. The cooling water is pumped to the circular-tube annulus, where it absorbs the heat of
Table 1 Summary of previous studies on condensation heat transfer of CO2 in smooth and microfin tubes. Literature
Tube geometry
Tsat [ C], Psat [MPa]
x [/], qre [kW/m2]
Gre [kg/(m2s)]
Kondou and Hrnjak [1] Jang and Hrnjak [16]
Smooth tube, di ¼ 6.1 mm Smooth tube, microfin tube di ¼ 6.1 mm Microfin tube di ¼ 5.67 mm
Psat ¼ 5e7 15, 25
0e1, qre ¼ 3e30 0.1e0.9
100e240 200e400
Psat ¼ 5, 6 MPa
0.1e1.0
e
Koyama et al. [15]
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
53
Fig. 1. Schematic diagram of experimental apparatus.
the condensing refrigerant. The mass flow rate of cooling water is controlled by adjusting the metering valve and pump speed. The inlet temperature of the cooling water at the test section is controlled by both the electric heater and the refrigeration unit. The mass flow rate of cooling water is also measured by a turbine type flow meter. 2.2. Test section The schematic diagram of the test sections is shown in Fig. 2. The test sections are horizontal double-tube type heat exchangers which are made by a smooth- and microfin-tube, respectively. The refrigerant flows through the inside of copper tube and the cooling water flows through the annulus in the counterflow direction. The test sections are the smooth- and microfin-tube having insidediameters/outside-diameters of 4.95 mm/6.35 mm and 4.6 mm/ 5.0 mm, respectively. The total length of two condensers is 2400 mm and contains 12 subsections along with 200 mm in length of one subsection. In order to measure accurately the temperature of cooling water in the test sections, a mixing chamber is installed at the outlet of each subsection and inlet and outlet of test section. Fig. 3 presents the characteristic geometries of the microfin tube used in this study. This microfin tube has a single set of 55 spiral fins with spiral angle (b) of 18 , fin height (e) of 0.2 mm, fin apex angle (g) of 40 . The test sections are well insulated by thermal insulation material with a thermal conductivity of 0.04 W/(mK). The outer tube is made of a clear copper tube with an internal diameter of 11.1 mm. In order to minimize heat gain and loss from the surroundings, fiberglass insulation material was used to cover the test section as shown in Fig. 4. At the central position of each subsection the outside wall temperature of the inner tube is measured with four copper-constantan thermocouples which are distributed circumferentially at the top, left, bottom, and right sides
of the tube as shown in Fig. 4 to take care of any circumferential temperature variation. In order to verify the test facility, preliminary tests were performed. To check the accuracy, the amount of heat loss was calculated from the energy balances between the inner copper tube side and the annular side. The heat loss was 3% of the total heat transfer rate. The results indicated that all sensors and the instruments worked satisfactorily. In addition, the single phase heat transfer coefficients were calculated and compared with Gnielinski’s [19] correlation. The comparison showed that the absolute average deviation was less than 10.9%. The uncertainty changed depending on the flow conditions, so their minimum to maximum ranges are shown in Table 2. The refrigerant temperatures are measured by T-type thermocouples at the inlet and outlet of each subsection. Temperature of the cooling water is measured at inlet and outlet of each subsection. The pressure at the inlet and outlet of each subsection is measured with a pressure transducer. The pressure drop of the test section is obtained from the difference of measured pressure at the inlet and outlet. Mass flow rate of the refrigerant is measured with the mass flow meter of Oval corporation, and mass flow rate of the cooling water is measured with the flow meter of Schlumberger corporation. 2.3. Experimental method and condition As stated above, the test rig can be used to conduct the condensation heat transfer and pressure drop test. The thermodynamic properties of the refrigerants under test were incorporated in the data-acquisition system consisting of a data logger, interface card (HPIB), and a compatible PC. The data was analyzed in real time using PC and a data reduction program (MS-Excel with a Visual Basic). All of the information about the test conditions and data were displayed on the monitor during the test, the test conditions were changed based
54
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Fig. 2. Schematic diagram of test sections. (a) Smooth tube, (b) Microfin tube.
on this information. All channels in the data-acquisition system were scanned five times. The data collection took about 2 min for each run. During the condensation heat transfer test, the preheater was used to adjust the refrigerant inlet quality. By knowing the inlet temperature of the refrigerant prior to the preheater and the power supplied to the preheater, the inlet quality of the refrigerant in the test section can be calculated. The exit quality of the refrigerant can be determined by applying an energy balance on the water side. The system was allowed to come to steady-state before any data were recorded. The steady-state condition of experiment was reached when the refrigerant inlet pressure/temperature/quality, exit temperature/ quality, mass flow rate and water mass flow rate and inlet and outlet temperature keep in stationary state. The test conditions in this study are summarized in Table 3. Test runs are conducted for various mass fluxes and inlet pressures of CO2. Mass fluxes are controlled from 400 to 800 kg/(m2s) by a variable speed magnetic gear pump. The saturation temperatures of CO2 are adjusted from 20 to 30 C.
2.4. Data reduction Defect signals for checking data are processed with the computer through the data logger. In this study, the thermodynamic and transport properties of all refrigerants are calculated using REFPROP (version 8.0) [20] made by NIST(National Institute of Standards and Technology). We can use the following equations to analyze the test data, using the above mentioned properties. The heat flux, qc, supplied to inside wall surface of the copper tube is calculated as Eq. (1):
qc ¼
Qc
p$di $dz
(1)
where, dz indicates the effective length of a subsection, di is inner diameter of the smooth tube and fin tip diameter of the microfin tube, respectively. Qc is the heat transfer rate gained by cooling water flowing through the annular. The circumferential local heat
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
55
Fig. 3. Characteristics geometries of microfin tube.
Fig. 4. Setting of temperature sensor and insulation material.
transfer coefficient of refrigerant is calculated by the following Eq. (2):
With the enthalpies calculated, subsection inlet vapor quality was given by Eq. (8).
hc;loc ¼
qc Tre Tre;w;in
(2)
where, hc,loc represents the local heat transfer coefficient in the subsection of condenser, Tre is bulk refrigerant temperature in condenser. Tre,w,in is determined from average of the measured outside wall temperature via one the equation of radial thermal conduction through the wall as shown in Eq. (3):
Tre;w;in ¼ Tre;w;out Qre;sub
lnðdo =di Þ ð2p$dz$kw Þ
(3)
Tw;top þ Tw;left þ Tw;right þ Tw;bottom 4
(4)
where Tw,top, Tw,left, Tw,right, and Tw,bottom are the measured temperature at the top, both the left and right sides and bottom wall, respectively. The subcooled refrigerant enthalpy upstream of preheater was determined by measuring its temperature and pressure. Then, the subsection inlet enthalpy could be calculated from the heat input of the preheater.
isub;in ¼ ipre þ Qpre =mr
(5)
(6)
The enthalpy of subsection outlet was calculated from Eq. (7).
isub;out ¼ isub;in þ Qc;sub =mr
The vapor quality of refrigerant is calculated as equation (9) at each subsection.
xsub;out ¼ xsub;in Qc;sub
mr $ifg
(9)
3. Results and discussions
The various physical properties of CO2 have an influence on the condensation heat transfer at subcritical conditions. Therefore, an accurate understanding of the physical properties is very important for calculating heat transfer in the design of CO2 condensers. All of these physical properties in Figs. 5e9 were obtained from REFPROP (version 8.0). Fig. 5 shows the specific heats of liquid and vapor phase with saturation temperature. These specific heats are nearly constant at 0e25 C, and increase rapidly near the critical point. These trends mean that the heat transfer coefficient has a maximum value as the
Table 2 Summary of the estimated uncertainty. Parameters
The condensing heat transfer rates in the subsections were obtained by the mass flow rate and the temperature difference of the cooling water.
Qc;sub ¼ mcs $cp;cs $ Tcs;out Tcs;in
(8)
3.1. Physical properties of CO2
where Tre,w,out is the outside wall surface temperature of copper tube as shown in Eq. (4) and Kw is the thermal conductivity of the copper tube.
Tre;w;out ¼
xsub;in ¼ isub;in il ifg
(7)
Measured quantities Temperature Pressure Pressure drop Mass flow of refrigerant Calculated quantities Mass flux Heat flux Heat transfer coefficient Enhancement factor
Uncertainty 0.1 C 2 kPa 2.8 kPa 1% 1.09% 1.31% to 3.17% 4.19% to 8.96% 9.76% to 13.42%
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Table 3 Experimental conditions.
1200
Refrigerant
CO2 (R744)
Test section
Smooth tube
Microfin tube
Inner diameter (outer diameter) of copper tube, [mm] Mass flux, [kg/(m2s)] Vapor quality, [/] Saturation temperature, [ C] Saturation pressure, [MPa]
4.95 (6.35)
4.6 (5.0)
100 Liquid density of CO Vapro densityof CO Density ratio of CO
1000
Density [kg/m3]
400, 600, 800 0.0e1.0 20, 25, 30 5.76, 6.43, 7.21
80
Density ratio of R134a 800
60 600 40 400
Density ratio
56
20
200
C.
temperature approaches near 30 One of the most important properties is the specific heat that has an effect on the condensation heat transfer coefficient which increases with the increasing specific heat. Fig. 6 presents the liquid and vapor density, and the ratio of liquid to vapor density with saturation temperature ranging from 0 to 30 C. The density ratio of CO2 is higher than that of R134a. The low density ratio of CO2 may give more homogenous two-phase flow than with other refrigerants. The liquid to vapor density ratio plays an important role in a condenser since it determines the flow pattern and thus the heat transfer coefficient. An effect of the low vapor density is the volume flow and flow velocity which increases the heat transfer coefficients. The higher vapor density indicates the high volumetric refrigeration capacity (VCR) of CO2. Fig. 7 illustrates the surface tension with respect to saturation temperature. This decrease with temperature and become zero at the critical point. The surface tension decreases with saturation temperature, which causes the decreased heat transfer coefficient at the near critical temperature (31.1 C). As shown in Fig. 7, the surface tension of CO2 is 2.5 times smaller than that of R134a at 0 C. Low surface tension of CO2 has an effect on the forced convective heat transfer during condensation. Especially, low surface tension becomes thinner the liquid film of the upper tube wall and thicker the liquid film of the bottom tube wall, and thus the convective heat transfer to the tube wall is enhanced. Fig. 8 displays the thermal conductivities of liquid and vapor phase with varying saturation temperature. A high thermal conductivity is essential for heat transfer coefficients for single- and two-phase flow. Especially, the liquid thermal conductivity increases the heat transfer in liquid film of flow patterns such as annular flow and annular-droplet flow. And while the vapor
0 0
10
15
20
25
0 35
30
o
Tsat [ C] Fig. 6. Density and density ratios of liquid and vapor phase of CO2 and R134a with saturation temperature.
thermal conductivity increases with saturation temperature, the liquid thermal conductivity decreases and thus increases at more than 20 C. This trend has an influence on the condensation heat transfer at the near critical temperature. Fig. 9 shows the liquid and vapor viscosity, and the ratio of liquid to vapor viscosity with saturation temperature of 0e30 C. These viscosities are important factors for the fluid flow behaviors, and convective heat transfer and pressure drop characteristics during condensation. Additionally, another important parameter to describe the effect of saturation temperature is the liquid viscosity. This viscosity is higher at lower saturation temperature, which results in the decrease of heat transfer coefficients for lower saturation temperature. As mentioned above, the physical properties of CO2 are quite different from those of conventional refrigerants. In case of applying CO2 condenser to small diameter and microfin tubes, the heat transfer in these tubes can be improved. And then the condenser using high pressure fluid like CO2 can be compact. 3.2. Flow pattern Even though the CO2 flow patterns at near critical temperatures in smooth and microfin tubes are not presented in open literature,
60
14
Liquid specific heat Vapro specific heat
CO R134a
12
Surface tension [mN/m]
50
Specific heat [kJ/(kgK)]
5
40
30
20
10
0
10 8 6 4 2 0
0
5
10
15
20
25
30
o
Tsat [ C] Fig. 5. Specific heats of liquid and vapor phase with saturation temperature.
35
0
5
10
15
20
25
o
Tsat [ C] Fig. 7. Surface tension of CO2 with saturation temperature.
30
35
120
100
100
80 Liquid thermal conductivity Vapor thermal conductivity 60
40
57 10 Liquid viscosity Vapor viscosity Viscosity ratio
8
80 6 60 4 40
Viscosity ratio
120
Viscosity [106Pa s]
Thermal conductivity [mW/(mK)]
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
2
20
20 0 0
5
0 0
5
10
15
20
25
30
10
15
20
25
30
0 35
Tsat [oC]
35
o
Tsat [ C]
Fig. 9. Viscosity and viscosity ratio of liquid and vapor phase with saturation temperature.
Fig. 8. Thermal conductivity of liquid and vapor phase with saturation temperature.
3.3. Local heat transfer it is necessary to predict the flow patterns with existing flow pattern maps in order to analyze the measured condensation heat transfer coefficients. This study was applied to the representative modified flow pattern map of Taitel and Dukler [21] recommended by Jang and Hrnjak [16]. As stated earlier, they investigated experimentally for condensation flow pattern of CO2 in the 6.1 mm diameter tube. They concluded that the flow map of Taitel and Dukler [21], shown below, correctly predicts 19 out of the 25 observations. It is useful for predicting flow pattern during condensation of CO2 in horizontally smooth and microfin tube. Fig. 10(a) and (b) presents the flow pattern predictions for the experiment conditions using the flow pattern map of Taitel and Dukler [21]. The parameters for x- and y-axis in Fig. 10(a) and (b) are defined by Eqs. (10)e(13).
F ¼
rv
ðrl rv Þ
0:5 $
jv ðdi $g$cos qÞ
"
rv $j2v $jl K ¼ ðrl rv Þ$g$cos q$nl T ¼
ðdp=dzÞl ðrl rv Þ$cos q$g
X ¼
dp dz
0:5
(10)
#0:5 (11)
0:5
0:5 dp dz v l
(12)
(13)
According to this map, two flow patterns are present during condensation: stratified wavy and annular. A very limited number of points fall in the stratified wavy region, while most of the data points for CO2 lay in the annular regions. These results suggest that annular flow can be considered as an important factor in the condensation heat transfer characteristics of CO2 in the smooth- and microfin-tube. Moreover, we believe that the condensation flow mechanisms of CO2 were similar to those of non-CO2. But it is necessary to confirm accurately flow characteristics during condensation of CO2 in horizontal smooth and microfin tubes. Therefore, additional research is needed to further elucidate the flow patterns and investigate the CO2 condensation flow mechanisms in smooth and microfin tubes.
3.3.1. The effect of mass flux Fig. 11 presents the measured heat transfer coefficient of condensing CO2 with the change of vapor qualities and mass fluxes at saturation temperatures of 20 C in a smooth- and microfin-tube. As shown in Fig. 11, the condensation heat transfer coefficients in two tubes have a tendency to obviously increase with increasing vapor quality. This is because mean density decreases with the increase of vapor quality and thus liquid and vapor phase velocity increases. Also, as shown in Fig. 11, the experimental data show noticeable increase with respect to mass flux at a given saturation temperature of 20 C. This trend can be explained by the increasing Reynolds number with an increase of mass flux. With the increase of mass flux, the enhancements of the heat transfer coefficients in the microfin tube are higher than those in the smooth tube. It can be explained that the turbulent flow of CO2 near the wall in the microfin tube is increased considerably compared to that in the smooth tube. As shown in Fig. 11, the enhancement of heat transfer coefficients at high quality is obviously higher than those at low quality. This is due to the major effect of a forced convective condensation heat transfer at liquid and vapor interface and at the tube wall. These heat transfer trends are almost identical to the study results of Park and Hrnjak [17]. They presented that the heat transfer is enhanced with the increase of mass flux and vapor quality which raises liquid and vapor phase velocity in a channel. These trends shown in Fig. 11 are typical characteristics of flow condensation in an annular flow pattern which was predicted by flow pattern maps as shown in Fig. 10. We can know from above results that the condensation heat transfer characteristics of CO2 with mass fluxes in the microfin tube are almost identical to those in the smooth tube. 3.3.2. The effect of saturation temperature Fig. 12 illustrates the effect of saturation temperature on the condensation heat transfer in the smooth- and microfin-tube as a function of vapor quality. In Fig. 12, the heat transfer coefficient for two tubes decreases with increasing saturation temperature. The primary effect of saturation temperature is related to CO2 thermophysical properties. Park and Hrnjak [17] used Akers et al. [22] correlation, as represented by Eq. (14), which is simple form and is widely used, to explain this trend with thermophysical
58
a
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
104
22
102 Annular-droplets
b
18
Bubbly
101 b
K
102
100 Intermittent (Slug and Plug)
Smooth tube G=400 kg/(m2s)
10-1
T or F
d Stratified wavy
6
2 0 0.0
10-2
10-1
100
101
102
101
a
103
d
2
10
100 Intermittent (Slug and Plug)
Microfin tube 2
T or F
d b
Stratified wavy
10-1
G=400 kg/(m s) 2
G=600 kg/(m s) G=800 kg/(m2s)
101
10-2
a c 100 10-3
10-2
Stratified smooth
c
10-1
101
100
10-3 102
Χ Fig. 10. Experimental data of CO2 on the flow pattern map by Taitel and Dukler in the smooth- and microfin tube. (a) Smooth tube, (b) Microfin tube.
kl
di
1=3 $cp;l $ 7=15
1=3
$Prl
0:5 0:8 r x þ1 $ l ð1 xÞ rv
0:5 0:8 r x þ1 $ l ð1 xÞ rv
1=3 $cp;l r $ l 7=15
ml
(16)
rv
22 20 18
$Re0:8 l $
1.0
From the results stated above, it can be concluded that Akers et al. [22] correlation cannot predict the condensation heat transfer coefficient of CO2 at saturation temperature above 20 C. Zilly et al. [23] presented the similar result for the CO2 condensation heat transfer inside horizontal tubes. They explained that the effect of saturation temperature is related to thermophysical properties such as density ratio of liquid to vapor, dynamic viscosity and surface tension. The explanation for earlier two properties is almost identical to the experimental result of Park and Hrnjak [17] mentioned above. They proposed that the surface tension is another important factor to describe the effect of saturation temperature. The decrease of saturation temperature
Smooth tube T =20 [ C] T =25 [ C] T =30 [ C]
Microfin tube T =20 [ C] T =25 [ C] T =30 [ C]
16
(14)
(15)
With the decrease of saturation temperature, liquid conductivity, kl, liquid viscosity, ml, and density ratio of liquid to vapor, rl/rv, increase, and liquid specific heat, cp,l, decreases. The contribution of the increased liquid conductivity and density ratio of liquid to vapor to heat transfer is higher than the effect of viscosity and specific heat, which results in higher coefficients at lower saturation temperature. But, at saturation temperature above 20 C, the explanation of Park and Hrnjak [17] cannot be applied for the effect of saturation
hc,loc [kW/(m2K)]
0.8
2=3
kl hz
properties. The relation between the condensation heat transfer coefficients and thermophysical properties is presented by Eq. (14) which is rearranged from Akers et al. correlation.
h ¼ 0:026$
0.6
temperature. As shown in Table 4 and Fig. 13, because the liquid conductivity decreases and increases with the increase of saturation temperature. In Eq. (16), the values including four thermophysical properties decrease with increasing saturation temperature. And the condensation heat transfer coefficients of CO2 increase with saturation temperature.
b Bubbly
K
0.4
Fig. 11. Heat transfer coefficient of CO2 with mass flux and vapor quality at saturation temperature of 20 C in the smooth- and microfin tube.
102 Annular-droplets
ml
0.2
x
Χ
hz
8
c
Stratified smooth
104
2=3
10
10-3
10-3
kl
12
4
100
b
T
10-2
a c
G =400 [kg/(m s)] G =600 [kg/(m s)] G =800 [kg/(m s)]
14
G=600 kg/(m2s) G=800 kg/(m2s)
101
Microfin tube
16
d
hc,loc [kW/(m2K)]
a
103
Smooth tube G =400 [kg/(m s)] G =600 [kg/(m s)] G =800 [kg/(m s)]
20
G =800 [kg/(m s)] 14 12 10 8 6 4 2 0 0.0
0.2
0.4
0.6
0.8
1.0
x Fig. 12. Heat transfer coefficient of CO2 with saturation temperature at mass flux of 800 kg/m2s in the smooth- and microfin tube.
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
59
Table 4 Variation of thermophysical properties with saturation temperature. Tsat[ C]
kl, [mW/(mK)]
ml, [106 Pa s]
rl/rv, [/]
cp,l, [kJ/(kgK)]
s, [mN/m]
85.683 80.789 95.356
66.148 57.048 43.768
3.982 2.927 1.719
4.263 6.467 35.338
1.203 0.552 0.054
r 1=3 k2=3 $cp;l = 7=15 $ l , l ml rv [Jm2/(kgs2K2)]
20 25 30
increases surface tension which can be regarded as work per surface area of a liquid droplet and higher surface tension may increase heat transfer coefficient. More energy from the condensation process can be stored as surface energy, which increases the condensation rate. This effect would also explain higher heat transfer coefficient at lower saturation temperature, assuming the increased surface tension dries the upper tube wall and the vapor phase can condensate directly to the tube wall. It can be seen from above results that the effect of saturation temperature is related to the thermal properties of CO2 which have an effect on the condensation heat transfer coefficient in the smooth- and microfin-tube. 3.3.3. The effect of tube geometry An enhancement factor (EF) is generally used as a measure of the heat transfer enhancement. EF is defined as the ratio (EF ¼ h/hplate) of the heat transfer coefficient of an enhanced surface to that of a plain surface at the same condition. But, in this study, for investigating the increase in the heat transfer coefficient of the microfin tube to smooth tube, we uses the enhancement factor (EF), which is defined as the ratio (EF ¼ hmicrofin/hsmooth) of the heat transfer coefficient in the microfin tube to that in the smooth tube. But, since the tube diameter of the smooth tube used in this study is different to that of microfin tube, the heat transfer areas (Amicrofin and Asmooth) of the smooth- and microfin-tube are included at a denominator and numerator of Eq. (17). The enhancement factor used in this study is an average one, EFavg given by Eq. (17).
EFavg
havg $A microfin ¼ havg $A smooth
(17)
Where, (havg$A)microfin and (havg$A)smooth are the product of the average heat transfer coefficient and area in the smooth and microfin tube, respectively.
8.891 9.024 15.396
Fig. 14 presents the enhancement factors (EFavg) with mass fluxes at a given condition. It can be seen from Fig. 14 that EFavg varies from 1.006 to 1.48 with increasing mass flux. It means that the heat transfer coefficients for the microfin tube are greater than those for the smooth tube in the entire quality range. The reasons for the condensation heat transfer enhancement in the microfin tube are summarized as follows: first of all the mere increase of the effective heat exchange area. Additionally, the turbulence induced in the liquid film and the surface tension effect on the liquid drainage promotes an early transition from the wavy-stratified flow to the annular flow. [4]. Also, the heat transfer enhancement at high mass flux is higher than that at low mass flux due to the main effect of the forced convection heat transfer as mentioned earlier. 3.4. Comparison of existing heat transfer correlations The condensation heat transfer of non-CO2 has been studied by a large number of researchers in horizontal smooth and microfin tubes. But the studies of the heat transfer of CO2 during condensation inside smooth and microfin tubes have been limited in open literature until now. Determination of accurate correlation is necessary for predicting the condensation heat transfer coefficient in the horizontal smooth- and microfin-tube. Therefore, the experimental data for CO2 inside the horizontal smooth- and microfin-tube were compared with related previous correlations. The correlations of Cavallini and Zecchin [12], Dobson et al. [11], Shah [13], Cavallini et al. [14] and Kondou and Hrnjak [1] are used for the comparison of the data in the smooth tube. The correlations of Han and Lee [24], Chamra et al. [25], Cavallini et al. [4], Haraguchi et al. [6], Jung et al. [26] and Honda et al. [27] are used for the microfin tube. The compared results for condensation heat transfer coefficients in the smooth tube are shown in Table 5 and Fig. 15, where
1.8
EFavg [/]
1.6
T =20 [ C]
T =20 [ C]
T =20 [ C]
T =25 [ C]
T =25 [ C]
T =25 [ C]
T =30 [ C]
T =30 [ C]
T =30 [ C]
1.4
1.2
1.0
0.8 200
400
600
800
Gre [kg/m2s] Fig. 13. Physical properties of CO2 with respect to saturation temperature from 20 to 30 C.
Fig. 14. Enhancement factor (EF) at an entire range of mass fluxes.
1000
60
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Table 5 The deviations between the calculated and experimental heat transfer coefficients in the smooth tube. 8 8
9 ! 9
= = n h
cal;i hexp;i n savg ¼ =N 100; sabs ¼
N 100 i ¼ 1
; ; : : hexp;i hexp;i i¼1
Author
Refrigerant
Table 6 The deviations between the calculated and experimental heat transfer coefficients in the horizontal microfin tube. 8 8
9 ! 9
= = n h
cal;i hexp;i n savg ¼ =N 100; sabs ¼
N 100 i ¼ 1
; ; : : hexp;i hexp;i i¼1
Author
Refrigerant
CO2
Cavallini and Zecchin [12] Dobson et al. [11] Shah [13] Cavallini et al. [14] Kondou and Hrnjak [1]
CO2
savg
sabs
53.2 44.5 41.7 28.8 20.7
53.2 44.5 41.7 28.8 20.7
statistical comparison of correlations, based on average and absolute deviation is given. Among these correlations listed in Table 5, Kondou and Hrnjak [1] correlation is proposed only for condensing CO2 in a horizontal smooth tube and the others are for non-CO2. All correlations proposed for non-CO2 showed enormous deviations with experimental data. The correlations of Cavallini and Zecchin [12], Dobson et al. [11] and Shah [13] failed in predicting the current experimental data with mean deviations of higher than 40%. The large deviation of these correlations is due to experimental conditions such as higher reduced pressure and mass flux, and different characteristics of CO2 such as lower liquid density, higher vapor density and lower surface tension etc. Cavallini et al. [14] correlation had an agreement to experimental data within 28.8%. The condensation latent heat and liquid thermal conductivity of CO2 at a given saturation temperature are slightly higher than those of various refrigerants used in Cavallini et al. correlation. The correlation of Kondou and Hrnjak [1] proposed for condensing CO2 is reasonably correlated to the current experimental data within 20.7%. They applied Cavallini et al. [14] equation using the approach of Fujii and Watabe [28], calculating the liquid properties at the mean temperature.
Han and Lee [24] Chamra et al. [25] Haraguchi et al. [6] Cavallini et al. [4] Jung et al. [26] Honda et al. [27]
sabs
42.7 34.5 46.2 21.1 28.8 36.7
42.7 35.5 46.2 21.1 28.8 36.7
Kondou and Hrnjak proposed the modified correlation that agrees well with the experimental data not only in subcritical region but also at lower pressure below 6.5 MPa. But they proposed their modified correlation over limited mass fluxes of Gre < 250 kg/ (m2s). And the reduced pressures in this study are much higher than the validity range of the original correlation of Cavallini et al. [14]. Thus, this correlation needs to improve for higher mass flux ranges of Gre > 250 kg/(m2s). As shown in above comparison results, it is difficult to apply existing correlations which are proposed for non-CO2 in the horizontal tubes to predict the condensation heat transfer coefficient of CO2. Thus new and reliable correlation is necessary for the optimal design of condenser using CO2. The comparison results for the heat transfer coefficients during condensation in the microfin tube are shown in Table 6 and Fig. 16. The correlations of Han and Lee [24], Chamra et al. [25], Haraguchi et al. [6], Jung et al. [26] and Honda et al. [27] were developed for nonCO2, Cavallini et al. [4] correlation were specifically developed from R22, R134a, R123, R410A and CO2 data taken inside microfin tubes. Most of the correlations proposed for non-CO2 showed large deviations with experimental data. The correlations of Han and Lee [24], Chamra et al. [25], Haraguchi et al. [6], and Honda et al. [27]
20
25
+30%
+30% 20
15
-30%
10
Cavallini and Zecchin Dobson et al.
5
-30%
hpred [kW/(m2K)]
hpred [kW/(m2K)]
savg
Shah
15
10 Han and Lee Chamra et al. Haraguchi et al. Cavallini et al. Jung et al. Honda et al.
5
Cavallini et al. Kondou and Hrnjak 0
0
0
5
10
15
20
2
hexp [kW/m K] Fig. 15. Comparison of the predicted and experimental heat transfer coefficients for CO2 condensing in the smooth tube.
0
5
10
15
20
25
hexp [kW/m2K] Fig. 16. Comparison of the predicted and experimental heat transfer coefficients for CO2 condensing in the microfin tube.
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
61
1.25
12
1.20
Penalty factor, [/]
Pressure drop, [kP/m]
10
8
6
4 Smooth tube Tre=30 [oC] Tre=25 [oC] o Tre=20 [ C]
2
Microfin tube
1.15
1.10
Microfin tube T =30 [ C] T =25 [ C] T =20 [ C]
Tre=30 [oC] o Tre=25 [ C] Tre=20 [oC] 1.05
0 400
600
400
800
failed in predicting the present experimental data with mean deviations of higher than 30%. The correlations of Cavallini et al. [4] and Jung et al. [26] averagely reasonably correlated to the current experimental data within 20%. Among six correlations, the comparison of the predicted values given by the Cavallini et al. [4] correlation to the experimental values showed the best satisfaction. For these comparison results, some researchers reported the following opinion. Zilly et al. [23] showed that the correlation of Haraguchi [6] is overpredicted, but Cavallin et al. [4] correlation is in good agreement with the experimental data. Also, Cavallini et al. [4] presented that only data by Haraguchi et al. [6] is underestimated more than 20%. Kondou and Hrnjak [1] showed that the experimental heat transfer coefficient was compared to the correlation for condensation flow of other refrigerants proposed by Cavallini et al. [4] and Haraguchi et al. [6]. These correlations were developed on experimental data obtained at much lower reduced pressures. Cavallini et al. [4] correlation showed the best agreement to the experimental heat transfer coefficient. Cavallini et al. [4] developed their correlation using the experimental CO2 data of Zilly et al. [23], measured in a limited saturation temperature of 15 to 25 C and mass flux of 200e400 kg/(m2s). Zilly et al. [23] experimental conditions are very different from those in Table 1. Therefore, it is difficult to apply Cavallini et al. [4] correlations to predict the condensation heat transfer coefficients of CO2 for high saturation temperatures in the smooth- and microfin-tube. Thus it is necessary to develop a new and reliable condensation heat transfer correlation which can be applied to the smooth and microfin tubes for various conditions. 3.5. Pressure drop The condensation pressure drop of CO2 in the smooth- and microfin-tube is shown in Fig. 17. As expected, the pressure drop in both tubes increases almost linearly as the mass flux increases. The pressure drop at 20 C is higher than at 30 C in the smooth- and microfin-tube. As represented in Figs. 6 and 9, this is because liquid viscosity increases and vapor density decreases with the decrease of saturation temperature. With the decrease of vapor density, the vapor phase velocity in a tube increases at an identical mass flux condition, which results in the increase of pressure drop. Also, the pressure drop in the microfin tube is higher than that in the smooth tube. This reason is the turbulent mixing effect and smaller tube diameter of microfin tube.
800 2
Gre, [kg/(m s)]
Gre, [kg/(m2s)] Fig. 17. Pressure drop with respect to mass flux and saturation temperature.
600
Fig. 18. The effect of mass flux and saturation temperature on penalty factor.
The pressure drop characteristics of microfin tubes were investigated in terms of the pressure drop penalty factor (PF). The PF is the ratio of pressure drop of the microfin tube and that of the smooth tube at the same operating condition. Therefore, the values of the PF in Fig. 18 are defined as follows:
PF ¼
ðDP$AÞmicrofin ðDP$AÞsmooth
(18)
Where, (DP$A)microfin and (DP$A)smooth are the product of the pressure drop and heat transfer area in the smooth and microfin tube, respectively. The PF in the present study varies from 1.14 to 1.23 with increasing mass flux as shown in Fig. 18. It means that the pressure drop of microfin tube is higher than that of smooth tube. The reason is the increased turbulent mixing and swirl effects of the microfin tube. As a result, although the pressure drop of CO2 in the microfin tube is higher than that in the smooth tube, the pressure drop of CO2 is much lower than conventional refrigerants such as R410a and R134a at an identical test conditions due to the significantly higher vapor density of CO2. This CO2 pressure drop characteristics are valid for microfin tubes [17].
4. Conclusions In this study, the condensation heat transfer experiments of CO2 were carried out in the horizontal smooth- and microfin-tube at mass fluxes from 400 to 800 kg/(m2s), saturation temperatures from 20 to 30 C, and vapor qualities from 0.1 to 0.9. The condensation heat transfer and pressure drop data in the horizontal smooth- and microfin-tube were measured and their applicability to the various condensation heat transfer coefficient models under different operating conditions was examined. This study performed tests for the validation of models with different operating conditions. The experimental results showed that annular flow almost dominates the major of condensation flow in the smooth- and microfin-tube and it can be considered importantly for heat transfer and pressure drop characteristics as well as new and reliable condensation heat transfer correlations in the smooth- and microfin-tube. The heat transfer coefficients in the smooth- and microfin-tube have similar trends to increase with respect to increased mass flux and decreasing saturation temperature. The average heat transfer
62
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
enhancement factor (EF) for the microfin tube is approximately from 1.006 to 1.48, and penalty factor (PF) varies from 1.14 to 1.23. Thus, the use of the microfin tube will be prosperous for the condensers using CO2 such as working fluid, if the fin geometry such as the fin shape and height can be optimized for the better performance. Most of the proposed correlations for non-CO2 failed to predict the condensation heat transfer coefficients in the smooth- and microfin-tube. However, for the overall experimental data in the microfin tube, the Cavallini et al. correlation predicted the present experimental data better than the other correlations. The correlation by Kondou and Hrnjak showed relatively good agreement with experimental data in the smooth tube. As a result, there are little researches for the condensation heat transfer and pressure drop in the horizontal smooth- and microfintube. Thus, the accurate condensation heat transfer and pressure drop data under various test conditions is necessary for the optimal design of CO2 condenser using the smooth- and microfin-tube. Furthermore, it is necessary to develop more accurate and reliable correlation to predict heat transfer characteristics during condensation in the smooth- and microfin-tube. Therefore, the dominant annular-flow pattern during CO2 condensation must be considered importantly for new correlations which can be also applied to higher saturation temperature and mass flux conditions. These correlations based on the annular flow model for the smooth- and microfin-tube will give better prediction accuracy. Acknowledgements The work presented in this paper is part of the project ‘Development of high efficient cooling and heating system using natural refrigerant of CO20 sponsored by Ministry of Commerce, Industry, and Energy. The support of these sponsors is gratefully acknowledged. References [1] C. Kondou, P. Hrnjak, Heat rejection from R744 flow under uniform temperature cooling in a horizontal smooth tube around the critical point, Int. J. Refrigeration 34 (2011) 719e731. [2] J.M. Cho, M.S. Kim, Experimental studies on the evaporative heat transfer and pressure drop of CO2 in smooth and micro-fin tubes of the diameter of 5 and 9.52 mm, Int. J. Refrigeration 30 (2007) 1e9. [3] Y.J. Kim, J. Jang, P. Hrnjak, M.S. Kim, An experimental study on condensation heat transfer of CO2 at low temperature in smooth and micro-fin tubes, in: Proceedings of SAREK (2003), pp. 50e55. [4] A. Cavallini, D. Del Col, S. Mancin, L. Rossetto, Condensation of pure and nearazeotropic refrigerants in microfin tubes: a new computational procedure, Int. J. Refrigeration 32 (2009) 162e174. [5] J. Yu, S. Koyama, Condensation heat transfer of pure refrigerants in microfin tubes, in: Proceeding of International Refrigeration Conference at Purdue. 1720 July, West Lafayette, Indiana, USA (1998), pp. 325e330. [6] H. Haraguchi, S. Koyama, J. Esaki, T. Fujii, Condensation heat transfer of refrigerants HCFC134a, HCFC123, and HCFC22 in a horizontal smooth tube
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19] [20]
[21]
[22] [23]
[24]
[25]
[26]
[27] [28]
and a horizontal microfin tube, in: Proceeding of 30th National Symposium of Japan, Yokohama, Japan (1993), pp. 343e345. M.A. Kedzierski, J.M. Goncalves, Horizontal convective condensation of alternative refrigerants within a micro-fin tube NISTIR 6095, U.S. Dept. Commerce, 1997. M.H. Kim, J.S. Shin, B.H. Lim, Condensation heat transfer of R22 and R410A in horizontal smooth and microfin tubes, zero Leakage-minimum charge, IIR-IIF, Stockholm, Sweden, 2002, Section B and E253e261. S. Koyama, R. Yonemoto, Experimental study on condensation of pure refrigerants in horizontal micro-fin tube - proposal of correlation for heat transfer and frictional pressure drop, in: Proceeding of International Refrigeration Conference at Purdue. 17e20 July, West Lafayette, Indiana, USA (2006). H.S. Wang, J.J. Wei, H. Honda, J.W. Rose, Condensation of New Refrigerants in Horizontal Microfin Tubes: a new correlation, in: Proceeding of International Congress Refrigeration, ICR07, Beijing, China (2007). M.K. Dobson, J.C. Chato, D.K. Hinde, S.P. Wang, Experimental evaporation and condensation of Refrigerant R-12 and R-134a, ASHRAE Trans. 100 (1994) 744e754. A. Cavallini, R. Zecchin, A Dimensionless correlation for heat transfer in forced convection condensation, in: , Proc, 6th Intl. Heat Transfer Conf. Tokyo, vol. 3, 1974, pp. 309e313. M.M. Shah, A general correlation for heat transfer during film condensation inside pipes, Int. J. Heat Mass Transfer 22 (1979) 547e556. A. Cavallini, D. Del Col, L. Doretti, M. Matkovic, L. Rossetto, C. Zilio, G. Censi, Condensation in horizontal smooth tubes: a new heat transfer model for heat exchanger design, Heat Trans. Eng. 27 (2006) 31e38. S. Koyama, C. Kondou, K. Kuwahara, An experimental study on condensation of CO2 in a horizontal micro-fin tube, 12th Int. Refrig. A/C Conf. at Purdue, Paper No. 2322, 2008. J. Jang, P.S. Hrnjak, Condensation of CO2 at Low Temperature inside Horizontal Microfinned Tubes, ACRC CR-56, University of Illinois at UrbanaChampaign, 2004. C.Y. Park, P. Hrnjak, R744 flow condensation heat transfer and pressure drop in multi-port microchannels at low temperatures, Int. journal Refrigeration 32 (2009) 1129e1139. N. Hayes, Amir Jokar, Zabid H. Ayub, Study of carbon dioxide condensation in chevron plate exchangers; heat transfer analysis, Int. J. Heat Mass Transfer 54 (2011) 1121e1131. V. Gnielinski, New equations for heat and mass transfer in turbulent pipe and channel flow, Int. Chem. Eng. 16 (1976) 359e368. E.W. Lemmon, M.L. Huber, M.O. McLinden, Reference Fluid Thermodynamic and Transport Properties, NIST Standard Reference Database 23 Version 8.0 (2007). Y. Taitel, A.E. Dukler, A model for predicting flow Regime transitions in horizontal and near horizontal gas-liquid flow, Am. Inst. Chem. Eng. J. 22 (1976) 47e55. W.W. Akers, H.A. Deans, O.K. Crosser, Condensing heat transfer within horizontal tubes, Chem. Eng. Prog. Symp. Ser. 55 (1959) 171e176. J. Zilly, J. Jang, P.S. Hrnjak, Condensation of CO2 at Low Temperature inside Horizontal Microfined Tubes, ACRA (Air Conditioning and Refrigeration Center) Report, March 2003, University of Illinois at Urbana-Campaign, 2003. D. Han, K.J. Lee, Experimental study on condensation heat transfer enhancement and pressure drop penalty factors in four microfin tubes, Int. J. Heat Mass Tran. 48 (2005) 3804e3816. L.M. Chamra, P.J. Mago, M.O. Tan, C.C. Kung, Modeling of condensation heat transfer of pure refrigerants in micro-fin tubes, Int. J. Heat Mass Tran. 48 (2005) 1293e1302. D.S. Jung, Y.M. Cho, K.H. Park, Flow condensation heat transfer coefficients of R22, R134a, R407C, and R410A inside plain and microfin tubes, Int. J. Refrigeration 27 (2004) 25e32. H. Honda, A.T. Wijayanta, N. Takata, Condensation of R407C in a horizontal microfin tube, Int. J. Refrigeration 28 (2005) 203e211. T. Fujii, M. Watabe, Laminar film condensation in the subcritical region e Gravity controlled condensation, JSME Trans. B. 53 (1987) 1801e1806 (in Japanese).
Contents lists available at SciVerse ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Condensation heat transfer characteristics of CO2 in a horizontal smooth- and microfin-tube at high saturation temperatures Chang-Hyo Son*, Hoo-Kyu Oh Department of Refrigeration and Air-Conditioning Engineering, College of Engineering, Pukyong National University, San 100, Yongdang-dong, Nam-gu, Pusan 608-739, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 May 2011 Accepted 9 December 2011 Available online 16 December 2011
The condensation heat transfer characteristics for CO2 at high saturation temperatures in a horizontal smooth- and microfin-tube were investigated by experiment with respect to condensation temperature and mass flux. The test sections consist of 2400 mm length with a horizontal copper tube of 4.6 mm (smooth) and 4.95 mm (microfin) inner diameter. The experiments were conducted at refrigerant mass flux of 400e800 kg/(m2s), and saturation temperature of 20e30 C. The main experimental results showed that annular flow almost dominated the major of condensation flow in the horizontal smoothand microfin-tube. The condensation heat transfer coefficients for the smooth- and microfin-tube increase with the decreasing saturation temperature and increasing mass flux. The average heat transfer enhancement factor (EF) for the microfin tube is approximately from 1.006 to 1.48, and penalty factor (PF) varies from 1.14 to 1.23. The experimental data in the smooth- and microfin-tube were compared against previous heat transfer correlations. Most of correlations failed to predict the experimental data. However, the correlations by Kondou and Hrnjak [1] and Cavallini et al. [4] showed relatively good agreement with experimental data in the smooth- and microfin-tube, respectively. But it is necessary to develop accurate and reliable correlation to predict condensation heat transfer coefficient at high saturation temperatures in the horizontal smooth- and microfin- tube. Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Keywords: Condensation heat transfer coefficient Carbon dioxide (CO2) Microfin tube Smooth tube Subcritical cycle High saturation temperature
1. Introduction During summer season, commercial refrigeration systems using CO2 spend most operating hours in a transcritical cycle, while these systems at autumn and spring season are operated as a subcritical cycle. [1] The operation conditions of the subcritical cycle are quite different from those of the transcritical one. In the summer season, the heat rejection temperatures of the transcritical system of CO2 are usually above the critical temperature of CO2 (31.1 C), the systems using CO2 will have to operate in the transcritical cycle. As such, the heat rejection takes place above the critical pressure in a so-called gas cooler (corresponding to the condenser in the subcritical systems), while the heat rejection process at autumn and spring season remains below the subcritical region. Consequently, the CO2 heat transfer characteristics of the heat rejection process in the subcritical cycle are considerably different from those in the transcritical cycle. Thus, it is necessary to understand the characteristics of the subcritical cycle to ensure
* Corresponding author. Tel.: þ82 51 629 6802; fax: þ82 51 611 6368. E-mail addresses: [email protected], [email protected] (C.-H. Son).
reliable system performance and to evaluate the annual performance. Heat transfer enhancement has been a substantial factor to obtain high energy efficiency in refrigeration and air-conditioning applications. Microfin tubes represent a technology that has been able to beneficially enhance condensation heat transfer without causing similar increases in pressure drop and refrigerant charge, both in single-phase and two-phase application. Condensation heat transfer in microfin tubes is increased of (i) the larger surface area (A), (ii) the thinning of the condensate film (d) by a redistribution of the liquid due to spiraling and surface tension forces, and (iii) the presence of film disturbances caused by the presence of the fins, all of which results in an increase of the thermal capacity performance of the heat exchanger [2,3]. For conventional refrigerants, flow condensation in smooth and microfin tubes was investigated by several researchers [4e14]. But, as shown in Table 1, the characteristics of CO2 flow condensation heat transfer in horizontal smooth and microfin tubes are not well known. However, in recent years, some researchers [15e18] have studied for CO2 condensation heat transfer inside smooth and microfin tubes. As shown in Table 1, most of the previous studies about flow condensation in tubes were performed at low temperatures below
1359-4311/$ e see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.applthermaleng.2011.12.017
52
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Nomenclature A cp d e EF dp/dz G g ifg j kw m N n P DP PF Pr Q Re T VCR X x dz
heat transfer area, m2 specific heat at constant pressure, kJ/(kgK) diameter, m fin height, mm enhancement factor frictional pressure gradient along tube axis, kPa/m mass flux, kg/(m2s) gravitational acceleration, m/s2 latent heat, kJ/kg superficial velocity, m/s thermal conductivity, kW/(mK) mass flow rate, kg/s number of data number of microfins pressure, kPa pressure drop, kPa penalty factor Prandtl number, mcp/k heat capacity, kW Reynold number, rvdi/m temperature, K volumetric refrigeration capacity, kJ/m3 Martinelli parameter vapor quality length of test section, m
0 C. Also, a few of CO2 condensation heat transfer coefficient data in smooth and microfin tubes have been reported in any publication [15,16] known to the authors. More experimental studies are necessary to develop CO2 condensation heat transfer database and correlations in smooth and microfin tubes. Accordingly, the purpose of this study is to present new experimental data and search the suitable existing predictions which describe the present data and also analyze the experimental data to find condensation heat transfer characteristics in the horizontal smooth- and microfin-tube. 2. Experimental apparatus and procedures 2.1. Test facility The experimental facility as schematically shown in Fig. 1 is designed to investigate the condensation heat transfer coefficient of CO2 in the smooth- and microfin-tube. Basically, the main loops of the system are a CO2 loop and a cooling water loop. Detailed descriptions of the two independent loops of the test facility are provided below. As shown in Fig. 1, the CO2 loop is composed of a magnetic gear pump, a mass flow meter, two preheaters, a condenser (test section for condensation heat transfer experiment), a subcooler and a receiver etc. The refrigerant then passes in series through the magnetic gear pump, the refrigerant flow meter, the preheater, and
Greek symbols b microfin apex angle, g microfin helix angle, m dynamic viscosity, Pa s r density, kg/m3 d film thickness, m s surface tension, mN/m, deviation n specific volume, m3/kg Subscripts abs mean avg average cal calculated cr critical point cs source water exp experimental i inner, inside in inlet l liquid out outlet pre preheater pred predicted re refrigerant sat saturation sub subsection v vapor w wall
enters the test section. The subcooled refrigerant is charged in the receiver where it is further cooled to increase its density. The refrigerant liquid in the receiver is circulated by a magnetic gear pump which can be regulated by means of an inverter, and then flows through a flow meter. The liquid of refrigerant enters the preheater, where it is heated up to the desired inlet quality and temperature before entering the test section. Electrically insulated heating wires are wrapped around the surface of copper tubes in the preheater. The preheater is insulated with glass fibers and rubber. The amount of heat loss from the preheater is calibrated through pretests with water; this is correlated to the voltage input. In passing through the test section, the two-phase refrigerant is completely condensed and subcooled by cooling water. Afterwards, the CO2 is subcooled by the subcooler and goes into the liquid receiver. Finally, the subcooled refrigerant is recirculated through the refrigerant loop. The subcooler is a counterflow heat exchanger with refrigerant flowing in the inner tube and the water flowing in the annulus. It is used to condense the refrigerant leaving the test section. The mass flow rate of refrigerant into the test section is adjusted by changing the speed of the gear pump. The refrigerant flow rate is measured by the mass flow meter. The pressure of the cycle is controlled by the charged amount of refrigerant. The cooling water loop is composed of a centrifugal pump, an inline electric heater and a heat exchanger. The cooling water is pumped to the circular-tube annulus, where it absorbs the heat of
Table 1 Summary of previous studies on condensation heat transfer of CO2 in smooth and microfin tubes. Literature
Tube geometry
Tsat [ C], Psat [MPa]
x [/], qre [kW/m2]
Gre [kg/(m2s)]
Kondou and Hrnjak [1] Jang and Hrnjak [16]
Smooth tube, di ¼ 6.1 mm Smooth tube, microfin tube di ¼ 6.1 mm Microfin tube di ¼ 5.67 mm
Psat ¼ 5e7 15, 25
0e1, qre ¼ 3e30 0.1e0.9
100e240 200e400
Psat ¼ 5, 6 MPa
0.1e1.0
e
Koyama et al. [15]
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
53
Fig. 1. Schematic diagram of experimental apparatus.
the condensing refrigerant. The mass flow rate of cooling water is controlled by adjusting the metering valve and pump speed. The inlet temperature of the cooling water at the test section is controlled by both the electric heater and the refrigeration unit. The mass flow rate of cooling water is also measured by a turbine type flow meter. 2.2. Test section The schematic diagram of the test sections is shown in Fig. 2. The test sections are horizontal double-tube type heat exchangers which are made by a smooth- and microfin-tube, respectively. The refrigerant flows through the inside of copper tube and the cooling water flows through the annulus in the counterflow direction. The test sections are the smooth- and microfin-tube having insidediameters/outside-diameters of 4.95 mm/6.35 mm and 4.6 mm/ 5.0 mm, respectively. The total length of two condensers is 2400 mm and contains 12 subsections along with 200 mm in length of one subsection. In order to measure accurately the temperature of cooling water in the test sections, a mixing chamber is installed at the outlet of each subsection and inlet and outlet of test section. Fig. 3 presents the characteristic geometries of the microfin tube used in this study. This microfin tube has a single set of 55 spiral fins with spiral angle (b) of 18 , fin height (e) of 0.2 mm, fin apex angle (g) of 40 . The test sections are well insulated by thermal insulation material with a thermal conductivity of 0.04 W/(mK). The outer tube is made of a clear copper tube with an internal diameter of 11.1 mm. In order to minimize heat gain and loss from the surroundings, fiberglass insulation material was used to cover the test section as shown in Fig. 4. At the central position of each subsection the outside wall temperature of the inner tube is measured with four copper-constantan thermocouples which are distributed circumferentially at the top, left, bottom, and right sides
of the tube as shown in Fig. 4 to take care of any circumferential temperature variation. In order to verify the test facility, preliminary tests were performed. To check the accuracy, the amount of heat loss was calculated from the energy balances between the inner copper tube side and the annular side. The heat loss was 3% of the total heat transfer rate. The results indicated that all sensors and the instruments worked satisfactorily. In addition, the single phase heat transfer coefficients were calculated and compared with Gnielinski’s [19] correlation. The comparison showed that the absolute average deviation was less than 10.9%. The uncertainty changed depending on the flow conditions, so their minimum to maximum ranges are shown in Table 2. The refrigerant temperatures are measured by T-type thermocouples at the inlet and outlet of each subsection. Temperature of the cooling water is measured at inlet and outlet of each subsection. The pressure at the inlet and outlet of each subsection is measured with a pressure transducer. The pressure drop of the test section is obtained from the difference of measured pressure at the inlet and outlet. Mass flow rate of the refrigerant is measured with the mass flow meter of Oval corporation, and mass flow rate of the cooling water is measured with the flow meter of Schlumberger corporation. 2.3. Experimental method and condition As stated above, the test rig can be used to conduct the condensation heat transfer and pressure drop test. The thermodynamic properties of the refrigerants under test were incorporated in the data-acquisition system consisting of a data logger, interface card (HPIB), and a compatible PC. The data was analyzed in real time using PC and a data reduction program (MS-Excel with a Visual Basic). All of the information about the test conditions and data were displayed on the monitor during the test, the test conditions were changed based
54
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Fig. 2. Schematic diagram of test sections. (a) Smooth tube, (b) Microfin tube.
on this information. All channels in the data-acquisition system were scanned five times. The data collection took about 2 min for each run. During the condensation heat transfer test, the preheater was used to adjust the refrigerant inlet quality. By knowing the inlet temperature of the refrigerant prior to the preheater and the power supplied to the preheater, the inlet quality of the refrigerant in the test section can be calculated. The exit quality of the refrigerant can be determined by applying an energy balance on the water side. The system was allowed to come to steady-state before any data were recorded. The steady-state condition of experiment was reached when the refrigerant inlet pressure/temperature/quality, exit temperature/ quality, mass flow rate and water mass flow rate and inlet and outlet temperature keep in stationary state. The test conditions in this study are summarized in Table 3. Test runs are conducted for various mass fluxes and inlet pressures of CO2. Mass fluxes are controlled from 400 to 800 kg/(m2s) by a variable speed magnetic gear pump. The saturation temperatures of CO2 are adjusted from 20 to 30 C.
2.4. Data reduction Defect signals for checking data are processed with the computer through the data logger. In this study, the thermodynamic and transport properties of all refrigerants are calculated using REFPROP (version 8.0) [20] made by NIST(National Institute of Standards and Technology). We can use the following equations to analyze the test data, using the above mentioned properties. The heat flux, qc, supplied to inside wall surface of the copper tube is calculated as Eq. (1):
qc ¼
Qc
p$di $dz
(1)
where, dz indicates the effective length of a subsection, di is inner diameter of the smooth tube and fin tip diameter of the microfin tube, respectively. Qc is the heat transfer rate gained by cooling water flowing through the annular. The circumferential local heat
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
55
Fig. 3. Characteristics geometries of microfin tube.
Fig. 4. Setting of temperature sensor and insulation material.
transfer coefficient of refrigerant is calculated by the following Eq. (2):
With the enthalpies calculated, subsection inlet vapor quality was given by Eq. (8).
hc;loc ¼
qc Tre Tre;w;in
(2)
where, hc,loc represents the local heat transfer coefficient in the subsection of condenser, Tre is bulk refrigerant temperature in condenser. Tre,w,in is determined from average of the measured outside wall temperature via one the equation of radial thermal conduction through the wall as shown in Eq. (3):
Tre;w;in ¼ Tre;w;out Qre;sub
lnðdo =di Þ ð2p$dz$kw Þ
(3)
Tw;top þ Tw;left þ Tw;right þ Tw;bottom 4
(4)
where Tw,top, Tw,left, Tw,right, and Tw,bottom are the measured temperature at the top, both the left and right sides and bottom wall, respectively. The subcooled refrigerant enthalpy upstream of preheater was determined by measuring its temperature and pressure. Then, the subsection inlet enthalpy could be calculated from the heat input of the preheater.
isub;in ¼ ipre þ Qpre =mr
(5)
(6)
The enthalpy of subsection outlet was calculated from Eq. (7).
isub;out ¼ isub;in þ Qc;sub =mr
The vapor quality of refrigerant is calculated as equation (9) at each subsection.
xsub;out ¼ xsub;in Qc;sub
mr $ifg
(9)
3. Results and discussions
The various physical properties of CO2 have an influence on the condensation heat transfer at subcritical conditions. Therefore, an accurate understanding of the physical properties is very important for calculating heat transfer in the design of CO2 condensers. All of these physical properties in Figs. 5e9 were obtained from REFPROP (version 8.0). Fig. 5 shows the specific heats of liquid and vapor phase with saturation temperature. These specific heats are nearly constant at 0e25 C, and increase rapidly near the critical point. These trends mean that the heat transfer coefficient has a maximum value as the
Table 2 Summary of the estimated uncertainty. Parameters
The condensing heat transfer rates in the subsections were obtained by the mass flow rate and the temperature difference of the cooling water.
Qc;sub ¼ mcs $cp;cs $ Tcs;out Tcs;in
(8)
3.1. Physical properties of CO2
where Tre,w,out is the outside wall surface temperature of copper tube as shown in Eq. (4) and Kw is the thermal conductivity of the copper tube.
Tre;w;out ¼
xsub;in ¼ isub;in il ifg
(7)
Measured quantities Temperature Pressure Pressure drop Mass flow of refrigerant Calculated quantities Mass flux Heat flux Heat transfer coefficient Enhancement factor
Uncertainty 0.1 C 2 kPa 2.8 kPa 1% 1.09% 1.31% to 3.17% 4.19% to 8.96% 9.76% to 13.42%
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Table 3 Experimental conditions.
1200
Refrigerant
CO2 (R744)
Test section
Smooth tube
Microfin tube
Inner diameter (outer diameter) of copper tube, [mm] Mass flux, [kg/(m2s)] Vapor quality, [/] Saturation temperature, [ C] Saturation pressure, [MPa]
4.95 (6.35)
4.6 (5.0)
100 Liquid density of CO Vapro densityof CO Density ratio of CO
1000
Density [kg/m3]
400, 600, 800 0.0e1.0 20, 25, 30 5.76, 6.43, 7.21
80
Density ratio of R134a 800
60 600 40 400
Density ratio
56
20
200
C.
temperature approaches near 30 One of the most important properties is the specific heat that has an effect on the condensation heat transfer coefficient which increases with the increasing specific heat. Fig. 6 presents the liquid and vapor density, and the ratio of liquid to vapor density with saturation temperature ranging from 0 to 30 C. The density ratio of CO2 is higher than that of R134a. The low density ratio of CO2 may give more homogenous two-phase flow than with other refrigerants. The liquid to vapor density ratio plays an important role in a condenser since it determines the flow pattern and thus the heat transfer coefficient. An effect of the low vapor density is the volume flow and flow velocity which increases the heat transfer coefficients. The higher vapor density indicates the high volumetric refrigeration capacity (VCR) of CO2. Fig. 7 illustrates the surface tension with respect to saturation temperature. This decrease with temperature and become zero at the critical point. The surface tension decreases with saturation temperature, which causes the decreased heat transfer coefficient at the near critical temperature (31.1 C). As shown in Fig. 7, the surface tension of CO2 is 2.5 times smaller than that of R134a at 0 C. Low surface tension of CO2 has an effect on the forced convective heat transfer during condensation. Especially, low surface tension becomes thinner the liquid film of the upper tube wall and thicker the liquid film of the bottom tube wall, and thus the convective heat transfer to the tube wall is enhanced. Fig. 8 displays the thermal conductivities of liquid and vapor phase with varying saturation temperature. A high thermal conductivity is essential for heat transfer coefficients for single- and two-phase flow. Especially, the liquid thermal conductivity increases the heat transfer in liquid film of flow patterns such as annular flow and annular-droplet flow. And while the vapor
0 0
10
15
20
25
0 35
30
o
Tsat [ C] Fig. 6. Density and density ratios of liquid and vapor phase of CO2 and R134a with saturation temperature.
thermal conductivity increases with saturation temperature, the liquid thermal conductivity decreases and thus increases at more than 20 C. This trend has an influence on the condensation heat transfer at the near critical temperature. Fig. 9 shows the liquid and vapor viscosity, and the ratio of liquid to vapor viscosity with saturation temperature of 0e30 C. These viscosities are important factors for the fluid flow behaviors, and convective heat transfer and pressure drop characteristics during condensation. Additionally, another important parameter to describe the effect of saturation temperature is the liquid viscosity. This viscosity is higher at lower saturation temperature, which results in the decrease of heat transfer coefficients for lower saturation temperature. As mentioned above, the physical properties of CO2 are quite different from those of conventional refrigerants. In case of applying CO2 condenser to small diameter and microfin tubes, the heat transfer in these tubes can be improved. And then the condenser using high pressure fluid like CO2 can be compact. 3.2. Flow pattern Even though the CO2 flow patterns at near critical temperatures in smooth and microfin tubes are not presented in open literature,
60
14
Liquid specific heat Vapro specific heat
CO R134a
12
Surface tension [mN/m]
50
Specific heat [kJ/(kgK)]
5
40
30
20
10
0
10 8 6 4 2 0
0
5
10
15
20
25
30
o
Tsat [ C] Fig. 5. Specific heats of liquid and vapor phase with saturation temperature.
35
0
5
10
15
20
25
o
Tsat [ C] Fig. 7. Surface tension of CO2 with saturation temperature.
30
35
120
100
100
80 Liquid thermal conductivity Vapor thermal conductivity 60
40
57 10 Liquid viscosity Vapor viscosity Viscosity ratio
8
80 6 60 4 40
Viscosity ratio
120
Viscosity [106Pa s]
Thermal conductivity [mW/(mK)]
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
2
20
20 0 0
5
0 0
5
10
15
20
25
30
10
15
20
25
30
0 35
Tsat [oC]
35
o
Tsat [ C]
Fig. 9. Viscosity and viscosity ratio of liquid and vapor phase with saturation temperature.
Fig. 8. Thermal conductivity of liquid and vapor phase with saturation temperature.
3.3. Local heat transfer it is necessary to predict the flow patterns with existing flow pattern maps in order to analyze the measured condensation heat transfer coefficients. This study was applied to the representative modified flow pattern map of Taitel and Dukler [21] recommended by Jang and Hrnjak [16]. As stated earlier, they investigated experimentally for condensation flow pattern of CO2 in the 6.1 mm diameter tube. They concluded that the flow map of Taitel and Dukler [21], shown below, correctly predicts 19 out of the 25 observations. It is useful for predicting flow pattern during condensation of CO2 in horizontally smooth and microfin tube. Fig. 10(a) and (b) presents the flow pattern predictions for the experiment conditions using the flow pattern map of Taitel and Dukler [21]. The parameters for x- and y-axis in Fig. 10(a) and (b) are defined by Eqs. (10)e(13).
F ¼
rv
ðrl rv Þ
0:5 $
jv ðdi $g$cos qÞ
"
rv $j2v $jl K ¼ ðrl rv Þ$g$cos q$nl T ¼
ðdp=dzÞl ðrl rv Þ$cos q$g
X ¼
dp dz
0:5
(10)
#0:5 (11)
0:5
0:5 dp dz v l
(12)
(13)
According to this map, two flow patterns are present during condensation: stratified wavy and annular. A very limited number of points fall in the stratified wavy region, while most of the data points for CO2 lay in the annular regions. These results suggest that annular flow can be considered as an important factor in the condensation heat transfer characteristics of CO2 in the smooth- and microfin-tube. Moreover, we believe that the condensation flow mechanisms of CO2 were similar to those of non-CO2. But it is necessary to confirm accurately flow characteristics during condensation of CO2 in horizontal smooth and microfin tubes. Therefore, additional research is needed to further elucidate the flow patterns and investigate the CO2 condensation flow mechanisms in smooth and microfin tubes.
3.3.1. The effect of mass flux Fig. 11 presents the measured heat transfer coefficient of condensing CO2 with the change of vapor qualities and mass fluxes at saturation temperatures of 20 C in a smooth- and microfin-tube. As shown in Fig. 11, the condensation heat transfer coefficients in two tubes have a tendency to obviously increase with increasing vapor quality. This is because mean density decreases with the increase of vapor quality and thus liquid and vapor phase velocity increases. Also, as shown in Fig. 11, the experimental data show noticeable increase with respect to mass flux at a given saturation temperature of 20 C. This trend can be explained by the increasing Reynolds number with an increase of mass flux. With the increase of mass flux, the enhancements of the heat transfer coefficients in the microfin tube are higher than those in the smooth tube. It can be explained that the turbulent flow of CO2 near the wall in the microfin tube is increased considerably compared to that in the smooth tube. As shown in Fig. 11, the enhancement of heat transfer coefficients at high quality is obviously higher than those at low quality. This is due to the major effect of a forced convective condensation heat transfer at liquid and vapor interface and at the tube wall. These heat transfer trends are almost identical to the study results of Park and Hrnjak [17]. They presented that the heat transfer is enhanced with the increase of mass flux and vapor quality which raises liquid and vapor phase velocity in a channel. These trends shown in Fig. 11 are typical characteristics of flow condensation in an annular flow pattern which was predicted by flow pattern maps as shown in Fig. 10. We can know from above results that the condensation heat transfer characteristics of CO2 with mass fluxes in the microfin tube are almost identical to those in the smooth tube. 3.3.2. The effect of saturation temperature Fig. 12 illustrates the effect of saturation temperature on the condensation heat transfer in the smooth- and microfin-tube as a function of vapor quality. In Fig. 12, the heat transfer coefficient for two tubes decreases with increasing saturation temperature. The primary effect of saturation temperature is related to CO2 thermophysical properties. Park and Hrnjak [17] used Akers et al. [22] correlation, as represented by Eq. (14), which is simple form and is widely used, to explain this trend with thermophysical
58
a
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
104
22
102 Annular-droplets
b
18
Bubbly
101 b
K
102
100 Intermittent (Slug and Plug)
Smooth tube G=400 kg/(m2s)
10-1
T or F
d Stratified wavy
6
2 0 0.0
10-2
10-1
100
101
102
101
a
103
d
2
10
100 Intermittent (Slug and Plug)
Microfin tube 2
T or F
d b
Stratified wavy
10-1
G=400 kg/(m s) 2
G=600 kg/(m s) G=800 kg/(m2s)
101
10-2
a c 100 10-3
10-2
Stratified smooth
c
10-1
101
100
10-3 102
Χ Fig. 10. Experimental data of CO2 on the flow pattern map by Taitel and Dukler in the smooth- and microfin tube. (a) Smooth tube, (b) Microfin tube.
kl
di
1=3 $cp;l $ 7=15
1=3
$Prl
0:5 0:8 r x þ1 $ l ð1 xÞ rv
0:5 0:8 r x þ1 $ l ð1 xÞ rv
1=3 $cp;l r $ l 7=15
ml
(16)
rv
22 20 18
$Re0:8 l $
1.0
From the results stated above, it can be concluded that Akers et al. [22] correlation cannot predict the condensation heat transfer coefficient of CO2 at saturation temperature above 20 C. Zilly et al. [23] presented the similar result for the CO2 condensation heat transfer inside horizontal tubes. They explained that the effect of saturation temperature is related to thermophysical properties such as density ratio of liquid to vapor, dynamic viscosity and surface tension. The explanation for earlier two properties is almost identical to the experimental result of Park and Hrnjak [17] mentioned above. They proposed that the surface tension is another important factor to describe the effect of saturation temperature. The decrease of saturation temperature
Smooth tube T =20 [ C] T =25 [ C] T =30 [ C]
Microfin tube T =20 [ C] T =25 [ C] T =30 [ C]
16
(14)
(15)
With the decrease of saturation temperature, liquid conductivity, kl, liquid viscosity, ml, and density ratio of liquid to vapor, rl/rv, increase, and liquid specific heat, cp,l, decreases. The contribution of the increased liquid conductivity and density ratio of liquid to vapor to heat transfer is higher than the effect of viscosity and specific heat, which results in higher coefficients at lower saturation temperature. But, at saturation temperature above 20 C, the explanation of Park and Hrnjak [17] cannot be applied for the effect of saturation
hc,loc [kW/(m2K)]
0.8
2=3
kl hz
properties. The relation between the condensation heat transfer coefficients and thermophysical properties is presented by Eq. (14) which is rearranged from Akers et al. correlation.
h ¼ 0:026$
0.6
temperature. As shown in Table 4 and Fig. 13, because the liquid conductivity decreases and increases with the increase of saturation temperature. In Eq. (16), the values including four thermophysical properties decrease with increasing saturation temperature. And the condensation heat transfer coefficients of CO2 increase with saturation temperature.
b Bubbly
K
0.4
Fig. 11. Heat transfer coefficient of CO2 with mass flux and vapor quality at saturation temperature of 20 C in the smooth- and microfin tube.
102 Annular-droplets
ml
0.2
x
Χ
hz
8
c
Stratified smooth
104
2=3
10
10-3
10-3
kl
12
4
100
b
T
10-2
a c
G =400 [kg/(m s)] G =600 [kg/(m s)] G =800 [kg/(m s)]
14
G=600 kg/(m2s) G=800 kg/(m2s)
101
Microfin tube
16
d
hc,loc [kW/(m2K)]
a
103
Smooth tube G =400 [kg/(m s)] G =600 [kg/(m s)] G =800 [kg/(m s)]
20
G =800 [kg/(m s)] 14 12 10 8 6 4 2 0 0.0
0.2
0.4
0.6
0.8
1.0
x Fig. 12. Heat transfer coefficient of CO2 with saturation temperature at mass flux of 800 kg/m2s in the smooth- and microfin tube.
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
59
Table 4 Variation of thermophysical properties with saturation temperature. Tsat[ C]
kl, [mW/(mK)]
ml, [106 Pa s]
rl/rv, [/]
cp,l, [kJ/(kgK)]
s, [mN/m]
85.683 80.789 95.356
66.148 57.048 43.768
3.982 2.927 1.719
4.263 6.467 35.338
1.203 0.552 0.054
r 1=3 k2=3 $cp;l = 7=15 $ l , l ml rv [Jm2/(kgs2K2)]
20 25 30
increases surface tension which can be regarded as work per surface area of a liquid droplet and higher surface tension may increase heat transfer coefficient. More energy from the condensation process can be stored as surface energy, which increases the condensation rate. This effect would also explain higher heat transfer coefficient at lower saturation temperature, assuming the increased surface tension dries the upper tube wall and the vapor phase can condensate directly to the tube wall. It can be seen from above results that the effect of saturation temperature is related to the thermal properties of CO2 which have an effect on the condensation heat transfer coefficient in the smooth- and microfin-tube. 3.3.3. The effect of tube geometry An enhancement factor (EF) is generally used as a measure of the heat transfer enhancement. EF is defined as the ratio (EF ¼ h/hplate) of the heat transfer coefficient of an enhanced surface to that of a plain surface at the same condition. But, in this study, for investigating the increase in the heat transfer coefficient of the microfin tube to smooth tube, we uses the enhancement factor (EF), which is defined as the ratio (EF ¼ hmicrofin/hsmooth) of the heat transfer coefficient in the microfin tube to that in the smooth tube. But, since the tube diameter of the smooth tube used in this study is different to that of microfin tube, the heat transfer areas (Amicrofin and Asmooth) of the smooth- and microfin-tube are included at a denominator and numerator of Eq. (17). The enhancement factor used in this study is an average one, EFavg given by Eq. (17).
EFavg
havg $A microfin ¼ havg $A smooth
(17)
Where, (havg$A)microfin and (havg$A)smooth are the product of the average heat transfer coefficient and area in the smooth and microfin tube, respectively.
8.891 9.024 15.396
Fig. 14 presents the enhancement factors (EFavg) with mass fluxes at a given condition. It can be seen from Fig. 14 that EFavg varies from 1.006 to 1.48 with increasing mass flux. It means that the heat transfer coefficients for the microfin tube are greater than those for the smooth tube in the entire quality range. The reasons for the condensation heat transfer enhancement in the microfin tube are summarized as follows: first of all the mere increase of the effective heat exchange area. Additionally, the turbulence induced in the liquid film and the surface tension effect on the liquid drainage promotes an early transition from the wavy-stratified flow to the annular flow. [4]. Also, the heat transfer enhancement at high mass flux is higher than that at low mass flux due to the main effect of the forced convection heat transfer as mentioned earlier. 3.4. Comparison of existing heat transfer correlations The condensation heat transfer of non-CO2 has been studied by a large number of researchers in horizontal smooth and microfin tubes. But the studies of the heat transfer of CO2 during condensation inside smooth and microfin tubes have been limited in open literature until now. Determination of accurate correlation is necessary for predicting the condensation heat transfer coefficient in the horizontal smooth- and microfin-tube. Therefore, the experimental data for CO2 inside the horizontal smooth- and microfin-tube were compared with related previous correlations. The correlations of Cavallini and Zecchin [12], Dobson et al. [11], Shah [13], Cavallini et al. [14] and Kondou and Hrnjak [1] are used for the comparison of the data in the smooth tube. The correlations of Han and Lee [24], Chamra et al. [25], Cavallini et al. [4], Haraguchi et al. [6], Jung et al. [26] and Honda et al. [27] are used for the microfin tube. The compared results for condensation heat transfer coefficients in the smooth tube are shown in Table 5 and Fig. 15, where
1.8
EFavg [/]
1.6
T =20 [ C]
T =20 [ C]
T =20 [ C]
T =25 [ C]
T =25 [ C]
T =25 [ C]
T =30 [ C]
T =30 [ C]
T =30 [ C]
1.4
1.2
1.0
0.8 200
400
600
800
Gre [kg/m2s] Fig. 13. Physical properties of CO2 with respect to saturation temperature from 20 to 30 C.
Fig. 14. Enhancement factor (EF) at an entire range of mass fluxes.
1000
60
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
Table 5 The deviations between the calculated and experimental heat transfer coefficients in the smooth tube. 8 8
9 ! 9
= = n h
cal;i hexp;i n savg ¼ =N 100; sabs ¼
N 100 i ¼ 1
; ; : : hexp;i hexp;i i¼1
Author
Refrigerant
Table 6 The deviations between the calculated and experimental heat transfer coefficients in the horizontal microfin tube. 8 8
9 ! 9
= = n h
cal;i hexp;i n savg ¼ =N 100; sabs ¼
N 100 i ¼ 1
; ; : : hexp;i hexp;i i¼1
Author
Refrigerant
CO2
Cavallini and Zecchin [12] Dobson et al. [11] Shah [13] Cavallini et al. [14] Kondou and Hrnjak [1]
CO2
savg
sabs
53.2 44.5 41.7 28.8 20.7
53.2 44.5 41.7 28.8 20.7
statistical comparison of correlations, based on average and absolute deviation is given. Among these correlations listed in Table 5, Kondou and Hrnjak [1] correlation is proposed only for condensing CO2 in a horizontal smooth tube and the others are for non-CO2. All correlations proposed for non-CO2 showed enormous deviations with experimental data. The correlations of Cavallini and Zecchin [12], Dobson et al. [11] and Shah [13] failed in predicting the current experimental data with mean deviations of higher than 40%. The large deviation of these correlations is due to experimental conditions such as higher reduced pressure and mass flux, and different characteristics of CO2 such as lower liquid density, higher vapor density and lower surface tension etc. Cavallini et al. [14] correlation had an agreement to experimental data within 28.8%. The condensation latent heat and liquid thermal conductivity of CO2 at a given saturation temperature are slightly higher than those of various refrigerants used in Cavallini et al. correlation. The correlation of Kondou and Hrnjak [1] proposed for condensing CO2 is reasonably correlated to the current experimental data within 20.7%. They applied Cavallini et al. [14] equation using the approach of Fujii and Watabe [28], calculating the liquid properties at the mean temperature.
Han and Lee [24] Chamra et al. [25] Haraguchi et al. [6] Cavallini et al. [4] Jung et al. [26] Honda et al. [27]
sabs
42.7 34.5 46.2 21.1 28.8 36.7
42.7 35.5 46.2 21.1 28.8 36.7
Kondou and Hrnjak proposed the modified correlation that agrees well with the experimental data not only in subcritical region but also at lower pressure below 6.5 MPa. But they proposed their modified correlation over limited mass fluxes of Gre < 250 kg/ (m2s). And the reduced pressures in this study are much higher than the validity range of the original correlation of Cavallini et al. [14]. Thus, this correlation needs to improve for higher mass flux ranges of Gre > 250 kg/(m2s). As shown in above comparison results, it is difficult to apply existing correlations which are proposed for non-CO2 in the horizontal tubes to predict the condensation heat transfer coefficient of CO2. Thus new and reliable correlation is necessary for the optimal design of condenser using CO2. The comparison results for the heat transfer coefficients during condensation in the microfin tube are shown in Table 6 and Fig. 16. The correlations of Han and Lee [24], Chamra et al. [25], Haraguchi et al. [6], Jung et al. [26] and Honda et al. [27] were developed for nonCO2, Cavallini et al. [4] correlation were specifically developed from R22, R134a, R123, R410A and CO2 data taken inside microfin tubes. Most of the correlations proposed for non-CO2 showed large deviations with experimental data. The correlations of Han and Lee [24], Chamra et al. [25], Haraguchi et al. [6], and Honda et al. [27]
20
25
+30%
+30% 20
15
-30%
10
Cavallini and Zecchin Dobson et al.
5
-30%
hpred [kW/(m2K)]
hpred [kW/(m2K)]
savg
Shah
15
10 Han and Lee Chamra et al. Haraguchi et al. Cavallini et al. Jung et al. Honda et al.
5
Cavallini et al. Kondou and Hrnjak 0
0
0
5
10
15
20
2
hexp [kW/m K] Fig. 15. Comparison of the predicted and experimental heat transfer coefficients for CO2 condensing in the smooth tube.
0
5
10
15
20
25
hexp [kW/m2K] Fig. 16. Comparison of the predicted and experimental heat transfer coefficients for CO2 condensing in the microfin tube.
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
61
1.25
12
1.20
Penalty factor, [/]
Pressure drop, [kP/m]
10
8
6
4 Smooth tube Tre=30 [oC] Tre=25 [oC] o Tre=20 [ C]
2
Microfin tube
1.15
1.10
Microfin tube T =30 [ C] T =25 [ C] T =20 [ C]
Tre=30 [oC] o Tre=25 [ C] Tre=20 [oC] 1.05
0 400
600
400
800
failed in predicting the present experimental data with mean deviations of higher than 30%. The correlations of Cavallini et al. [4] and Jung et al. [26] averagely reasonably correlated to the current experimental data within 20%. Among six correlations, the comparison of the predicted values given by the Cavallini et al. [4] correlation to the experimental values showed the best satisfaction. For these comparison results, some researchers reported the following opinion. Zilly et al. [23] showed that the correlation of Haraguchi [6] is overpredicted, but Cavallin et al. [4] correlation is in good agreement with the experimental data. Also, Cavallini et al. [4] presented that only data by Haraguchi et al. [6] is underestimated more than 20%. Kondou and Hrnjak [1] showed that the experimental heat transfer coefficient was compared to the correlation for condensation flow of other refrigerants proposed by Cavallini et al. [4] and Haraguchi et al. [6]. These correlations were developed on experimental data obtained at much lower reduced pressures. Cavallini et al. [4] correlation showed the best agreement to the experimental heat transfer coefficient. Cavallini et al. [4] developed their correlation using the experimental CO2 data of Zilly et al. [23], measured in a limited saturation temperature of 15 to 25 C and mass flux of 200e400 kg/(m2s). Zilly et al. [23] experimental conditions are very different from those in Table 1. Therefore, it is difficult to apply Cavallini et al. [4] correlations to predict the condensation heat transfer coefficients of CO2 for high saturation temperatures in the smooth- and microfin-tube. Thus it is necessary to develop a new and reliable condensation heat transfer correlation which can be applied to the smooth and microfin tubes for various conditions. 3.5. Pressure drop The condensation pressure drop of CO2 in the smooth- and microfin-tube is shown in Fig. 17. As expected, the pressure drop in both tubes increases almost linearly as the mass flux increases. The pressure drop at 20 C is higher than at 30 C in the smooth- and microfin-tube. As represented in Figs. 6 and 9, this is because liquid viscosity increases and vapor density decreases with the decrease of saturation temperature. With the decrease of vapor density, the vapor phase velocity in a tube increases at an identical mass flux condition, which results in the increase of pressure drop. Also, the pressure drop in the microfin tube is higher than that in the smooth tube. This reason is the turbulent mixing effect and smaller tube diameter of microfin tube.
800 2
Gre, [kg/(m s)]
Gre, [kg/(m2s)] Fig. 17. Pressure drop with respect to mass flux and saturation temperature.
600
Fig. 18. The effect of mass flux and saturation temperature on penalty factor.
The pressure drop characteristics of microfin tubes were investigated in terms of the pressure drop penalty factor (PF). The PF is the ratio of pressure drop of the microfin tube and that of the smooth tube at the same operating condition. Therefore, the values of the PF in Fig. 18 are defined as follows:
PF ¼
ðDP$AÞmicrofin ðDP$AÞsmooth
(18)
Where, (DP$A)microfin and (DP$A)smooth are the product of the pressure drop and heat transfer area in the smooth and microfin tube, respectively. The PF in the present study varies from 1.14 to 1.23 with increasing mass flux as shown in Fig. 18. It means that the pressure drop of microfin tube is higher than that of smooth tube. The reason is the increased turbulent mixing and swirl effects of the microfin tube. As a result, although the pressure drop of CO2 in the microfin tube is higher than that in the smooth tube, the pressure drop of CO2 is much lower than conventional refrigerants such as R410a and R134a at an identical test conditions due to the significantly higher vapor density of CO2. This CO2 pressure drop characteristics are valid for microfin tubes [17].
4. Conclusions In this study, the condensation heat transfer experiments of CO2 were carried out in the horizontal smooth- and microfin-tube at mass fluxes from 400 to 800 kg/(m2s), saturation temperatures from 20 to 30 C, and vapor qualities from 0.1 to 0.9. The condensation heat transfer and pressure drop data in the horizontal smooth- and microfin-tube were measured and their applicability to the various condensation heat transfer coefficient models under different operating conditions was examined. This study performed tests for the validation of models with different operating conditions. The experimental results showed that annular flow almost dominates the major of condensation flow in the smooth- and microfin-tube and it can be considered importantly for heat transfer and pressure drop characteristics as well as new and reliable condensation heat transfer correlations in the smooth- and microfin-tube. The heat transfer coefficients in the smooth- and microfin-tube have similar trends to increase with respect to increased mass flux and decreasing saturation temperature. The average heat transfer
62
C.-H. Son, H.-K. Oh / Applied Thermal Engineering 36 (2012) 51e62
enhancement factor (EF) for the microfin tube is approximately from 1.006 to 1.48, and penalty factor (PF) varies from 1.14 to 1.23. Thus, the use of the microfin tube will be prosperous for the condensers using CO2 such as working fluid, if the fin geometry such as the fin shape and height can be optimized for the better performance. Most of the proposed correlations for non-CO2 failed to predict the condensation heat transfer coefficients in the smooth- and microfin-tube. However, for the overall experimental data in the microfin tube, the Cavallini et al. correlation predicted the present experimental data better than the other correlations. The correlation by Kondou and Hrnjak showed relatively good agreement with experimental data in the smooth tube. As a result, there are little researches for the condensation heat transfer and pressure drop in the horizontal smooth- and microfintube. Thus, the accurate condensation heat transfer and pressure drop data under various test conditions is necessary for the optimal design of CO2 condenser using the smooth- and microfin-tube. Furthermore, it is necessary to develop more accurate and reliable correlation to predict heat transfer characteristics during condensation in the smooth- and microfin-tube. Therefore, the dominant annular-flow pattern during CO2 condensation must be considered importantly for new correlations which can be also applied to higher saturation temperature and mass flux conditions. These correlations based on the annular flow model for the smooth- and microfin-tube will give better prediction accuracy. Acknowledgements The work presented in this paper is part of the project ‘Development of high efficient cooling and heating system using natural refrigerant of CO20 sponsored by Ministry of Commerce, Industry, and Energy. The support of these sponsors is gratefully acknowledged. References [1] C. Kondou, P. Hrnjak, Heat rejection from R744 flow under uniform temperature cooling in a horizontal smooth tube around the critical point, Int. J. Refrigeration 34 (2011) 719e731. [2] J.M. Cho, M.S. Kim, Experimental studies on the evaporative heat transfer and pressure drop of CO2 in smooth and micro-fin tubes of the diameter of 5 and 9.52 mm, Int. J. Refrigeration 30 (2007) 1e9. [3] Y.J. Kim, J. Jang, P. Hrnjak, M.S. Kim, An experimental study on condensation heat transfer of CO2 at low temperature in smooth and micro-fin tubes, in: Proceedings of SAREK (2003), pp. 50e55. [4] A. Cavallini, D. Del Col, S. Mancin, L. Rossetto, Condensation of pure and nearazeotropic refrigerants in microfin tubes: a new computational procedure, Int. J. Refrigeration 32 (2009) 162e174. [5] J. Yu, S. Koyama, Condensation heat transfer of pure refrigerants in microfin tubes, in: Proceeding of International Refrigeration Conference at Purdue. 1720 July, West Lafayette, Indiana, USA (1998), pp. 325e330. [6] H. Haraguchi, S. Koyama, J. Esaki, T. Fujii, Condensation heat transfer of refrigerants HCFC134a, HCFC123, and HCFC22 in a horizontal smooth tube
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19] [20]
[21]
[22] [23]
[24]
[25]
[26]
[27] [28]
and a horizontal microfin tube, in: Proceeding of 30th National Symposium of Japan, Yokohama, Japan (1993), pp. 343e345. M.A. Kedzierski, J.M. Goncalves, Horizontal convective condensation of alternative refrigerants within a micro-fin tube NISTIR 6095, U.S. Dept. Commerce, 1997. M.H. Kim, J.S. Shin, B.H. Lim, Condensation heat transfer of R22 and R410A in horizontal smooth and microfin tubes, zero Leakage-minimum charge, IIR-IIF, Stockholm, Sweden, 2002, Section B and E253e261. S. Koyama, R. Yonemoto, Experimental study on condensation of pure refrigerants in horizontal micro-fin tube - proposal of correlation for heat transfer and frictional pressure drop, in: Proceeding of International Refrigeration Conference at Purdue. 17e20 July, West Lafayette, Indiana, USA (2006). H.S. Wang, J.J. Wei, H. Honda, J.W. Rose, Condensation of New Refrigerants in Horizontal Microfin Tubes: a new correlation, in: Proceeding of International Congress Refrigeration, ICR07, Beijing, China (2007). M.K. Dobson, J.C. Chato, D.K. Hinde, S.P. Wang, Experimental evaporation and condensation of Refrigerant R-12 and R-134a, ASHRAE Trans. 100 (1994) 744e754. A. Cavallini, R. Zecchin, A Dimensionless correlation for heat transfer in forced convection condensation, in: , Proc, 6th Intl. Heat Transfer Conf. Tokyo, vol. 3, 1974, pp. 309e313. M.M. Shah, A general correlation for heat transfer during film condensation inside pipes, Int. J. Heat Mass Transfer 22 (1979) 547e556. A. Cavallini, D. Del Col, L. Doretti, M. Matkovic, L. Rossetto, C. Zilio, G. Censi, Condensation in horizontal smooth tubes: a new heat transfer model for heat exchanger design, Heat Trans. Eng. 27 (2006) 31e38. S. Koyama, C. Kondou, K. Kuwahara, An experimental study on condensation of CO2 in a horizontal micro-fin tube, 12th Int. Refrig. A/C Conf. at Purdue, Paper No. 2322, 2008. J. Jang, P.S. Hrnjak, Condensation of CO2 at Low Temperature inside Horizontal Microfinned Tubes, ACRC CR-56, University of Illinois at UrbanaChampaign, 2004. C.Y. Park, P. Hrnjak, R744 flow condensation heat transfer and pressure drop in multi-port microchannels at low temperatures, Int. journal Refrigeration 32 (2009) 1129e1139. N. Hayes, Amir Jokar, Zabid H. Ayub, Study of carbon dioxide condensation in chevron plate exchangers; heat transfer analysis, Int. J. Heat Mass Transfer 54 (2011) 1121e1131. V. Gnielinski, New equations for heat and mass transfer in turbulent pipe and channel flow, Int. Chem. Eng. 16 (1976) 359e368. E.W. Lemmon, M.L. Huber, M.O. McLinden, Reference Fluid Thermodynamic and Transport Properties, NIST Standard Reference Database 23 Version 8.0 (2007). Y. Taitel, A.E. Dukler, A model for predicting flow Regime transitions in horizontal and near horizontal gas-liquid flow, Am. Inst. Chem. Eng. J. 22 (1976) 47e55. W.W. Akers, H.A. Deans, O.K. Crosser, Condensing heat transfer within horizontal tubes, Chem. Eng. Prog. Symp. Ser. 55 (1959) 171e176. J. Zilly, J. Jang, P.S. Hrnjak, Condensation of CO2 at Low Temperature inside Horizontal Microfined Tubes, ACRA (Air Conditioning and Refrigeration Center) Report, March 2003, University of Illinois at Urbana-Campaign, 2003. D. Han, K.J. Lee, Experimental study on condensation heat transfer enhancement and pressure drop penalty factors in four microfin tubes, Int. J. Heat Mass Tran. 48 (2005) 3804e3816. L.M. Chamra, P.J. Mago, M.O. Tan, C.C. Kung, Modeling of condensation heat transfer of pure refrigerants in micro-fin tubes, Int. J. Heat Mass Tran. 48 (2005) 1293e1302. D.S. Jung, Y.M. Cho, K.H. Park, Flow condensation heat transfer coefficients of R22, R134a, R407C, and R410A inside plain and microfin tubes, Int. J. Refrigeration 27 (2004) 25e32. H. Honda, A.T. Wijayanta, N. Takata, Condensation of R407C in a horizontal microfin tube, Int. J. Refrigeration 28 (2005) 203e211. T. Fujii, M. Watabe, Laminar film condensation in the subcritical region e Gravity controlled condensation, JSME Trans. B. 53 (1987) 1801e1806 (in Japanese).