International Journal of Heat and Mass Transfer 60 (2013) 241–251

Contents lists available at SciVerse ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: [Link]/locate/ijhmt

Parametric study and field synergy principle analysis of H-type finned tube
bank with 10 rows
Yu Jin, Gui-Hua Tang, Ya-Ling He, Wen-Quan Tao ⇑
Key Laboratory of Thermo Fluid Science & Engineering, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an, Shannxi 710049, China

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, three-dimensional numerical studies are performed for heat transfer and pressure drop
Received 16 May 2012 characteristics of H-type finned tube bank with 10 rows by software FLUENT. The effects of seven geo-
Received in revised form 13 November 2012 metric parameters (tube row number, fin thickness, slit width, fin height, fin pitch, spanwise tube pitch
Accepted 14 November 2012
and longitudinal tube pitch) and Reynolds number are examined. It is found that for the tube bundle
Available online 1 February 2013
studied, the heat transfer and fluid flow are in the developing region, and they become periodically fully
developed after the 10th row. Among the seven geometric parameters spanwise tube pitch has the most
Keywords:
important effect and slit width has the least important effect, with other five parameters in between. The
Waste heat recovery
H-type finned tube
results are also analyzed from the view point of field synergy principle. It is found that the effects of the
Heat transfer eight parameters on heat transfer and fluid flow characteristics can be well described by the field synergy
Pressure drop principle. Correlations of Nu and Eu for the 10-row tube bundle are presented.
Parametric study Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
Field synergy principle
Numerical simulation

1. Introduction Luve Contardo experimental facilities. Mon and Gross [10] investi-
gated the effects of the fin spacing on four-row annular-finned
In order to improve the thermal performance of heat exchang- tube bundles in staggered and in-line arrangements by three-
ers, it is necessary to enhance heat transfer on the side of heat ex- dimensional numerical study. Cheng et al. [11] numerical designed
changer where the thermal resistance is dominant in the overall slotted fin surface with field synergy principle. Qu et al. [12] stud-
heat transfer process. The extended surface has been proved to ied three-dimensional strip fin surfaces with X-arrangement. He
be an effective method for enhancing gas-side heat transfer. Fin- et al. [13] presented a three-dimensional numerical simulation to
and-tube heat exchangers are the typical equipment with ex- study the heat transfer and fluid flow characteristics of the plain
tended surface and widely used in many engineering applications, fin-and-tube heat exchangers in laminar flow. The effects of Rey-
such as air conditioning units, compressor intercoolers, boiler nolds number, fin pitch, the number of tube rows, the spanwise
economizers, etc. pitch and the longitudinal pitch on finned tube heat exchangers
A lot of experimental and numerical studies have been con- were analyzed. Zhou and Tao [14] studied the strip fin with radial
ducted on airside heat transfer performances of fin-and-tube heat strips. Tao et al. [15] and Jin et al. [16] made an optimized design of
exchangers. A brief review is presented below. Wang et al. [1–6] two-row slotted fin surface with X-shape strip arrangement. Tao
made extensive experiments on heat transfer and pressure drop et al. [17] studied the laminar heat transfer and fluid flow charac-
characteristics of wavy fin-and-tube heat exchangers. Madi et al. teristics of wavy fin heat exchangers with elliptic/circular tubes.
[7] investigated the effects of fin thickness, fin pitch and the num- The results were also analyzed from the view point of field synergy
ber of tube rows on the airside performance of wavy fin-and-tube principle. Xie et al. [18] numerically studied the airside laminar
heat exchangers. Halici et al. [8] experimentally investigated the heat transfer and fluid flow characteristics of plain fin-and-tube
effect of the number of tube rows on the heat transfer, mass trans- heat exchangers with large number of large-diameter tube rows.
fer and fluid flow characteristics of plain fin-and-tube heat The effects of Reynolds number, the number of tube rows, tube
exchangers. Lozza and Merlo [9] performed the heat transfer mea- diameter, tube pitch, fin pitch and fin materials were examined.
surements of the plate fin, the wavy fin and the louvered fin in the The above-mentioned different kinds of fin-and-tube heat transfer
surfaces usually have small pitches between two adjacent fins and
are mainly used in air conditioning industry because air is a clean
⇑ Corresponding author. Tel./fax: +86 029 82669106.
working medium.
E-mail address: wqtao@[Link] (W.-Q. Tao).

0017-9310/$ - see front matter Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
[Link]
242 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251

Nomenclature

A area, m2 DT the log mean temperature difference, K

C1eC2eCl turbulence model constants u velocity, m s1
cp specific heat, J kg1 K1 W slit width, m
D tube outside diameter, m
Eu Euler number Greek symbols
Fp fin pitch, m U heat transfer rate, W
Ft fin thickness, m k thermal conductivity, W m1 K1
h heat transfer coefficient, W m2 K1 l viscosity, kg m1 s1
H fin height, m m kinematic viscosity, m2 s1
k turbulent kinetic energy, m2 s2 q density, kg m3
Nu average Nusselt number b local intersection angle, Deg.
p pressure, Pa e turbulent energy dissipation rate, m2 s3
Pr Prandtl number
P pumping power, W Subscripts
Dp pressure drop, Pa in inlet
Re Reynolds number out outlet
S1 spanwise tube pitch, m max maximum
S2 longitudinal tube pitch, m up upstream
S modulus of the mass rate-of-stress tensor, s1 w wall
Sij mean stress rate, s1 eff effective
T temperature, K

For industrial gases, such as gas of a boiler furnace and dis- importance for the application in the heat exchangers of waste
charged gases from many engineering furnaces, discrete fins with heat recovery. In addition, some heat transfer areas of the fin in
rectangular shape are often used to enhance the gas convective the separation zone are removed to reduce the negative effect on
heat transfer. In recent years, with the developments in manufac- heat transfer. Liu et al. [19] conducted experimental study on the
turing techniques, the configurations of fins become more compli- characteristics of heat transfer and fluid flow for H-type finned
cated, among which are the widely used H-type finned tubes in tube. Yu et al. [20] performed experimental tests to study the heat
boiler economizer. H-type finned tube as shown in Fig. 1 is derived, transfer and flow resistance characteristics at air-side of single or
in large part, from the rectangle-type finned tube. Because of its double H-type finned tube banks. The flow and temperature fields
unique groove structure in fin surface, H-type finned tube has were also obtained by numerical simulation and the results agreed
excellent anti-wear and anti-fouling performance, which is of great well with that of their experimental test. Zhang et al. [21] carried

Fig. 1. Schematic configuration of an H-type finned tube heat transfer surface.

 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251 243

out numerical simulation on heat transfer characteristics of H-type Table 1

finned tube. The effects of the number of tube rows and longitudi- Geometric parameters of the basic H-type finned tube in physical model.

nal space on heat transfer coefficient were also analyzed. However, H/mm W/mm Fp/mm Ft/mm D/mm S1/mm S2/mm N
so far no systematic study has been conducted for the effects of 73.4 15 16.875 2.5 38 108 120 10
geometric parameters on heat transfer and pressure drop charac-
teristics of H-type finned tube.
 
As far as the mechanism of heat transfer enhancement is con- De @ @e e e2
q ¼ ap leff þ C 1 e l t S2  C 2 e q  R ð2Þ
cerned, in 1998, Guo et al. [22] proposed a novel concept about Dt @xi @xi k k
the enhancement of convective heat transfer for parabolic flow 2

and showed that the reduction of the intersection angle between where effective viscosity leff = l + lt, and lt ¼ qC l ke with
the velocity and temperature gradient can effectively enhance Cl = 0.0845. R is the rate of strain term given by
the heat transfer. This concept is now called as field synergy prin- C l qg3 ð1  g=g0 Þ e2
ciple. A lot of experimental and numerical studies [23–29] have R¼ ð3Þ
1 þ bg3 k
been conducted, showing that the field synergy principle could
well explain the mechanism of the convective heat transfer where g = sk/e, g0 = 4.38, b = 0.012 and S2 = 2SijSij is the modulus of
 
@u
enhancement. The examples in the above mentioned references the rate of strain tensor expressed as Sij ¼ 12 @u @x
i
þ @xj . The RNG
j i
are all in laminar flow. Zeng and Tao [30] demonstrated that the model gives C1e = 1.42 and C2e = 1.68.
field synergy principle is also valid for turbulent flow. In order to In order to ensure the inlet uniformity, the computational do-
shorten the length of the article, the detail content of the field syn- main is extended upstreamly by five times of the original heat
ergy principle which can be found in above mentioned references, transfer zone for the entrance section. At the exit of heat transfer
is not presented here. region, the domain was extended by 10 times of the original heat
In this paper, numerical study of an H-type finned tube surfaces transfer zone in order to make sure that the local one-way method
is performed. The effects of eight parameters: number of tube can be used for the numerical treatment at the domain outlet.
rows, fin thickness, slit width, fin height, fin pitch, spanwise tube The computational domain includes six boundaries: inlet, out-
pitch, longitudinal tube pitch and Reynolds number on heat trans- let and four symmetrical boundary surfaces (top, bottom, front
fer and fluid flow characteristics are examined. The results are ex- and back). At the inlet boundary, dry air entering the computa-
plained from the view point of the field synergy principle. tional domain is assumed to have uniform velocity uin, tempera-
In the following passage, the geometric model and the computa- ture Tin (420 K) and the turbulent intensity I (5%) with the
tional domain of the studied H-type fin will firstly be provided with velocity components in the y and z directions being zero. The
adopted turbulence model and boundary conditions. Then the def- fluid region is comprised of the inlet, outlet and bundle zone
initions of the studied parameters are offered. Model validation and and the solid region includes the fin. At the solid surfaces, no-
parameter effects of numerical results are presented in details in slip conditions for the velocity are specified. Heat convection
the results and discussion part. Finally some conclusions are made. to the fin and heat conduction in the fin is considered. At the
tube surface constant temperature Tw (360 K) is assigned. At
2. Model description and numerical method the symmetry planes, heat flux is assumed zero. At the fluid
symmetry plane the normal velocity component and the normal
2.1. Physical model first derivatives of other variables are zero, while at the solid
symmetry plane all velocity components and the normal first
A schematic diagram of an H-type multiple-row finned tube derivative of temperature are zero.
heat transfer surface is shown in Fig. 1. The geometric parameters A three-dimensional and steady-state numerical simulation
of the basic H-type finned tube are listed in Table 1. Fig. 1(b) gives a was conducted by using the commercial software FLUENT [33]
top view of the computation domain of the 10-row H-type finned with a second-order discretization scheme for both convective
tube. Fig. 1(c) presents the side view of the computational domain and diffusive terms. When the residual of each variable is less than
shown by the shaded area, and its top and bottom are the symmet- 1.0  105, the numerical computation is regarded converged.
rical lines of the two neighboring fins.
2.3. Parameter definition
2.2. Governing equations and boundary conditions
In the numerical data reduction, the following characteristic
The fluid is assumed to be incompressible with constant prop- and non-dimensional parameters are introduced.
erty and the gas flow is in turbulence and steady condition. Due quin D
to the relatively larger heat transfer coefficient between the cooling Re ¼ ð4Þ
l
water and inner wall of the tube and the large thermal conductivity
hD
of the tube wall, the tube is assumed to be at constant temperature. Nu ¼ ð5Þ
However, the temperature distribution in the fin surface has to be k
  
calculated. Therefore, the computations are of conjugated type, in U ¼ H H ð6Þ
out in
which both the temperature in the solid fin surface and in the fluid
2Dp
is to be determined simultaneously [31]. The tubes and fins are Eu ¼ ð7Þ
made of carbon steel with constant thermal conductivity. qu2in N
The governing equations in Cartesian coordinates include mass, Dp ¼ pin  pout ð8Þ
momentum and energy conservation equations. They can be found ðT out  T w Þ  ðT in  T w Þ
in many references and are not presented here for simplicity. Only DT ¼   ð9Þ
ln TToutTT w
the turbulence model is presented below. in w

The RNG k  e turbulence model [32] is adopted:
 
U
h¼ ð10Þ
Dk @ @k A DT
q ¼ ap leff þ l t S2  q e ð1Þ
Dt @xi @xi P ¼ Dpuin Af ð11Þ
244 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251


In the above equations, uin is the average oncoming flow velocity. H
in

and H are the fluid enthalpy rates of inlet and outlet positions of
out
the computational domain. pin and pout are the bulk pressure at inlet
and outlet section of fin surfaces respectively. Af is the frontal area.
A is the total air side heat transfer area (both tube and fin surface).
Tw is the average temperature of total air side heat transfer area. uc
is the air velocity in the minimum flow cross-section of the tube
row. P is the pumping power.
To analyze the heat transfer and the fluid flow characteristics of
H-type finned tube from the view point of field synergy principle,
the local intersection angle between the velocity and temperature
gradient is determined as follow.
u  @T
@x
þ v  @T@y
þ w  @T
@z
b ¼ arccos ! ð12Þ
j U j  jrTj
And the module-averaged synergy angle of the computation do-
Fig. 2. Variation of the predicted Nu number with grid number systems.
main of the fin area can be obtained by using numerical integration,
P !
j U ji  jrTji  cos bi  dmi
bm ¼ arccos P ! ð13Þ
j U ji  jrTji  dmi Table 2
Geometric parameters of the H-type finned tube for model validation.

Here dvi is the volume element of the control volume. It is interest- H/mm W/mm Fp/mm Ft/mm D/mm S1/mm S2/mm N
ing to note that there are several average schemes from local syn- 84 15 20 2 38 90 90 10
ergy angle to the domain average synergy angle, including
arithmetic mean, volume weight mean, and module-volume weight
average mean [34]. And it was shown that even though the absolute
values of different definitions are different the variation trends of
different definitions are the same. And it is the variation trend that
is used to guide our analysis. However, according to the full under-
standing of the field synergy principle shown in [23], the module-
volume weight average mean is the most appropriate one in which
the effects of local |U|,|rT|, synergy angle and volume are all re-
flected. Hence, this definition is adopted here for comparison.

2.4. Grid independent solution

In order to validate the grid-independency of the solution, five

different grid systems are investigated, including about 160 000,
403 000, 807 000, 1 371 000 and 1 843 000 cells. The predicted aver-
age Nu numbers of the whole tube bundle from the five grid sys-
tems are shown in Fig. 2, from which we can see that the
solution of the grid system of 1 371 000 can be regarded grid-inde-
pendent. This grid system is adopted in our computation. Fig. 3. Correlation-numerical comparison of Nu number and Eu number for model
validation.
2.5. Validation of the computational model

In order to validate the reliability of the computational model 3. Numerical results and discussion
and numerical method, numerical simulation is carried out for
the H-type finned tube heat transfer surface. The geometric param- In the following passage, the comparison between rectangle fin
eters for the H-type finned tube are shown in Table 2, which are and H-type fin is first analyzed, followed by the effects of number
taken from Ref. [19]. The average air velocity ranges from 1 m/s of tube rows, fin thickness, slit width, fin height, fin pitch, spanwise
to 10 m/s and the corresponding Re number ranges from 3834 to tube pitch and longitudinal tube pitch and Reynolds number on
33 072. The predicted Nusselt number and Euler number are com- heat transfer and fluid flow respectively. Then the simulation re-
pared with the corresponding experimental correlations estab- sults are discussed and analyzed from the view point of field syn-
lished by Liu et al. [19] as shown in Fig. 3. For the readers’ ergy principle.
convenience, the correlation of the average Nu number and Euler
number are presented as Eqs. (14) and (15) respectively. From
3.1. Comparison between rectangle-type finned tubes with H-type
Fig. 3, the maximum deviation in the Nusselt number and Euler
finned tube
number are less than 9.5% with the average deviation being around
6.5%. The good agreement between the simulation results and
H-type finned tube is mainly derived from the rectangle-type
tested results indicates that the numerical model is reliable.
finned tube. So this paper first briefly compares H-type finned tube
Nu ¼ 0:09152Re0:7013 Pr 0:33 ð14Þ with the rectangle-type finned tube. Fig. 4(a) presents the variation
0:0449 of heat transfer rate versus Re number for the rectangle-type tube
Eu ¼ 0:2963Re ð15Þ and H-type finned tube, respectively. From Fig. 4(a), we can see
 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251 245

that the heat transfer rate of rectangle-type tube is a bit larger than number was taken as 1.5  104. It may be noted here that in the
H-type finned tube, since the heat transfer area of the rectangle- following presentation of numerical results for examining the ef-
type finned tube is a bit larger than that of H-type finned tube. fects of geometric parameters, the numerical studies were always
Fig. 4(b) shows the relations between Nu number and Re number conducted for Re = 1.5  104 and with one parameter varying in a
of rectangle-type tube and H-type finned tube. We can see that certain range and the other parameters remained the same as H-
the Nu number of H-type finned tube is a bit larger than that of type finned tube defined in the physical model. In addition all
rectangle-type finned tube, due to the removing of some heat non-dimensional geometric parameters were based on the tube
transfer area in the separation zone. In Fig. 4(c), the variation of outside diameter. For the simplicity of presentation this will not
the pressure drop versus Re number is presented. Because the sur- be restated anymore.
face area of the H-type finned tube is a bit smaller than that of the Fig. 5 shows the effects of number of tube rows on bundle
rectangle-type finned tube, the pressure drop of the H-type finned average Nusselt number, intersection angle, pressure drop, Euler
tube is lower than that of the rectangle-type finned tube. Fig. 4(d) number, heat transfer rate per unit frontal area and heat transfer
shows that heat transfer rate per pumping power of the rectangle- rate per pumping power. The total pressure drop and the heat
type finned tube is a bit higher than that of H-type finned tube. It is transfer rate per unit frontal area both increase with the tube
worth noting that in this paper, the medium is clean air, so the ef- number obviously, and the increasing trends are gradually weak-
fects of wear and fouling of heat transfer surface on heat transfer ened. The character of gradually weakening with tube number of
and fluid flow are not considered. Because of the unique groove the heat transfer rate is caused by the fact that the local Nu
structure in fin surface, H-type finned tube has excellent anti-wear number decreases with the increase of the number of tube rows,
and anti-fouling performance, which makes H-type finned tube while this variation character of the pressure drop leads to the
more appropriate used with industry gas. descending variation of the per row Euler number with tube
row number. The increase of average synergy angle with tube
3.2. The effect of number of tube rows number is consistent with the Nusselt number variation. The
heat transfer rate per unit pumping power increases with the in-
For examining the effect of number of tube rows on the heat crease of the tube row number which is due to the fact that
transfer and fluid flow, in the simulation the number of tube rows with the increase of tube row number the positive effect of
varied from 1 to 10. The other parameters remained the same as the total heat transfer rate increase is much larger than the neg-
H-type finned tube defined in the physical model. The Reynolds ative effect of the total pressure drop increase.

Fig. 4. Variations of heat transfer rate, Nu, pressure drop and heat transfer rate per pumping power with Re for rectangle-type and H-type finned tube (a) heat transfer rate,
(b) Nu, (c) pressure drop, (d) heat transfer rate per pumping power.
246 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251

Fig. 5. The effects of number of tube rows on Nusselt number, intersection angle, pressure drop, Euler number, heat transfer rate per unit frontal area and heat transfer rate
per pumping power (Re = 15 000).

The variation of tube average Nusselt number for each row of 3.3. The effect of fin thickness
tubes is presented in Fig. 6. It can be seen that the per-tube local Nus-
selt number decreases with the tube number and approaches a con- For examining the effect of fin thickness on the heat transfer
stant when N P 9. The relative difference between the local Nusselt and fluid flow, simulations were conducted for the non-dimen-
number of tube 9 and 10 is less than 1%, hence when N P 10 the local sional fin thickness from 0.026 to 0.105.
heat transfer can be regarded as periodically fully developed. This The variations of Nusselt number, intersection angle, pressure
implies that the average heat transfer and pressure drop character- drop, heat transfer rate per unit frontal area and heat transfer
istics obtained in this paper are for the developing region of the tube rate per pumping power with fin thickness are shown in
bundle with 10 streamwise tube rows. Further study is needed to re- Fig. 8. First, it can be seen that Nu number increases with the
veal the heat transfer and pressure drop characteristics in the peri- increase of fin thickness, while the synergy angle decreases
odically fully developed region. This study has been underway in quickly with the increase of fin thickness. Such results are con-
the authors’ group and the results will be reported elsewhere. sistent with [30], where the effects of fin thickness on turbulent
In the present study, in order to clarify whether the intersection heat transfer was studied from the view point of FSP. Second,
angles are dependent on the adopted turbulence model, two differ- both the pressure drop and the heat transfer rate per unit frontal
ent models, namely standard k  e model and RNG k  e model are area increase with the increase of fin thickness. With the in-
chosen to simulate the heat transfer and fluid flow of H-type finned crease in fin thickness the average air flow velocity in the fin coil
tube. The variations of the intersection angle b with the number of increases, which leads to the increase of heat transfer rate and
tube rows are provided in Fig. 7. It can be seen that these two lines pressure drop. The heat transfer rate per unit pumping power
have the same variation tendency and the average intersection an- decreases with the increase of fin thickness. This is because
gle b obtained by two models are almost identical when N P 5. As the fact that pressure drop is proportional to u(1.752.0) while
indicated in [34], the most important character of the synergy an- for the convective heat transfer the exponent of velocity is less
gle is its variation trend, rather than the absolute value. The qual- than 0.8. Therefore, during the design of the H-type finned tube,
itatively good agreement between the results obtained by two fin thickness should be made as thin as possible, provided that
different models shows that the result of field synergy principle the heat transfer requirement is met and the manufacturing is
analysis is independent on the adopted turbulence model. realizable.

Fig. 6. Nu number of each row of 10-row H-type finned tube. Fig. 7. Variation of the intersection angle b with the number of tube rows.
 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251 247

Fig. 8. The effects of fin thickness on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area and heat transfer rate per pumping power
(N = 10).

3.4. The effect of slit width The effects of fin height on Nusselt number, intersection angle,
pressure drop, heat transfer rate per unit frontal area and heat trans-
For examining the effect of slit width on the heat transfer and fer rate per pumping power are presented in Fig. 10. Following fea-
fluid flow, during the simulation the non-dimensional slit width tures may be noted. First, it can be seen that the Nu number
varied from 0.158 to 0.632. decreases and the average intersection angle b increases with the in-
Fig. 9 presents the relations between Nusselt number, intersec- crease of fin height, and this is because a longer fin height leads to a
tion angle, pressure drop, heat transfer rate per unit frontal area higher thickness of the thermal boundary layer. Both the pressure
and heat transfer rate per pumping power and slit width. It can drop and the heat transfer rate per unit frontal area increase with
be seen that only the Nu number increases with the increase of slit the increase of fin height. The heat transfer area increases with the
width, while all other parameters decrease with different descend- increase of fin height, which leads to the increase of heat transfer
ing trends. When the slit width is quite narrow, the pressure drop rate and pressure drop. In this paper, Tw is the average temperature
decreases rapidly with its increase. With further increase in the slit of total air side heat transfer area which means that the fin efficiency
width the descending trend of the pressure drop soon becomes is implicitly considered in the temperature difference in Eq. (10)
mild. As for the synergy angle, heat transfer rate per unit frontal from which the heat transfer coefficient is evaluated. Therefore in
area and heat transfer rate per pumping power, they decrease with Eq. (10) A is the total air side heat transfer area (both tube and fin sur-
the increase in the slot width almost linearly. face). It can be seen that the variation tendency of Nu is opposite
with that of the pressure drop. The increasing tendency of the heat
3.5. The effect of fin height transfer rate per unit frontal area is larger than that of the pressure
drop at first, and then the increasing tendency of the pressure drop
The effect of fin height on the heat transfer and pressure drop is larger than that of the heat transfer rate per unit frontal area. So
was examined within a variation range of the non-dimensional there exists an optimal fin height (H/D of 1.45) at which the heat
fin height from 1.32 to 2.36 during the simulation. transfer rate per unit pumping power is the maximum.

Fig. 9. The effects of slit width on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area and heat transfer rate per pumping power
(N = 10).
248 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251

Fig. 10. The effects of fin height on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area and heat transfer rate per pumping power
(N = 10).

3.6. The effect of fin pitch heat transfer surface with the increase in fin pitch is linear, making
the heat transfer coefficient being enhanced in somewhat with the
For examining the effect of fin pitch on the heat transfer and increase in fin pitch. The heat transfer rate per unit pumping power
fluid flow, during the simulation the non-dimensional fin pitch increases with the increase of fin pitch and the increasing tendency
varied from 0.158 to 0.474. is weakened gradually. Within the variation range of the fin pitch
The relations between Nusselt number, intersection angle, pres- (from 0.158 to 0.474) the optimal fin pitch in the aspect of pump-
sure drop, heat transfer rate per unit frontal area and heat transfer ing power does not exist, but such variation trend can be observed.
rate per pumping power and fin pitch are provided in Fig. 11. It can Therefore, a larger fin pitch of the H-type fin tube surface is in favor
be seen that the Nu number increases with the increase of fin pitch of increasing the heat transfer rate per pumping power.
while the average intersection b decreases with the increase of fin
pitch. Both the pressure drop and the heat transfer rate per unit 3.7. The effect of spanwise tube pitch
frontal area decrease with the increase of fin pitch. The increase
of fin pitch leads to lower velocity in the fin coil and less heat The non-dimensional spanwise tube pitch varied from 2.24 to
transfer surfaces per unit frontal area, leading to a lower pressure 3.42 to examine its effect.
drop and heat transfer rate. It should be noted that even though the The effects of spanwise tube pitch on Nusselt number, intersec-
heat transfer rate per unit frontal area decreases with the increase tion angle, pressure drop, heat transfer rate per unit frontal area
in fin pitch, the Nu number varies in the opposite way. The reason and heat transfer rate per pumping power are shown in Fig. 12.
may be explained as follows. The reduction of heat transfer rate With the increase of the spanwise tube pitch, the velocity in the
with the decrease on velocity is not linear while the reduction of fin coil decreases, so both Nusselt number and pressure drop

Fig. 11. The effects of fin pitch on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area and heat transfer rate per pumping power
(N = 10).
 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251 249

Fig. 12. The effects of spanwise tube pitch on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area, heat transfer rate per unit volume and
heat transfer rate per pumping power (N = 10).

decrease, and the synergy angle increases. As indicated before, the row becomes less and less with the increase of the longitudinal
dependency of pressure drop on velocity is much stronger than tube pitch which leads to a higher velocity of the stream impinging
that of heat transfer, hence in the variation range of S1/D studied, on the tube wall, leading to a higher pressure drop and heat trans-
the reduction of Nu is only about 38%, while that of pressure drop fer rate. The heat transfer rate per unit volume decreases with the
is as large as 75%. Such variation trends of heat transfer and pres- increase of longitudinal tube pitch because of the fact that the heat
sure drop leads to the increase of the heat transfer rate per unit transfer area per unit volume decreases with the increase of longi-
pumping power increases with the spanwise tube pitch. tudinal tube pitch. In the variation range of S2/D studied, the
growth of heat transfer rate per unit frontal area is about 9%, while
3.8. The effect of longitudinal tube pitch that of pressure drop is as large as 18% which leads to the decrease
of the heat transfer rate per unit pumping power.
The effect of longitudinal tube pitch on the heat transfer and From the above presentation of the effects of seven geometric
pressure drop was examined within a variation range of the non- parameters on the heat transfer and pressure drop characters it
dimensional longitudinal tube pitch varied from 2.37 to 3.95 dur- can be observed that the most important parameter is spanwise
ing the simulation. tube pitch, and the least important one is slit width with tube
The effects of longitudinal tube pitch on Nusselt number, inter- row number, fin thickness, fin height, fin pitch and longitudinal
section angle, pressure drop, heat transfer rate per unit frontal tube pitch in between.
area, heat transfer rate per unit volume and heat transfer rate
per pumping power are shown in Fig. 13. It can be seen that Nu
number increases with the increase of longitudinal tube pitch, 3.9. The effect of Re number
while the synergy angle decreases quickly with the increase of lon-
gitudinal tube pitch. Both the pressure drop and heat transfer rate Finally the effect of Re number on the heat transfer and pressure
per unit frontal area increase with the increase of longitudinal tube drop are examined. For this purpose simulations were performed
pitch. The effect of wake region of the former tube row on the latter for the inlet velocity from 1 to 10 m/s, corresponding to a variation

Fig. 13. The effects of longitudinal tube pitch on Nusselt number, intersection angle, pressure drop, heat transfer rate per unit frontal area, heat transfer rate per unit volume
and heat transfer rate per pumping power (N = 10).
250 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251

Fig. 14. The effects of Re number on Nusselt number, intersection angle, pressure drop, Euler number, heat transfer rate per unit frontal area and heat transfer rate per
pumping power (N = 10).

range of Re from 2.1  103 to 2.1  104. The other parameters re-
mained the same.
Fig. 14 shows the variations of Nusselt number, intersection an-
gle, pressure drop, Euler number, heat transfer rate per unit frontal
area and heat transfer rate per pumping power with Re. The varia-
tion trends of all the six parameters with Re are expectable with
following two features worth further mentioning. First, the average
intersection angle b increases with the increase of Re, which im-
plies the deterioration of the synergy between the temperature
gradient and velocity with Re. For an angle as large as 88 degree,
a minor change in the angle will result in an appreciable difference
in cosine, and this result agrees with all previous works [11–17].
Second, the stronger dependence of pressure drop with velocity
than that of Nu leads to a descending variation of the heat transfer
rate per unit pumping power decreases with Re. However, when Re
is larger than 10 000, this parameter almost remains constant. Thus
the inlet velocity is an essential factor affecting the comprehensive
performance of the H-type finned tube heat exchanger.

4. Multiple correlation

Based on the above series of parametric computations, it is not

sufficient to end up only with the numerical results, but the data of
heat transfer and pressure drop should be compressed in compact
forms so that some practical correlations can be obtained. Accord-
ing to the similarity theory [35] the following correlation forms are
assumed:
 C 3  C 4  C 5  C 6  C 7  C8
Fp Ft S1 S2 H W
Nu ¼ C 1 ReC 2 ð16Þ
D D D D D D
 C 3  C 4  C 5  C 6  C 7  C8
F p F t S 1 S 2 H W
Eu ¼ C 1 ReC 2 ð17Þ
D D D D D D

The eight coefficients C1, C2, C3, C4, C5, C6, C7, and C8 should be deter-
mined by means of multiple regression technique [36]. The final re-
sults are as follows:
 0:389  0:165  1:108  0:293  0:624  0:029
Fp Ft S1 S2 H W
Nu ¼ 1:66Re0:585 ð18Þ
D D D D D D
 0:693  0:375  3:026  0:388  1:835  0:002 Fig. 15. Comparison between predicted results and numerical data (a) Nu and (b)
Fp Ft S1 S2 H W
Eu ¼ 11:63Re0:157 ð19Þ Eu.
D D D D D D

The predicted results from the correlations and original data are
Re ¼ 2100—21 000; D ¼ 38 mm; F p =D ¼ 0:158—0:474; F t =D
compared in Fig. 15. All of the deviations but one is within 10%
and the average deviation being around 2.3%, indicating that the ¼ 0:026—0:105; S1 =D ¼ 2:24—3:42; S2 =D ¼ 2:37—3:95; H=D
correlations are of sufficient accuracy. The application ranges of ¼ 1:32—2:36; W=D ¼ 0:158—0:632:
the present correlations are listed as follow:
 Y. Jin et al. / International Journal of Heat and Mass Transfer 60 (2013) 241–251 251

It is worth noting that Eqs. (18) and (19) provide the averaged char- [11] Y.P. Cheng, Z.G. Qu, W.Q. Tao, Y.L. He, Numerical design of efficient slotted fin
surface based on the field synergy principle, Numer. Heat Transfer, Part A 45
acteristics of heat transfer and pressure drop of the H-type finned
(6) (2004) 517–538.
tube bank of 10 rows. As a whole, the fluid flow and heat transfer [12] Z.G. Qu, W.Q. Tao, Y.L. He, Three dimensional numerical simulation on laminar
in the bank are in the developing region. Further study is needed heat transfer and fluid flow characteristics of strip fin surfaces with X-
to obtain the heat transfer and pressure drop characteristics of arrangement of strips, ASME J. Heat Transfer 126 (4) (2004) 697–707.
[13] Y.L. He, W.Q. Tao, F.Q. Song, W. Zhang, Three-dimensional numerical study of
the H-type finned tube bank for the fully developed region. heat transfer characteristics of plain plate fin-and-tube heat exchangers from
view point of field synergy principle, Int. J. Heat Fluid Flow 26 (3) (2005) 459–
473.
5. Conclusion [14] J.J. Zhou, W.Q. Tao, Three-dimensional numerical simulation and analysis of
the airside performance of slotted fin surfaces with radial strips, Eng. Comput.
In this paper, three-dimensional numerical studies were per- 22 (7/8) (2005) 940–957.
[15] W.Q. Tao, W.W. Jin, Y.L. He, Z.G. Qu, C.C. Zhang, Optimum design of two-row
formed for heat transfer and fluid flow characteristics of H-type
slotted fin surface with X-shape strip arrangement positioned by ‘‘Front Coarse
finned tube bank with 10 rows of tubes by software FLUENT. The and Rear Dense’’ principle, Part 1: physical and mathematical models and
following conclusions can be made. numerical methods, Numer. Heat Transfer, Part B 50 (8) (2006) 731–
750.
[16] W.W. Jin, Y.L. He, Z.G. Qu, C.C. Zhang, W.Q. Tao, Optimum design of two-row
1. Among the seven geometric parameters spanwise tube pitch slotted fin surface with X-shape strip arrangement positioned by ‘‘Front Coarse
has the most important effect and slit width has the least and Rear Dense’’ principle, Part 2: results and discussion, Numer. Heat
important effect, with other five parameters in between. Transfer, Part B, Part A 50 (8) (2006) 751–772.
[17] Y.B. Tao, Y.L. He, Z.G. Wu, W.Q. Tao, Three-dimensional numerical study and
2. The per-tube local Nusselt number decreases with the tube field synergy principle analysis of wavy fin heat exchangers with elliptic tubes,
number and approaches a constant when N P 9. When Int. J. Heat Fluid Flow 28 (6) (2007) 1531–1544.
N P 10 the local heat transfer can be regarded as periodically [18] G.N. Xie, Q.W. Wang, B. Sunden, Parametric study and multiple correlations on
air-side heat transfer and friction characteristics of fin-and-tube heat
fully developed. exchangers with large number of large-diameter tube rows, Appl. Therm.
3. The variation trends of Nu with the eight parameters are in good Eng. 29 (1) (2009) 1–16.
agreement with the field synergy principle. [19] Y.Z. Liu, Y.C. Yuan, S.Y. Xu, W.B. Wu, Experimental study on the characteristics
of heat transfer and flow resistance for H-type finned tube banks, J. Univ.
4. The numerical data for Nu and Eu can be well correlated by Eqs. Shanghai Sci. Technol. 26 (5) (2004) 457–460.
(18) and (19), respectively, which may be useful to engineering [20] X.N. Yu, Y.C. Yuan, Y.F. Ma, H.L. Liu, Experimental tests and numerical
design for clean gas situation. simulation on heat transfer and resistance characteristics of H-type finned
tube banks, J. Chin. Soc. Power Eng. 30 (6) (2010) 433–438.
[21] Z.X. Zhang, Y.G. Wang, Q.X. Zhao, Numerical simulation and verification on
heat transfer characteristics of H-type finned tubes, J. Chin. Soc. Power Eng. 30
Acknowledgments (5) (2010) 368–371.
[22] Z.Y. Guo, D.Y. Li, B.X. Wang, A novel concept for convective heat transfer
enhancement, Int. J. Heat Mass Transfer 41 (14) (1998) 2221–2225.
We gratefully acknowledge the support of this work by the Na- [23] Z.Y. Guo, W.Q. Tao, R.K. Shah, The field synergy (coordination) principle and its
tional Basic Research Program of China (973 Program, No. applications in enhancing single phase convective heat transfer, Int. J. Heat
2013CB228304, 2011CB710702) and thank doctorate students Li Mass Transfer 48 (9) (2005) 1797–1807.
[24] Z.G. Qu, W.Q. Tao, Y.L. He, Three-dimensional numerical simulation on laminar
Chen, Zhi-Qiang Yu, Zhao-Hui Li, Hu Zhang, and Ming-Jie Li for heat transfer and fluid flow characteristics of strip fin surface with x-
their assistance in this study. arrangement of strips, J. Heat Transfer – Trans. ASME 126 (5) (2004) 697–
707.
[25] S. Shen, W. Liu, W.Q. Tao, Analysis of field synergy on natural convective heat
References transfer in porous media, Int. Commun. Heat Mass Transfer 30 (8) (2003)
1081–1090.
[1] C.C. Wang, W.L. Fu, C.T. Chang, Heat transfer and friction characteristics of [26] W.Q. Tao, Z.Y. Guo, B.X. Wang, Field synergy principle for enhancing
typical wavy fin-and-tube heat exchangers, Exp. Thermal Fluid Sci. 14 (2) convective heat transfer – its extension and numerical verifications, Int. J.
(1997) 174–186. Heat Mass Transfer 45 (18) (2002) 3849–3856.
[2] C.C. Wang, J.Y. Chang, N.F. Chiou, Effects of waffle height on the air-side [27] W.Q. Tao, Y.L. He, Z.G. Qu, Y.P. Cheng, Applications of the field synergy
performance of wavy fin-and-tube heat exchangers, Heat Transfer Eng. 20 (3) principle in developing new type heat transfer enhanced surfaces, J. Enhanced
(1999) 45–56. Heat Transfer 11 (4) (2004) 435–451.
[3] C.C. Wang, J.Y. Jang, N.F. Chiou, A heat transfer and friction correlation for wavy [28] W.Q. Tao, Y.L. He, Q.W. Wang, Z.G. Qu, F.Q. Song, A unified analysis on
fin-and-tube heat exchangers, Int. J. Heat Mass Transfer 42 (10) (1999) 1919– enhancing single phase convective heat transfer with field synergy principle,
1924. Int. J. Heat Mass Transfer 45 (24) (2002) 4871–4879.
[4] C.C. Wang, W.H. Tao, Y.J. Du, Effect of waffle height on the air-side performance [29] J.M. Wu, W.Q. Tao, Investigation on laminar convection heat transfer in fin-
of wavy fin-and-tube heat exchangers under dehumidifying conditions, Heat and-tube heat exchanger in aligned arrangement with longitudinal vortex
Transfer Eng. 21 (5) (2000) 17–26. generator from the viewpoint of field synergy principle, Appl. Therm. Eng. 27
[5] C.C. Wang, Y.M. Hwang, Y.T. Lin, Empirical correlations for heat transfer and (14–15) (2007) 2609–2617.
flow friction characteristics of herringbone wavy fin-and-tube heat [30] M. Zeng, W.Q. Tao, Numerical verification of the field synergy principle for
exchangers, Int. J. Refrig. – Rev. Int. Du Froid 25 (5) (2002) 673–680. turbulent flow, J. Enhanced Heat Transfer 11 (4) (2004) 453–459.
[6] C.C. Wang, J.S. Liaw, B.C. Yang, Airside performance of herringbone wavy fin- [31] W.Q. Tao, Numerical Heat Transfer, second ed., Xi’an Jiaotong University Press,
and-tube heat exchangers – data with larger diameter tube, Int. J. Heat Mass Xi’an, China, 2001.
Transfer 54 (5–6) (2011) 1024–1029. [32] V. Yakhot, S.A. Orszag, Renormalization group analysis of turbulence. I. Basic
[7] M. Abu Madi, R.A. Johns, M.R. Heikal, Performance characteristics correlation theory, J. Sci. Comput. 1 (1) (1986) 7–20.
for round tube and plate finned heat exchangers, Int. J. Refrig. – Rev. Int. Du [33] F. Incorporated, Fluent 6.2 User’s Guide, in: Fluent Incorporated Lebanon, NH,
Froid 21 (7) (1998) 507–517. USA, 2004.
[8] F. Halici, I. Taymaz, M. Gunduz, The effect of the number of tube rows on heat, [34] J.J. Zhou, Research on the Mechanisms of Heat Transfer Enhancement and
mass and momentum transfer in flat-plate finned tube heat exchangers, Saving Energy for Compact Heat Exchanger Surfaces and Its Optimization
Energy 26 (11) (2001) 963–972. (Desertation), Xi’an Jiaotong Univesity, Xi’an, China, 2005.
[9] G. Lozza, U. Merlo, An experimental investigation of heat transfer and friction [35] S.M. Yang, W.Q. Tao, Heat Transfer, fourth ed., Higher Education Press, Beijing,
losses of interrupted and wavy fins for fin-and-tube heat exchangers, Int. J. China, 2006.
Refrig. – Rev. Int. Du Froid 24 (5) (2001) 409–416. [36] J.G. Orme, T. Combs-Orme, Multiple Regression with Discrete Dependent
[10] M.S. Mon, U. Gross, Numerical study of fin-spacing effects in annular-finned Variables, Oxford University Press, New York, USA, 2009.
tube heat exchangers, Int. J. Heat Mass Transfer 47 (8–9) (2004) 1953–1964.