Applied Thermal Engineering 27 (2007) 20662073

www.elsevier.com/locate/apthermeng

Heat transfer and pressure drop correlations for the wavy n and
at tube heat exchangers
Dong Junqi
a

a,*

, Chen Jiangping a, Chen Zhijiu a, Zhou Yimin b, Zhang Wenfeng

Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200030, China
b
Zhejing Yinlun Machine Co. Ltd., Zhejiang 317200, China
Received 23 August 2006; accepted 29 November 2006
Available online 12 January 2007

Abstract
A total of 11 cross-ow heat exchangers having wavy n and at tube were studied experimentally. A series of tests were conducted
for air side Reynolds number in the range of 8006500 with dierent n pitches, n lengths and n heights, at a constant tube-side water
ow rate of 2.5 m3/h. The air side thermal performance data were analyzed using the eectiveness-NTU method. The characteristics of
heat transfer and pressure drop for dierent geometry parameters were reported in terms of Colburn j-factor and Fanning friction factor
f, as a function of Re. The eects of n pitch, n height and n length on the performance of heat transfer and pressure drop were examined. The general correlations for j and f factors were derived by multiple linear regression analysis and F test of signicance. The correlations for j and f factors can predict 95% of the experimental data within 10%.
 2006 Elsevier Ltd. All rights reserved.
Keywords: Flat tube heat exchangers; Performance testing; Wavy n; Correlation

1. Introduction
Extended or nned surfaces are widely used in compact
heat exchanger to enhance the heat transfer and reduce the
size. Common among these are automobile radiators,
charge air coolers, automobile air-conditioning evaporators and condensers to meet the demand for saving energy
and resources. In these applications, the heat transfer is
normally limited by the thermal resistance on the air side
of the heat exchangers. Therefore, various augmented surfaces have been developed to improve air side heat transfer
performance. Typical n geometries are plain ns, wavy
ns, oset strip ns, perforated ns and multi-louvered
ns, which, besides increasing the surface area density of
the exchanger, also improve the convection heat transfer
coecients. Of these, wavy ns are particularly attractive
for their simplicity of manufacture and potentials for

Corresponding author. Tel.: +86 21 62933242; fax: +86 21 62932601.

E-mail address: [email protected] (D. Junqi).

1359-4311/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2006.11.012

enhanced thermal-hydraulic performance. The air side

thermal hydraulic performance of wavy n and round tube
heat exchangers have been studied by many researchers
[14]. However, the study of wavy n and at tube heat
exchanger is very limited [4]. The air-side thermal hydraulic
performance of wavy n and at tube heat exchangers
depends on the complex geometry of wavy n such as the
corrugation aspect ratio (2A/L), n spacing ratio (Fp/2A),
ow length ratio (Ld/L) and ow cross-section aspect ratio
(Fp/Fh) [5].
The surface corrugations of the wavy ns consist of triangular, sinusoidal, and trapezoidal patterns, and the ow
behavior has been studied both experimentally and computationally. Asako and Faghri [6] numerically investigated
two-dimensional steady laminar ow with Re = 100
1000, and heat transfer in plate channels with triangularproled wall corrugations that are maintained at a uniform
temperature. Subsequently, triangular corrugations with
round corners were considered [7]. Zhang et al. [5] numerical investigated the eects of wall-corrugation aspect
ratios and n space ratios on the vortex structure and

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

2067

Nomenclature
A
2A
Ac
Ao
Af
C
C*
Cp
De
d
Fh
Fp
f
h
hw
j
k
kc
ke
Ld
L
l
l0
m0
m_
NTU
Nu
N
Pr
Dp

area (m2)
twice of wavy n amplitude (mm)
minimum free-low area for air side (m2)
total air side heat transfer surface area (m2)
n surface area (m2)
heat capacity rate (W/k)
capacity ratio
specic heat at constant pressure (J/kg K)
hydraulic diameter of n entrance (m)
hydraulic diameter of tube hole (mm)
n height (mm)
n pitch (mm)
fanning friction factor
n heat transfer coecient of air side (W/m2 K)
heat transfer coecient of water side (W/m2 K)
Colburn j factor
thermal conductivity (W/(m K))
abrupt contraction pressure-loss coecient
abrupt expansion pressure-loss coecient
wavy n length (mm)
wavy n wavelength (mm)
length of at tube (mm)
Eq. (9), l 0 = Fp
h/2

Eq. (9), m0 2h=k f d

mass ow rate (kg/s)
number of transfer units
Nusselt number
number of the experimental data
Prandtl number
air side pressure drop in inlet and outlet of heat
exchanger (Pa)

enhanced heat transfer for low rate with Re = 1001000.

Likewise, Metwally and Manglik [8] have investigated
two-dimensional periodically developed laminar ow and
heat transfer in sinusoidal wavy channel with dierent corrugation aspect ratios. More complex three-dimensional,
cross-corrugated have also been computationally modeled
in a few recent studies [911]. Goldstein and Sparrow [12]
rst studied the corrugated channels with triangular waves
used the naphthalene sublimation method. Rush et al. [13]
conducted ow visualization test for sinusoidal wavy passages to investigate the local heat transfer and ow behavior of the uid in the laminar and transitional ow region.
Using the visualization methods, they reported that the
ow eld is characterized as steady and unsteady and
the location of the onset of mixing is found to depend on
the Reynolds number and channel geometry. In all, it has
generally been observed that wall corrugations induce a
steady vortex or swirl ow in the trough region of the wavy
wall in the low Reynolds number. This results in ow mixing and boundary layer disruption and thinning, thereby
signicantly enhancing the heat transfer.

Q
Qa
Qw
Re
Rew
T1
T2
u
u1
v
d
dwall
gf
ga
e
s
U

average heat transfer rate (W)

air side heat transfer rate (W)
water side heat transfer rate (W)
air side Reynolds number based on n entrance
diameter (u1De/v)
water side Reynolds number based on at tube
hole hydraulic diameter
inlet temperature (C)
outlet temperature (C)
air frontal air velocity (m/s)
maximum air velocity in the n (m/s)
viscosity (m2/s)
n thickness
tube wall thickness
n eciency
air side heat transfer surface eectiveness
eectiveness
contraction ratio of the n array
j or f factor

Subscripts
a
air side
cor
correlation
exp
experimental
f
n
Min
minimum value
Max
maximum value
w
water side
wall
tube wall

However, there are few researchers carrying out full

scale experiments studying the thermal hydraulic performance of wavy n and at tube heat exchangers. The earliest experimental data are given in the classical Kays and
London [14] sourcebook, though it includes only three
types wavy n. The tests covered a range of Reynolds number from 400 to 8000. And after that, little experimental
data of wind tunnel can be found in the public literatures
[11]. As for the general correlations for heat transfer and
pressure drop of the wavy ns, Jacobi et al. in 2001 [4]
reported that it is can not be obtained in the open literatures. General correlations for heat transfer and pressure
drop of the wavy n are not available in the literatures as
quoted by Jacobi et al. [4]. Hence, the generation of the
correlations for heat transfer and pressure drop for wavy
ns is the main objective of the present work.
The present study investigates experimentally the thermal hydraulic performance for wavy n and at tube heat
exchangers with 11 samples. The wavy ns are triangular
prole with round corners for dierent n pitches, n
heights and n lengths. The heat transfer coecients and

2068

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

averaged. The air pressure drop across the heat exchangers

and the nozzles were, respectively, measured by precision
dierential pressure transducers, whose accuracies were
0.4% and 0.25%. The model of the two dierential pressure
transducers are WIDERPLUS-DP and C268, and both of
them are mad in China. The air ow measuring station was
a multiple nozzle code tester based on the ISO 5167
standard.
The hot water ow loop consisted of a storage tank, a
100 kW electric heater, a centrifugal pump, a control unit
and a ow meter. The purpose of this loop was to transfer
heat to the air owing through the heat exchangers. The
temperature of the hot water in the water tank was measured by pre-calibrated RTDs (Pt-100 X) and was controlled by the temperature controller. Its accuracy was
within 0.1 C. After heating the water to the required temperature, the hot water was pumped out of the storage
tank, delivered to the heat exchanger and then returned
to the storage tank. The water temperatures at the inlet
and outlet of the heat exchanger were measured by two
per-calibrated RTDs (Pt-100 X) which have an accuracy
of 0.1 C.

pressure drop for the heat exchangers with dierent geometrical congurations are reported in terms of Colburn j
factor and Fanning friction factor f, as a function of Reynolds numbers based on the n entrance hydraulic diameter. The general correlations for j and f factors are
developed.
2. Experimental
2.1. Experimental set-up
Fig. 1 shows the schematic diagram of the wind tunnel
used in the study. Air and hot water were used as working
uids. The main components of the systems were the heat
exchangers, water ow loop, air supply, instrumentations
and data acquisition systems. The wind tunnel system
was designed to suck room air over the nned side of the
heat exchangers by a 15 kW centrifugal fan. The speed of
the fan could be adjusted by a frequency inverter. The tunnel was a rectangular duct 270 220 mm in cross-section.
To minimize heat loss to the surroundings, the tunnel surface was insulated with a 10 mm thick glass wool layer.
Being supported by stands of perforated steel plate, the
tunnel system was kept 75 cm above the oor level of the
laboratory.
The inlet and exit temperature across the air side of the
heat exchangers were measured by two T-type thermocouple meshes. The inlet measuring mesh consists of eight
thermocouples while the exit mesh contains sixteen thermocouples. These thermocouples were pre-calibrated which
have an accuracy of 0.1 C. The measuring points were
located at positions as described in the ASHRAE standard.
These data signals were individually recorded and then

18

2.2. Test heat exchangers

Figs. 2 and 3 indicate geometrical conguration and terminology of wavy n and at tube heat exchangers. The
number of the tested wavy n and at tube heat exchanger
samples is 11. Table 1 shows the specications of the wavy
ns tested in this study. All tested samples core sizes are
about 250 200 mm due to the dierent n height. All
tested ns were checked before brazing and overall heat
exchangers quality after brazing was excellent.

17

12

14
1

10

8
11

15

13
Data acquisition system

1, air inlet
2, honey cone straightener
3, T/C inlet temperature
measuring station
4, pressure tap (inlet)
5, test unit
6, pressure tap (outlet)
7, T/C outlet temperature
measuring station
8, setting means
9, static pressure tap

10,
11,
12,
13,
14,
15,
16,
17,
18,

multiple nozzle plate

variable exhaust fan system
air outlet
difference pressure tap nozzle
inlet temperature tap water
outlet temperature tap water
data acquisition system
hot water tank
water pump

Fig. 1. Schematic diagram of the wind tunnel test apparatus.

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

2069

water ow rate. The choice of the water ow rate is based

on the principle that the water side thermal resistance is less
than 20% and the temperature drop in the tube side is
higher than 2.0 C [15]. The water ow rate were maintained at 2.5 m3/h. All the experimental data was obtained
on basis of the heat balance which is less than 3%.
2.4. Data reduction
Heat transfer rate required for the calculation of air-side
heat transfer coecient can be expressed as

Fig. 2. The wavy n photo.

Q Qw Qa =2;
Qw m_ w C pw T w1  T w2 ;

2.3. Test condition and method

The tested wavy n and at tube heat exchanger was
installed in the test system. In this work, the exchanger
height was less than the tunnel dimensions, and the bypass
ows were eliminated by a thin layer of foam plastic sandwiched between the heat exchanger core and tunnel edge.
Upon completion of the hot water side links, the water tube
was completely insulated with a 15 mm thick layer of glass
wool. The test was performed in a range of Reynolds number, which is based on hydraulic diameter of n entrance
and maximum air velocity [14], of 8006500. The inlet
water temperature was maintained at 90 C with a constant

Qa m_ a C pa T a2  T a1 :

1
2

Eectiveness-NTU method can be used for obtaining airside heat transfer coecient. The equation for both uids
unmixed is [15],
NTU0:22 expC  NTU0:78  1
;
C
C min
m_ a C pa
e Q=Qmax ; C 

;
C max m_ w C pw
e 1  exp

3
4

We can obtain overall heat transfer coecient (UA) for the

heat exchangers as
UA m_ a C pa NTU:

Fp

Flat Tube

Fh

Fin
A

Fin

Ld

Flat Tube

A-A

Fig. 3. Wavy n and at tube exchangers.

Table 1
Specication of wavy n parameters (mm)
No.

Fin pitch (Fp)

Fin height (Fh)

Fin length (Ld)

Fin thickness (d)

Wavy amplitude (2A)

Wavelength (L)

1
2
3
4
5
6
7
8
9
10
11

2.0
2.25
2.5
2.0
2.25
2.5
2.0
2.25
2.5
2.0
2.0

8.0
8.0
8.0
8.0
8.0
8.0
7.0
7.0
7.0
8.0
10.0

65.0
65.0
65.0
53.0
53.0
53.0
43.0
43.0
43.0
43.0
43.0

0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2

1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5

10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

The air-side heat transfer coecient can be obtained from

the following equation, assuming zero water side fouling
resistance. Due to the contact resistance of the at tube
and n is a resource of uncertainty, the eect of contact
resistance is included in the derived air side resistance
6

Note that the second term of the right-hand-side of the

Eq. (6) indicates the water side thermal resistance and the
third term means the tube wall thermal resistance. For
water-side heat transfer coecients, the Gnielinskis correlation [16,17] for fully developed turbulent ow for the pipe
ow is used. That is
"
 2=3 #
d
0:87
0:4
Nuw 0:012Rew  280Pr 1
:
7
l
The surface eectiveness and n eciency for the dry surface of wavy ns are [17]:

gf

Af
1  gf ;
Ao

tanhm0 l0
;
m 0 l0

s
2h
0
m
;
kf d

8
l F h =2:

From the Eqs. (6)(9), the heat transfer coecient h can be

obtained using the iterative calculation, since the n eciency is the function of heat transfer coecient h.
The Colburn j factor is dened as
h
Pr2=3
10
j
qu1 C p

0.12

Fp=2.00mm Ld=65mm
Fp=2.25mm Ld=65mm
Fp=2.50mm Ld=65mm

0.012

0.08

0.01

The core friction of the heat exchangers was reduced to obtain the Fanning friction factor f. In present study, the pressure drop equation proposed by Kays and London [14],
including the entrance and exit pressure losses, was used to
evaluate the friction factor. The air is treaded as incompressible uid, and the density of air is treated as constant according to average air temperature. The simple equation is
 

Ac
2Dp
f
 kc  ke ;
11
qu21
Ao
ui D e
12
Re
v

0.008
0.04
0.006

0.004
600

800 1000

2000

4000

6000

8000

Re

Fig. 4. Eect of n pitch on the j and f factor.

0.014

0.12

Fh=7.0mm Ld=43mm
Fh=8.0mm Ld=43mm
Fh=10.0mmLd=43mm

0.012

0.08
0.01

j factor

According to the geometry parameters of heat exchanger and the graph given by Kays and London [14], the kc
and ke are 0.4 and 0.2. Accounting for all instrument
errors, property uncertainties, and geometry tolerances,
the uncertainties for the j and f factors are 9.2% and
8.1%, respectively [18].

0.014

0.008
0.04
0.006

3. Results and discussion

3.1. j and f factors versus Reynolds number
600

The air side heat transfer and friction characteristics of

the tested wavy n and at tube heat exchangers are pre-

f factor

ga 1 

f factor

1
1
1
dwall



:
hga Ao UA hw Aw k wall Awall

sented in terms of the Colburn factor j factor and friction

factor f, which are plotted versus the Reynolds number
based on the n entrance hydraulic diameter. Figs. 46
illustrates the eects of n pitch, n height and n length
on the performance of heat exchangers having dierent
geometry parameters. From these gures, it is observed
that the geometry parameters of wavy ns have signicant
eect on the j factor as a function of Re.
Fig. 4 shows that the j and f factors increase with
increasing n pitch at the same Re, in which the wavy have
the same n length of 65.0 mm and n height of 8.0 mm.
The cause of this result from the phenomena is that, when
the n pitch increases, the air ow inside the corrugated
ow channel can be mixed better at the same Reynolds
number. And this better mixing leads to an increase of
the heat transfer coecient. The explanation can be conrmed by the results of Manglik and Zhang [11] with 3D
numerical simulation method for wavy ns. At the same
time, the phenomena also result in the pressure drop
increase. The behaviors of hydraulic thermal performance
with n pitch are dierent, compared with the conventional
nned tube heat exchangers. Wongwises [3] reported that

j factor

2070

800

2000

4000

6000

Re

Fig. 5. Eect of n height on the j and f factor.

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

2071

0.010

0.012

0.12

Ld=65mm Fh=8.0mm
Ld=53mm Fh=8.0mm

0.009

+10%

0.08
0.008

f factor

-10%
j cor

j factor

0.009

0.04
0.006

0.007

0.006

0.005

600

800

2000

4000

6000

Re

0.004
0.004

0.005

0.006

0.007

0.008

0.009

0.010

j exp

Fig. 6. Eect of n length on the j and f factor.

0.08

3.2. Empirical correlations

Based on the previous discussion, it is obvious from the
test data that no single curve can be expected to describe
the complex behaviors about the heat transfer and friction
characteristics of the wavy n. For easier engineer calculation, the empirical correlations for j and f factors were performed by multiple linear regression and F signicance test
[20] on basis of 154 experimental data with dierential
wavy n geometries. The corresponding correlations are
given as follows:

+10%
0.07

0.06

-10%

f cor

the n pitch has no signicant eect on the heat transfer

and pressure drop characteristics of ned tube heat
exchangers.
Fig. 5 presents the eect of n height on j and f factors
of wavy n with n pitch of 2.0 mm and n length of
43.0 mm as a function of Re. From Fig. 7, it is interesting
to observe that the j factor increase with increasing n
height, while the eects of n height on the f factor is little.
The result is dierent to the theory that the n height has
little eect on the characteristics of heat transfer which
was adopted by many numerical researchers whose numerical model is 2D [58]. And the result is similar to that of
Manglik and Zhang [11] 3D numerical simulation, who
reported that, with increasing ow cross-section aspect
ratio, the spatial coverage and strength of the counterrotating vortices are seen to increase, resulting in convective mixing and enhanced heat transfer.
From the Fig. 6 can be observed the eects of n length
on the j and f factors against the Re. In the comparison, the
wavy ns have the same n height of 8.0 mm and n pitch
of 2.0 mm. The j and f factors decrease with increasing n
length under the same Re. The reason may be that the eect
of uid entrance is relatively obvious for wavy n. This
result is in accord with the conclusions of Yasar [19],
who utilized the numerical simulation and reported that
the uid ow and heat transfer become periodically fully
development after 35 cycles.

0.05

0.04

0.03
0.03

0.04

0.05

0.06

0.07

0.08

fexp

Fig. 7. Comparison of experimental data and correlation for j and f

factor.

Correlation of the heat transfer performance of the

wavy ns:
 0:1284  0:153  0:326
Fp
Ld
0:2309 F p
j 0:0836Re
:
13
Fh
2A
L
Correlation of the frictional performance of the wavy ns
 0:3703  0:25  0:1152
Fp
Ld
0:309 F p
f 1:16Re
:
14
Fh
2A
L
Fig. 7 shows the comparison of j and f factors of the
experimental results with those of the proposed correlations. For the heat transfer and friction f factor correlations, Eqs. (13) and (14) can predict 95% of the
experimental data within 10%. The mean deviations of
the correlations Eqs. (13) and (14) are 4.4% and 5.1%,
and the average deviations are 0.4% and 0.3%, according to the Eqs. (15) and (16) [21].
!
1 X /cor  /exp
Average deviation
 100%;
15
N
/exp
!
1 X j/cor  /exp j
 100%:
16
Mean deviation
N
/exp

2072

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

0.018

Kays and London 11.44-3/8W

Present Correlation 11.44-3/8W
Kays and London 11.5-3/8W
Present Correlation 11.5-3/8W

j factor

0.015

0.012

0.009

0.006

1000

2000

3000

4000

5000

6000

7000

8000

Re

0.12

Kays and London 11.44-3/8W

Present Correlation 11.44-3/8W
Kays and London 11.5-3/8W
Present Correlation 11.5-3/8W

heat exchangers. The eects of n pitch, n height and

n length on the thermal hydraulic performance are examined. On the basis of previous discussions, the following
conclusions are made:
1. The j and f factors decrease with increasing Re, in the
tested range of Re, Re = 8006500. And the j and f factors increases with n space increasing at the same Re;
the j factor increases with n height, while the n height
has little eect on the f factor as a function of Re.
2. Correlations of heat transfer and pressure drop for the
wavy ns are developed. The proposed correlations give
fairly good predictive ability against the present test
data. The mean deviations of the correlations for j and
f factors are 4.4% and 5.1%, and the average deviations
are 0.4% and 0.3%, respectively.
Acknowledgements

0.09

f factor

The authors acknowledge the nancial support of Zhejiang Yinlun Machine Co. Ltd. We are grateful to Dr.
Niu and Xianhui Zhang for providing the valuable comments. It would not have been possible to carry out this
study without their help.

0.06

References

0.03
1000

2000

3000

4000

5000

6000

7000

8000

Re

Fig. 8. Comparison of present correlation and other experiment data.

Fig. 8 presents comparison of present correlation and

other experiment data for wavy n. The Kays and London
experiment data is only database for the wavy n from the
public literatures, although it only includes three dierent
wavy n geometries. The present correlations of j and f factor compared with the experiment data of Kays and London based on two wavy ns, which are named 11.43/
8 W and 11.53/8 W [14]. As for the j factors and f factors,
compared with experiment data, the average deviation are
22.6% and 11.5%. This indicates that the present correlations predict the values of j and f factors are smaller than
those of Kays and London under the same Re. We think
the main reasons for the dierence are that the proles of
wavy ns are not identical to those of Kays and London.
The wavy ns proles in this study are the triangular proles with round corners, however the Kays and Londons
wavy n proles are sinusoidal. Another, the fact should
not be neglected that the experimental heat exchangers
cores manufacture irregularities are not identical due to
dierent manufacture, which also result in the dierence.
4. Conclusions
The present experimental study reports the air side thermal hydraulic performance of the wavy n and at tube

[1] C.C. Wang, W.L. Fu, C.T. Chang, Heat transfer and friction
characteristics of typical wavy n and tube heat exchangers, Exp.
Thermal Fluid Sci. 14 (2) (1997) 174186.
[2] C.C. Wang, Y.T. Lin, C.J. Lee, Y.J. Chang, Investigation of wavy n
and tube heat exchangers: a contribution to databank, Exp. Heat
Transfer 12 (1999) 7389.
[3] S. Wongwises, Y. Chokeman, Eect of n pitch and number of tube
rows on the air side performance of herringbone wavy n and tube
heat exchangers, Energy Convers. Manage. 46 (2005) 22162231.
[4] A.M. Jacobi, Y. Park, D. Tafti, X. Zhang, An assessment of the state
of the art, and potential design improvement, for at-tube heat
exchangers in air conditioning and refrigeration applications-Phase I,
ARTI-21 CR/20020-01, Available to US Department of Energy.
[5] J. Zhang, J. Kundu, R.M. Manglik, Eect of n waviness and spacing
on the lateral vortex structure and laminar heat transfer in wavyplate-n cores, Int. J. Heat Mass Transfer 47 (2004) 17191730.
[6] Y. Asako, M. Faghri, Finite-volume solutions for laminar ow and
heat transfer in a corrugated duct, J. Heat Transfer 109 (3) (1987)
627634.
[7] Y. Asako, H. Nakamura, Heat transfer and pressure drop characteristic in a corrugated duct with rounded corners, Int. J. Heat Mass
Transfer 31 (1988) 12371245.
[8] H.M. Metwally, R.M. Manglik, Enhanced heat transfer due to
curvature-induced lateral vortices in laminar ows in sinusoidal
corrugated-plate channels, Int. J. Heat Mass Transfer 47 (2004) 2283
2292.
[9] D.R. Sawyers, M. Sen, H.C. Chang, Heat transfer enhancement in
three-dimensional corrugated channel ow, Int. J. Heat Mass
Transfer 41 (1998) 35593573.
[10] M. Ciafalo, J. Stasiek, M.W. Collins, Investigation of ow and heat
transfer in corrugated passages. II. Numerical simulation, Int. J. Heat
Mass Transfer 39 (1996) 165192.
[11] R.L. Manglik, J. Zhang, Low Reynolds number forced convection in
three-dimensional wavy-plate-n compact channels: n density
eects, Int. J. Heat Mass Transfer 48 (2005) 14391449.

D. Junqi et al. / Applied Thermal Engineering 27 (2007) 20662073

[12] J.L. Goldstein, E.M. Sparrow, Heat mass transfer characteristics for
ow in a corrugated wall channel, J. Heat Transfer 99 (1977) 187195.
[13] T.A. Rush, A.M. Newell, A.M. Jacobi, An experimental study of ow
and heat transfer in sinusoidal wavy passages, Int. J. Heat Mass
Transfer 42 (1999) 15411553.
[14] W.M. Kays, A.L. London, Compact Heat Exchangers, third ed.,
McGraw-Hill, New York, 1984.
[15] C.C. Wang, R.L. Webb, Data reduction for air-side performance of
n-and-tube heat exchangers, J. Exp. Thermal Fluid Sci. 21 (4) (2000)
218226.
[16] V. Gnielinski, New equation for heat and mass transfer in turbulent
pipe and channel ow, Int. Chem. Eng. 16 (1976) 359368.

2073

[17] S.M. Yang, W.Q. Tao, Heat Transfer, third ed., China, Beijing, 1998.
[18] R.J. Moat, Describing the uncertainties in experimental results, Exp.
Thermal Fluid Sci. 1 (1988) 317.
[19] I. Yasar, P. Cem, The eect of channel height on the enhanced heat
transfer characteristics in a corrugated heat exchanger channel, Appl.
Thermal Eng. 23 (2003) 979987.
[20] Z.L. Ding, Error Theory and Data Processing, Harbin Institute of
Technology Press, 2002.
[21] Y.J. Chang, C.C. Chang, A generalized heat transfer correlation for
louver n geometry, Int. J. Heat Mass Transfer 40 (3) (1997) 533544.