Experimental investigation into the thermal-flow performance characteristics of an evaporative cooler

Applied Thermal Engineering 30 (2010) 492–498

Contents lists available at ScienceDirect

Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng

Experimental investigation into the thermal-flow performance characteristics of an evaporative cooler J.A. Heyns *, D.G. Kröger 1 Department of Mechanical Engineering, University of Stellenbosch, Stellenbosch 7600, South Africa

a r t i c l e

i n f o

Article history: Received 20 January 2009 Accepted 14 October 2009 Available online 20 October 2009 Keywords: Evaporative cooling Heat and mass transfer

a b s t r a c t This study investigates the thermal-flow performance characteristics of an evaporative cooler. The derivation of the Poppe [1] and Merkel [2] analysis for evaporative coolers are presented and discussed. Performance tests were conducted on an evaporative cooler consisting of 15 tube rows with 38.1 mm outer diameter galvanized steel tubes arranged in a 76.2 mm triangular pattern. From the experimental results, correlations for the water film heat transfer coefficient, air–water mass transfer coefficient and air-side pressure drop are developed. The experimental tests show that the water film heat transfer coefficient is a function of the air mass velocity, deluge water mass velocity as well as the deluge water temperature, while the air–water mass transfer coefficient is a function of the air mass velocity and the deluge water mass velocity. It was found that the correlations obtained for the water film heat transfer coefficient and the air–water mass transfer coefficient compare well with the correlations given by Mizushina et al. [3]. Relatively little published information is available for predicting the air-side pressure drop across deluged tube bundles. The present study shows that the pressure drop across the bundle is a function of the air mass velocity and the deluge water mass velocity. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction Various authors have investigated the performance characteristics of evaporative coolers and condensers. It is however found that there are measurable differences in the empirical correlations suggested by these authors for determining the film heat transfer coefficient, hw, and the air–water mass transfer coefficient, hd. It is further found that relatively little information is available for determining the air-side pressure drop across deluged tube bundles. At this stage, there is no general accepted methodology for the analysis of evaporative coolers and condensers. Inconsistencies in the analyses result in discrepancies between the empirical correlations. Furthermore, most of the correlations recommended in literature are facility specific, where the tube diameter and spacing as well as the number of tube rows are fixed. It would be inaccurate to use these correlations in the analysis of a system with different parameters. This work forms part of ongoing research on hybrid (dry/wet) cooling for industrial application. The proposed system incorporates triangular spaced tubes with an outside diameter of 38.1 mm. As there is uncertainty over the accuracy of existing cor-

* Corresponding author. Address: P.O. Box 36613, Menlopark 0102, South Africa. Tel.: +27 82 611 9963; fax: +27 12 361 7860. E-mail addresses: [email protected] (J.A. Heyns), [email protected] (D.G. Kröger). 1 Address: Private Bag X1, 7602 Matieland, South Africa. 1359-4311/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.applthermaleng.2009.10.010

relations for the analysis of the proposed system, it was deemed necessary to evaluate the thermal-flow performance characteristics experimentally. In this study the methodology for the analysis of evaporative coolers is evaluated, the governing equations are derived and the restrictions associated with the different methods are discussed. In Section 5, experimental results of the water film heat transfer coefficient, the air–water mass transfer coefficient as well as the pressure drop over a deluged tube bundle are presented and discussed. 2. Literature review The first practical design procedure for the evaluation of counterflow evaporative coolers was given by Parker and Treybal [4]. Their model makes use of the Lewis factor to find the relationship between the heat and mass transfer at the air–water interface and assumes that the Lewis factor is equal to unity. They further assumed that the amount of water evaporated is negligibly small and that the saturated air enthalpy is a linear function of the temperature. This makes it possible to integrate the differential equations simultaneously over the height of the tube bundle. In their work, Parker and Treybal [4] tested tube bundles consisting of 19 mm outside diameter tubes arranged on a triangular pitch. Mizushina et al. [3] experimentally investigated the characteristics of evaporative coolers and determined the applicable heat and

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

493

Nomenclature A cp d G h hd i ifg k L m NTU n P p Q T U w

area (m2) specific heat at constant pressure (J/(kg K)) diameter (m) mass velocity (kg/(s m2)) heat transfer coefficient (W/(m2 K)) mass transfer coefficient (kg/(m2 s)) enthalpy (J/kg) latent heat (J/kg) thermal conductivity (W/(m K)) length (m) mass flow rate (kg/s) number of transfer units number pitch (m) pressure (N/m2) heat transfer rate (W) temperature (°C or K) overall heat transfer coefficient (W/(m2 K)) humidity ratio (kg water vapor/kg dry air)

Greek letters C water flow rate per unit length (kg/(s m)) l dynamic viscosity (kg/(m s))

mass transfer coefficients. A similar approach to that of Parker and Treybal [4] is followed, but the differential equations are integrated numerically. Mizushina et al. [3] conducted their tests on tube bundles with diameters varying between 12 and 40 mm. Niitsu et al. [5] conducted testes on bundles of both plain and finned tubes. They studied the film heat transfer coefficient and mass transfer coefficient, along with the air-side pressure loss over the deluged tube bundle. The plain tubes they tested had an outside diameter of 16 mm. Dreyer [6] conducted an extensive study on evaporative coolers and condensers. He considered a detailed one-dimensional analytical model, similar to the one suggested by Poppe and Rögener [1] that describes the physics of the heat and mass transfer processes, along with a simplified model, utilizing the assumptions made in a Merkel-type analysis. Dreyer [6] further investigated the heat and mass transfer correlations suggested by various authors in the literature and compared them graphically. Zalewski and Gryglaszeski [7] developed a mathematical model which is similar to the Poppe method described by Dreyer [6]. They suggested the use of correlations given by Tovaras et al. [8] for calculating the heat transfer coefficient between the tube and the deluge water and adapted data given by Grimison [9] for the heat transfer over dry tube banks to determine the convective heat transfer coefficient from the deluge water to the moist air. In view of the difference between their theoretical prediction and their experimental results, they modified the mass transfer coefficient correlation by introducing a correction factor. Further experimental work was conducted by Ettouney et al. [10] and Hasan and Siren [11]. Ettouney et al. [10] compared the performance of an evaporative condenser with the performance of the same system when it is operated as an air-cooled condenser. They showed that the thermal performance of the evaporative condenser is up to 60% higher compared to when the unit is air-cooled. Hasan and Siren [11] did a comparative study between plain and finned tube evaporative coolers, comparing the coolers based on the amount of heat rejected to the air-side pressure drop per unit length. More recently Stabat and Marchio [12] and Ren and Yang [13] investigated the use of an analytical model based on the effective-

m q

kinematic viscosity (m2/s) density (kg/m3)

Subscripts a air, or based on air-side area mixture of dry air and water vapour av b bundle c convective heat transfer d diameter i inlet, or inside int air–water interface l longitudinal, or lateral m mean, mass transfer, or mixture o outlet, or outside p constant pressure, or process water r row s saturated t total, tube, transversal tr tube rows, or tubes per row v vapour w water, or wall wb wet bulb

ness NTU-method to evaluate the performance characteristics of an evaporative cooler.

3. Analysis In an evaporative cooler, water (process water) is cooled inside the tubes, while deluge water is sprayed over the bundle of staggered horizontal plain tubes. In a process of non-adiabatic heat and mass transfer, the deluge water evaporates into the air passing through the bundle. In the present analysis of an evaporative cooler, the following initial assumptions are made:  It is a steady state process.  Since the temperature differences between the air and deluge water are relatively small, heat transfer by radiation is neglected.  If the tube surfaces are uniformly wetted, the air flow and thermal states are uniformly distributed at the inlet and uniformity is maintained throughout the bundle, the problem can be analyzed in one dimension.  If it is assumed that the re-circulating deluge water circuit is insulated from the surroundings and that pump work can be neglected, the temperature of the deluge water at the inlet and outlet of the tube bundle is the same.  At the air–water interface surface, the air temperature approaches the temperature of the deluge water and the humidity of the air at the interface corresponds to that of a saturated air–vapour mixture. By employing these assumptions and following an approach similar to Poppe and Rögener [1] and Dreyer [6], an analytical model of the evaporative cooler can be derived from basic principles. Consider an elementary control volume about a tube as shown in Fig. 1. Due to the one-dimensional characteristic of the unit, the properties of the air and water at any horizontal crosssection are assumed to be constant. Evaporation of the downward flowing water occurs at the air–water interface.

494

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

The convective transfer of sensible heat at the interface is given

Tw

ia + dia

mw

ma (1 + w + dw )

by

dQ c ¼ hðT w  T a ÞdA

ms

ð10Þ

Substituting Eqs (8) and (10) into Eq. (6), find the total enthalpy transfer at the air–water interface, i.e.

is + dis

dQ ¼ iv hd ðwsw  wÞdA þ hðT w  T a ÞdA

The enthalpy of the saturated air at the air–water interface evaluated at the local bulk water film temperature is

ms is

imasw ¼ cpa T w þ wsw ðifgwo þ cpv T w Þ ¼ cpa T w þ wsw iv

dA

Tw + dTw

ia

which may be written as

mw + dmw

ma (1 + w )

imasw ¼ cpa T w þ wiv þ ðwsw  wÞiv

The mass balance applicable to the control volume is

ð1Þ or

ð2Þ

where ma is the mass flow rate of the dry air. The energy balance over the control volume gives

þ ma ðima þ dima Þ þ mp cpp ðT p þ dT p Þ ð3Þ where the deluge water temperature Tw and the process water temperature Tp is in °C. Neglect the second order terms and simplify Eq. (3) to

ð4Þ

where ima refers to the enthalpy of the air–vapor mixture per unit mass of dry air, which can be expressed as

ima ¼ cpa T a þ wðifgwo þ cpv T a Þ

ð5Þ

The latent heat, ifgwo, is evaluated at 0 °C and the specific heats, cpv and cpa at Ta/2 °C. If the moist air is un-saturated, the total enthalpy transfer at the air–water interface consists of an enthalpy transfer due to the difference in vapour concentration and the difference in temperature,

dQ ¼ dQ m þ dQ c

dmw ¼ hd ðwsw  wÞdA

ð7Þ

where wsw is the saturated humidity ratio of the air evaluated at the bulk water film temperature. The corresponding enthalpy transfer at the air–water interface due to the difference in the vapour concentration is then

dQ m ¼ iv dmw ¼ iv hd ðwsw  wÞdA

ð8Þ

The enthalpy of the water vapour, iv, calculated at the local bulk water film temperature, is given by

iv ¼ ifgwo þ cpv T w where Tw is in °C and cpv is evaluated at Tw/2 °C.

T w  T a ¼ ½ðimasw  ima Þ  ðwsw  wÞiv =cpam

ð9Þ

ð15Þ

where cpam = cpa + wcpv. Substitute Eq. (15) into Eq. (11) and find



h

cpma hd

 ðimasw  ima Þ þ 1 

h cpma hd

  iv ðwsw  wÞ dA

ð16Þ

Noting that the enthalpy transfer must be equal to the enthalpy change of the moist air stream

1 dQ ma     hd h h iv ðwsw  wÞ dA ðimasw  ima Þ þ 1  ¼ cpma hd ma cpma hd

dia ¼

ð17Þ

The heat transfer from the process water to the deluge water is given by

dQ ¼ UðT p  T w ÞdA

ð18Þ

where U is the overall heat transfer coefficient between the process water inside the tubes and the deluge water on the outside.

 U¼

1 do lnðdo =di Þ do þ þ hw 2kt d i hp

 ð19Þ

The change in the temperature of the process water can be written as

ð6Þ

where the subscripts m and c refer to the enthalpies associated with the mass transfer and convective heat transfer. The mass flow rate of the deluge water evaporating into the air stream is expressed as

ð14Þ

or

dQ ¼ hd

ma ima þ mw cpw T w þ mp cpp T p ¼ ðmw þ dmw Þcpw ðT w þ dT w Þ

1 ðma dima  cpw T w dmw  mp dT p Þ mw cpw

ð13Þ

imasw  ima  ðcpa þ wcpv ÞðT w  T a Þ þ ðwsw  wÞiv

ma ð1 þ wÞ þ mw þ mp ¼ ma ½1 þ ðw þ dwÞ þ ðmw þ dmw Þ þ mp

dmw ¼ ma dw

ð12Þ

Subtracting Eq. (5) from Eq. (13) and ignoring the small difference in the specific heats, the equation can be simplified as follows:

Fig. 1. Control volume for the evaporative cooler.

dT w ¼

ð11Þ

dT p ¼ 

dQ mp cpp

ð20Þ

Substituting Eq. (18) into Eq. (20) yields

dT p ¼

U ðT p  T w ÞdA mp cpp

ð21Þ

For the case where the moist air is not saturated, Eqs. (2), (4), (7), (17), and (21) describe the processes that take place in the control volume of the evaporative cooler. The model can be simplified by making use of the assumptions of a Merkel-type analysis: Firstly, it is assumed that the amount of deluge water that evaporates is small compared to the mass flow rate of the deluge water and secondly the Lewis factor, which gives the relation between the heat and mass transfer, is equal to unity. The Lewis factor can be expressed as Lef = h/(cpmahd). For the Merkel-type analysis the governing Eqs. (4), (17), and (21) become:

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

dT w ¼ ðma dima þ mp cpp dT p Þ=mw cpw dia ¼

hd ðimasw  ima ÞdA ma

dT p ¼

U ðT p  T w ÞdA mp cpp

ð22Þ ð23Þ

ð24Þ

If the evaporative cooler is evaluated using the iterative stepwise Merkel-type analysis, it is found that the three governing equations must describe four unknown parameters. The Merkeltype analysis is often used in evaluating the thermal performance characteristics of fills or packs, in wet-cooling towers. In the analysis of wet-cooling towers, the Merkel integral is numerically integrated over the deluge water temperature, Tw (using for example the four-point Chebyshev integration technique) [14]. If the inlet and outlet deluge water temperature of evaporative heat exchanger is the same, then the solution of the numerical integral is trivial. Kröger [15] suggests the use of the simplified Merkel-type analysis, where a constant mean deluge water temperature is assumed and only the inlet and outlet values of the parameters are evaluated. It is possible to solve the simplified Merkel-type analysis analytically if the assumption of Merkel is made that the outlet air is saturated. For the simplified Merkel-type analysis the governing Eqs. (23) and (24) are unchanged. Assuming a constant mean deluge water temperature through the cooler, integrate Eq. (24) between the inlet and outlet conditions and find

T po ¼ T wm þ ðT pi  T wm ÞeUA=mp cpp ¼ T wm þ ðT pi  T wm ÞeNTUp

ð25Þ

where Twm is the mean deluge water temperature and NTUp = UA/ mpcpp. Similarly for Eq. (23)

iao ¼ imaswm  ðimaswm  iai ÞeNTUa

ð26Þ

where NTU a ¼ mhda Aa and Aa is the air–water interface area. The heat transfer rate of the evaporative cooler is given by the following equation:

Q ¼ ma ðiao  iai Þ ¼ mp cpp ðT pi  T po Þ

ð27Þ

Substitute Eqs. (25) and (26) into Eq. (27) and simplify to find

T wm ¼ T pi 

ma ðimaswm  imai Þ½1  expðNTU a Þ mp cpp ½1  expðNTU p Þ

ð28Þ

The heat transfer coefficient at the tube–water interface, hw, and air–water mass transfer coefficient, hd, are obtained experimentally under different operating conditions, employing the governing Eqs. (25) and (26) as well as Eqs. (19), (27), and (28). Although the simplified Merkel analysis does not predict the amount of water lost by evaporation as accurately as the Poppe analysis, it will predict the heat rejection rate of an evaporative cooler correctly if the abovementioned experimentally obtained transfer coefficients are used in the analysis. 4. Apparatus A schematic layout of the apparatus and the placement of the measurement equipment is shown in Fig. 2. The tube bundle consists of nr = 15 rows of externally galvanized steel tubes. The tubes are L = 0.65 m long and are arranged in a triangular pattern at a transversal pitch of Pt = 76.2 mm as shown in Fig. 3. The outside diameter of the tubes is do = 38.1 mm and the inside diameter is di = 34.9 mm. There are ntr = 8 tubes per tube row.

495

To ensure uniform flow of the air through the tube bundle, inactive half tubes are installed at the sides of the tube bundle. In the spray frame, a header distributes or divides the deluge water into several conduits or lateral branches. Each lateral branch consists of a two perforated stainless steel tubes, one placed inside the other. With this configuration it is possible to establish a uniform pressure distribution in the lateral branches (perforated tubes) and achieve a uniform water distribution. The mass flow rate of the deluge water is measured using a thin-plate orifice. The deluge water temperature is measured in the collecting basin and also before it enters the spray frame. The deluge water temperatures through tube bundle are measured with thermocouples placed after each tube row; the mean deluge water temperature, Twm, is taken as the average of these thermocouple readings. The inlet temperature of the warm process water, Tpi, is measured as it enters the top inlet header. After flowing through the tubes the process water exits the bundle at the bottom outlet header, where its temperature, Tpo, is measured. Process water mass flow rate, mp, is obtained at the start of each set of tests by means of the displacement method. The process water is heated in a 72 kW geyser. The test section is connected to the inlet of an atmospheric open-loop induced draft wind tunnel, drawing air over the tube bundle. By measuring the pressure drop over an elliptical flow nozzle located in the wind tunnel the air mass flow rate can be determined. The pressure drop over the tube bundle is measured with the aid of a differential pressure transducer. Dry bulb and wet bulb temperatures of the ambient air, are measured at the inlet of the tube bundle, while the atmospheric pressure is read from a mercury column barometer. 5. Results and observations To ensure a good deluge water distribution Niitsu et al. [5] recommends that the water loading should not be
496

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

Fig. 2. Schematic layout of the apparatus.

ntr = 8

Pt = 76.2 mm

L = 0.65 m

m .2 m 76

nr = 15

d o = 38.1 mm Fig. 3. Tube bundle layout and tube dimensions.

include the effect of the deluge water temperature. The present experimental results show that the deluge water mass flow rate has the greatest influence on the film heat transfer coefficient, hw, but this coefficient is also a function of the air mass flow rate and the deluge water temperature as given by Eq. (29). Parker and Treybal [4] state that hw increases linearly with the deluge 0:3 , as given in water temperature. This is well approximated by T w Eq. (29), over the range tested. The experimental results of the film

heat transfer coefficient as a function of the deluge water mass velocity is shown in Fig. 5. From the experimental results the following correlation for the film heat transfer coefficient are derived, 0:35 0:3 hw ¼ 470 G0:1 a Gw T w

ð29Þ

for 0.7 < Ga < 3.6 kg/(m2 s), 1.8 < Gw < 4.7 kg/(m2 s) and 35 < Twm < 53 °C.

497

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

over the staggered tubes is similar to the flow of cooling water through fills and packs used in wet-cooling towers. The mass transfer coefficient for fills or packs is typically given in terms of the air and the water flow rate. Mizushina et al. [3] gives the mass transfer coefficient in terms of an air and deluge water Reynolds numbers. From the present experimental results it follows that the air–water mass transfer coefficient is a function of the air mass velocity and the deluge water mass velocity as given by Eq. (30). The experimental results of the air-mass transfer coefficient as a function of the air mass velocity are shown in Fig. 7. From the experimental results the following correlation for the air–water mass transfer coefficient are derived,

Inlet

Ga = 2.3 kg/m2s, Gw = 3.4 kg/m2s Ga = 1.2 kg/m2s, Gw = 1.9 kg/m2s

Out let

Mean deluge water temperaure, Twm

Ta = 18 oC , Twb = 15oC pa = 101 000 N/m2 25

30

35

40

45

50

Deluge water temperature, Tw , oC

1000

Ta = 16 - 22 oC , Twb = 14 - 18 oC pa = 100 000 - 101 400 N/m2

800 700 600 500

0.12

400

––– Equation (29)

300

* Experimental data

200 100

---

10 %

0 0

1

2

3

4

Ta = 16 - 22 oC , Twb = 14 - 18 oC pa = 100 000 - 101 400 N/m2

0.1

5

Deluge water mass velocity, Gw , kg/sm2 Fig. 5. Experimental results of water film heat transfer coefficient.

hde / Gw0.2 = 0.038 Ga0.73

hw / Ga0.1 Twm0.3 = 470 Gw0.35

ð30Þ

for 0.7 < Ga < 3.6 kg/(m2 s) and 1.8 < Gw < 4.7 kg/(m2 s). In Fig. 8, Eq. (30) is compared to the correlation given by Mizushina et al. [3]. Relatively little published information is available for predicting the pressure drop across deluged tube bundles. The correlation of Niitsu et al. [5] for the air-side pressure drop over a deluged tube bundle with 16 mm diameter tubes is given in terms of the air flow rate as well as the deluge water flow rate. From the present experimental results it follows that the air-side pressure drop over the tube bundle is a function of the air mass velocity and the deluge

Fig. 4. Variation in the deluge water temperature.

900

hd ¼ 0:038 G0:73 G0:2 a w

0.08 0.06

––– Equation (30)

0.04

*

Experimental data

0.02

---

In Fig. 6, Eq. (29) is compared to the correlation given by Mizushina et al. [3]. The correlation compares well at an air mass velocity of Ga = 3.2 kg/(m2 s) (as stated earlier the correlation given by Mizushina et al. [3] is not dependent on the air mass velocity). The correlations recommended by Parker and Treybal [4] and Niitsu et al. [5] for the air–water mass transfer coefficient, are only a function of the air mass velocity. The flow of the deluge water

10 %

0 0

1

2

3

4

Air mass velocity, Ga , kg/sm2 Fig. 7. Experimental results of air–water mass transfer coefficient.

0.16

Gw30, = 1.7 Eq Gw kg/m2s = 1.7 kg/m2s

Ta = 15.6oC, Twb = 10oC, pa = 101000 N/m2, Twm = 45oC, do = 38.1 mm

3500

Mass transfer coefficient, hd , kg/m2s

Film heat transfer coefficient, hw , W/m2K

4000

3000 2500 2000 1500 1000

Mizushina (1967) 2 Eq =29, Ga 0.8Gkg/m2s a = 0.8 kg/(m s)

500

Ga 2 kg/m2s Eq =29, Ga = 2.0 2 Ga 3.2Gkg/m2s Eq =29, a = 3.2 kg/(m s) kg/(m2s)

0.14 2 Eq Gw30, = 3Gkg/m2s w = 3.0 kg/m s

0.12

Gw30, = 4.5 Eq Gwkg/m2s = 4.5 kg/m2s

0.1

Mizushina (1967) (G kg/m2s) (Gw 3 kg/m2s) w ==3.0

0.08 0.06 0.04

Ta = 15.6oC, Twb = 10oC, pa = 101000 N/m2, Twm = 45oC, do = 38.1 mm

0.02 0

0 0

2

4

6

Deluge water mass velocity, Gw , kg/m2s Fig. 6. Water film heat transfer coefficient correlation.

8

0

0.5

1

1.5

2

2.5

3

3.5

Air mass velocity, Ga , kg/m2s Fig. 8. Air–water interface mass transfer coefficient correlation.

4

498

J.A. Heyns, D.G. Kröger / Applied Thermal Engineering 30 (2010) 492–498

measured upstream of the elliptical nozzles in the wind tunnel. Although there is some uncertainty in the measurements, it is found that the measured outlet air temperature for most tests is approximately 2 °C higher than the predicted saturation temperature. The outlet air is thus un-saturated and the humidity ratio lower than predicted. The amount of water evaporated can be approximated by mw(evap) = ma (wso  wi).

120

––– Equation (31)

p / Gw0.22 = 10.2 Ga1.8

100

*

80

Experimental data

---

10 %

60

6. Conclusion

40 20

Ta = 16 - 22 oC , Twb = 14 - 18 oC pa = 100 000 - 101 400 N/m2

0 0

0.5

1

1.5

2

2.5

3

3.5

4

Air mass velocity, Ga , kg/sm2 Fig. 9. Experimental results of pressure drop over deluged tube bundle.

Two-phase pressure drop, Δp, N/m2

160 140

Eq 31, Gw = 1.7 kg/m2s

120

Eq 31, Gw = 3.0 kg/m2s

100

Eq 31, Gw = 4.5 kg/m2s

Performance tests were conducted on an evaporative cooler consisting of 15 tubes rows with 38.1 mm outer diameter galvanized steel tubes arranged in a 76.2 mm triangular pattern. Correlations for the water film heat transfer coefficient, the air–water mass transfer coefficient and the air-side pressure drop are developed from experimental results. The present experimental results show that the film heat transfer coefficient (Eq. (29)) is a function of the air mass velocity, deluge water velocity as well as the deluge water temperature, while the air–water mass transfer coefficient (Eq. (30)) and the air-side pressure drop (Eq. (31)) is a function of the air mass velocity and the deluge water mass velocity. The correlations for the water film heat transfer coefficient and the air–water mass transfer coefficient compare well with the correlations recommended by Mizushina et al. [3].

80

References

60

[1] M. Poppe, H. Rögener, Evaporative Cooling Systems, VDI-Wärmeatlas, Section Mh. [2] F. Merkel, Verdunstungskühlung VDI Zeitschrift 70 (70) (1925) 123–128. [3] T. Mizushina, R. Ito, H. Miyasita, Experimental study of an evaporative cooler, International Chemical Engineering 7 (4) (1967) 727–732. [4] R.O. Parker, R.E. Treybal, Heat, mass transfer characteristics of evaporative coolers, AIChE Chemical Engineering Progress Symposium Series 57 (32) (1961) 138–149. [5] Y. Niitsu, K. Naito, T. Anzai, Studies on characteristics and design procedure of evaporative coolers, Journal of SHASE 43, Japan (1969). [6] A.A. Dreyer, Analysis of Evaporative Coolers and Condensers, MSc Thesis, University of Stellenbosch, Republic of South Africa, 1988. [7] W. Zalewski, P.A. Gryglaszeski, Mathematical model of heat and mass transfer processes in evaporative fluid coolers, Chemical Engineering and Processing 36 (1997) 271–280. [8] N.V. Tovaras, A.V. Bykov, A.V. Gogolin, Heat exchange at film water flow under operating conditions of evaporative condenser, Holod Teh 1 (1984) 25–29. [9] E.D. Grimson, Correlation and utilization of new data on flow resistance and heat transfer in cross flow of gasses over tube bank, Transactions of the American Society of Mechanical Engineers 59 (1937) 583–594. [10] H.M. Ettouney, H.T. El-Dessouky, W. Bouhamra, B. Al-Azmi, Performance of evaporative condensers, Heat Transfer Engineering 22 (2001) 41–55. [11] A. Hasan, K. Siren, Performance investigation of plain and finned tube evaporatively cooled heat exchangers, Applied Thermal Engineering 23 (2002) 325–340. [12] P. Stabat, D. Marchio, Simplified model for indirect-contact evaporative cooling-tower behaviour, Applied Energy 78 (2004) 433–451. [13] C. Ren, H. Yang, An analytical model for the heat an mass transfer processes in an indirect evaporative cooling with parallel/counter flow configurations, International Journal of Heat and Mass Transfer 49 (2005) 617–627. [14] J.C. Kloppers, D.G. Kröger, A critical investigation into the heat and mass transfer analysis of counterflow wet-cooling towers, International Journal of Heat and Mass Transfer 48 (2005) 765–777. [15] D.G. Kröger, Air-cooled Heat Exchangers and Cooling Towers, PennWell Corporation, Tulsa, Oklahoma, USA, 2004.

40

Ta = 15.6 oC, Twb = 10 oC, pa = 101000 N/m2 Twm = 45 oC, do = 38.1 mm

20 0

0

0.5

1

1.5

2

2.5

3

3.5

4

Air mass velocity, Ga , kg/m2s Fig. 10. Air-side pressure drop correlation.

water mass velocity as given by Eq. (31). The experimental results of the pressure drop over the tube bundle as a function of the air mass velocity are shown in Fig. 9. From the experimental results the following correlation for the air-side pressure drop are derived, 0:22 Dp ¼ 10:2 G1:8 a Gw

ð31Þ 2

2

for 0.7 < Ga < 3.6 kg/(m s) and 1.8 < Gw < 4.7 kg/(m s). Eq. (31) is shown in Fig. 10 for different air and deluge water mass velocities. The correlations developed for the water film heat transfer coefficient, air–water mass transfer coefficient and the air-side pressure drop are only valid for a deluged tube bundle consisting of 15 tube rows with 38.1 mm diameter tubes arranged on a 76.2 mm triangular pitch. For the simplified Merkel-type analysis the assumption is made that the outlet air is saturated. The dry bulb temperature is