Three-Dimensional Numerical

Study of Flow and Heat Transfer

Enhancement Using Vortex
Generators in Fin-and-Tube Heat
P. Chu Exchangers
Y. L. He
e-mail: [email protected] In this paper, a three-dimensional numerical investigation was performed for heat trans-
fer characteristics and flow structure of full scale fin-and-tube heat exchangers with
W. Q. Tao rectangular winglet pair (RWP). For the Reynolds number ranging from 500 to 880, the
baseline configuration (without RWP) is compared with three enhanced configurations
State Key Laboratory of Multiphase Flow in (with RWP): inline-1RWP case, inline-3RWP case, and inline-7RWP case. It was found
Power Engineering, that the air-side heat transfer coefficient improved by 28.143.9%, 71.387.6%, and
School of Energy and Power Engineering, 98.9131% for the three enhanced configurations, with an associated pressure drop
Xian Jiaotong University, penalty increase of 11.325.1%, 54.472%, and 88.8121.4%, respectively. An overall
Xian, performance comparison was conducted by using the London area goodness factor. It is
Shaan xi 710049, China revealed that among the three enhanced configurations, the inline-1RWP case obtains the
best overall performance, and the inline-3RWP case is better than the inline-7RWP case.
The numerical results were also analyzed on the basis of the field synergy principle to
provide fundamental understanding of the relation between local flow structure and heat
transfer augmentation. It was confirmed that the reduction in the average intersection
angle between the velocity vector and the temperature gradient was one of the essential
factors influencing heat transfer enhancement. The analysis also provides guidelines for
where the enhancement technique is highly needed. DOI: 10.1115/1.3139185

Keywords: longitudinal vortex generator, heat transfer enhancement, fin-and-tube heat

exchanger

1 Introduction vortex generators LVGs. In recent years, the implementation of

vortex generators in fin-and-tube heat exchangers has received
Fin-and-tube heat exchangers are widely used in various engi-
more and more attention. Fiebig 1 numerically and experimen-
neering fields such as heating, ventilating, air conditioning, and
tally investigated embedded vortices in an internal flow. Deb and
refrigeration HVACR systems, automobiles, petroleum, elec-
Biswas 2 numerically analyzed heat transfer characteristics and
tronics cooling, and chemical industry. Owing to the requirement
flow structure in laminar and turbulent flows through a rectangular
of smaller volume, less cost, quiet, and high efficiency in opera- channel containing built-in vortex generators by means of solu-
tion, high efficiency compact heat exchanger is very necessary to tions of the full NavierStokes and energy equations. Jacobi and
achieve these goals. The total thermal resistance for such kind of Shah 3 gave an excellent review on heat transfer surface en-
heat exchangers is comprised of three parts: the air-side convec- hancement through the use of longitudinal vortices LVs. Biswas
tive resistance, the wall conductive resistance, and the liquid-side et al. 4 experimentally and numerically examined the flow struc-
convective resistance. The heat transfer coefficient on the air-side ture and heat transfer effects of longitudinal vortices in a channel
is typically low due to the thermophysical property of air and flow. Liou et al. 5 experimentally investigated heat transfer and
relative low frontal velocity. Thus, the air-side thermal resistance fluid flow in a square duct with 12 different shaped vortex gen-
is the dominant part to heat transfer and efforts to improve the erators. Gentry and Jacobi 6 experimentally explored the heat
performance of these heat exchangers should focus on the air-side transfer enhancement by delta wing-generated vortices in flat plate
surfaces. and developing channel flows. More recently, Joardar and Jacobi
Vortex generation is a new and innovative strategy, which holds 7 experimentally evaluated the potential of winglet type vortex
promise in air-side heat transfer enhancement. Vortex generators generator arrays for air-side heat transfer enhancement of a com-
VGs such as wings and winglets can introduce vortices into the pact plain fin-and-tube heat exchanger by full scale wind tunnel
flow field causing heat transfer enhancement. Vortex generators testing. They 8 also numerically investigated the flow and heat
can be punched, mounted, or embossed on a heat transfer surface. transfer enhancement using an array of delta winglet vortex gen-
When fluid flows through vortex generators, vortices are gener- erators in a fin-and-tube heat exchanger. OBrien et al. 9 experi-
ated due to the friction and separation on the edge of the vortex mentally investigated the forced convection heat transfer in a nar-
generator. Extensive studies were done on heat transfer character- row rectangular duct fitted with an elliptical tube and one or two
istics and flow structure for heat exchangers with longitudinal delta winglet pairs. They found that the addition of the single
winglet pair to the oval tube geometry yielded significant heat
transfer enhancement, averaging 38% higher than the oval tube
Contributed by the Heat Transfer Division of ASME for publication in the JOUR-
NAL OF HEAT TRANSFER. Manuscript received September 9, 2008; final manuscript
with no winglet case.
received March 2, 2009; published online June 25, 2009. Review conducted by Frank From the traditional point of view, the reasons why the vortex
Cunha. generators such as wings or winglets can augment heat transfer

Journal of Heat Transfer Copyright 2009 by ASME SEPTEMBER 2009, Vol. 131 / 091903-1

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 Fig. 1 Schematic of the core region of a fin-and-tube heat ex-
changer with RWPs

are attributed to intensified fluid mixing, boundary layer modifi-

cation, and flow destabilization. Guo et al. 10 proposed a novel
concept for convective heat transfer enhancement of parabolic
flow, and they indicated that the reduction in the intersection angle Fig. 3 Winglet type vortex generator dimensions and the
placement with respect to the tube
between the velocity and temperature gradient can effectively en-
hance the heat transfer. Furthermore, Tao and co-workers 11,12
proved this new concept for elliptic flow with a not too small
Peclet number greater than 100. Qu et al. 13 numerically de- winglet pairs and their placement with respect to the tube. In Fig.
signed several kinds of efficient slotted fin surfaces using the field 4, X is the streamwise direction, Y is the spanwise direction, and Z
synergy principle. In the present study, the effects of three param- stands for the fin pitch direction. Figure 4a gives a top view of
eters the Reynolds number, the number of vortex generators, and the computational domain for a fin-and-tube heat exchanger with
the tube bank arrangement on heat transfer characteristics and rectangular winglet pairs RWPs, and Fig. 4b presents a side
fluid flow structure of fin-and-tube heat exchangers with vortex view of the computational domain. Two neighboring fins form a
generators are explored. The heat conduction in the fins and the channel with H = 3.63 mm, B = 12.7 mm, and L = 177.8 mm. The
thickness of the vortex generators are considered. The temperature first tube, with diameter D = 10.67 mm, is located at X
distribution in the fins will be determined by solving the conjugate = 12.7 mm from the inlet of the flow channel. The fin material is
heat transfer problem. A full scale model is adopted in our numeri- aluminum and the fin thickness is Ft = 0.18 mm. The air inlet
cal study, which involves all the tube rows in a seven-row fin-and- temperature is 310.6 K and the tube temperature is 291.77 K.
tube heat exchanger with vortex generators. On the basis of the Because of the geometry character of the symmetry of the fin-and-
field synergy principle, we will explore the fundamental mecha- tube heat exchanger, the region in the x y plane outlined by the
nism for heat transfer enhancement. dashed lines in Fig. 4a is selected as the computational domain.
Due to the high heat transfer coefficient inside the tube and the
2 Model Descriptions high thermal conductivity of the tube wall, the tube temperature is
set as constant. However, the temperature distribution on the fin
2.1 Physical Model. The schematics of a fin-and-tube heat surface is unknown and will be determined during the iteration
exchanger with longitudinal vortex generators is shown in Fig. 1. process. In order to solve this conjugated problem, the computa-
There are four basic vortex generatorsdelta wing, rectangular tional domain should contain the whole fin surface during the
wing, delta winglet pair, and rectangular winglet pair see Fig. 2. numerical simulation. Therefore, the region in the x z plane
In this present study, we adopt the rectangular winglet pair as the outlined by the dashed lines in Fig. 4b is taken as the computa-
vortex generator. Figure 3 shows the dimensions of rectangular tional domain due to the geometry character of periodicity.

Fig. 2 Four basic vortex generator forms

091903-2 / Vol. 131, SEPTEMBER 2009 Transactions of the ASME

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 Fig. 8 Local temperature distributions on the middle cross section for the baseline case
and the inline-3RWP case at Re= 850

the inlet region can be primarily attributed to the large tempera- the size of the wake region will be reduced and the heat transfer is
ture gradient and the thin boundary layer at the entrance region. augmented. Comparing the curves for the inline-1RWP case, the
With the growth of velocity and thermal boundary along the chan- inline-3RWP case, and the inline-7RWP case, it can be found that
nel, the Nusselt number decreases. For the baseline case, the curve there are seven pairs of peaks around tube 1 the locations for the
shows seven peaks the locations for the peaks are X = 25.3 mm, peak pair are X = 25.3 mm and 30.7 mm, tube 2 X = 46.9 mm
54.1 mm, 79.2 mm, 104.4 mm, 129.5 mm, 154.6 mm, and 181.6 and 54.1 mm, tube 3 X = 72.0 mm and 79.2 mm, tube 4 X
mm at the same intervals along the longitudinal direction, corre- = 98.9 mm and 106.2 mm, tube 5 X = 124.1 mm and 131.3
sponding to the junctions of fin and the seven tubes. This is mm, tube 6 X = 149.3 mm and 156.4 mm, and tube 7 X
mainly due to the formation of horseshoe vortices at the fin-and- = 169.0 mm and 179.7 mm for the inline-7RWP case, and there
tube junction. After the first tube, for the baseline configuration, are three pairs of peaks around the tube 1 X = 25.3 mm and 30.7
the curve decreases abruptly until it meets the second tube in the mm, tube 3 X = 73.8 mm and 81.0 mm, and tube 5 X
channel. However, for the enhanced configurations, the curves = 124.1 mm and 131.3 mm for the inline-3RWP case. However,
decrease a little and rise to a new peak located at X = 30.7 mm there is one pair of peaks for the inline-1RWP case around tube 1
before the second tube. This is mainly due to the formation of X = 25.3 mm and 30.7 mm. But for the baseline configuration,
longitudinal vortices around the RWPs in the vicinity of the first there is only one peak in the vicinity of the tubes. It is interesting
tube. It can be seen clearly from the figure that the average Nus- to note that the location of the pair of peaks is corresponding to
selt number in the wake region of the tube for the baseline case is the location of the tubes and RWPs. The presence of the peak pair
apparently lower than that for the enhanced case. For the inline- around the tube may be due to the formation of horseshoe vortices
7RWP case, the curve in the wake region of the tube is slightly by the fin-and-tube junction and the formation of LVs by RWPs.
higher than that for the inline-1RWP case and the inline-3RWP After the fifth tube, the heat performance of the baseline configu-
case. This can be attributed to the generation of the strong swirl- ration gradually approaches that of the enhanced configurations.
ing flow by RWPs. The strong swirling flow can drag the fluid in Figure 10a presents the variation in the air-side heat transfer
the wake region to the mainstream and enhance the mixing. The coefficient hair versus the Re number. It can be seen from Fig.
wake will be suppressed by the high momentum swirling flow and 10a that the heat transfer coefficients hair for both the baseline
case and the enhanced cases increase with increasing Reynolds
number. With a higher value of Reynolds number, the thermal

Fig. 9 Longitudinal distributions of line-weighted average

Nusselt number on the RWP-mounted fins for the baseline case Fig. 10 Variations in the air-side heat transfer coefficient hair
and three enhanced cases at Re= 850 and the pressure drop P versus the Re number

Journal of Heat Transfer SEPTEMBER 2009, Vol. 131 / 091903-5

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 Fig. 6 Different configurations for fin-and-tube heat exchang-
Fig. 5 Experimental numerical comparison of hair and P for
ers with and without RWPs
model validation

3 Validation of Numerical Results which delay the boundary layer separation from the tube, reduce
the size of recirculation region, accelerate the local flow velocity,
In order to validate the independency of solution on the grid,
intensify the mixing between cold fluid and hot fluid, disrupt the
three different grid systems are investigated, which include about
thermal boundary layer, and thus enhance heat transfer. From Fig.
800,000, 1,260,000, and 1,500,000 cells, respectively, for the fin-
7b we can see that the point of separation on the tube travels
and-tube heat exchanger with one RWP. The change in the aver-
downstream, and the size of the recirculation region is reduced.
aged Nu number is less than 3% among the three different grid
Figure 8 illustrates the local temperature distributions on the
systems. For the present study, the final grid number is selected to
middle cross section parallel to the y z plane for the baseline
be about 1,260,000. Similar validations are also conducted for
case and the inline-3RWP case at Re= 850. Comparing Fig. 8a
other cases.
with Fig. 8b we can see that the temperature distribution in the
In order to validate the reliability of the numerical method be-
vicinity of the inlet region is almost identical for the baseline case
ing used, the numerical simulation is conducted for a fin-and-tube
and the inline-3RWP case. As the air approaches the RWPs, the
heat exchanger with the same geometrical configurations as pre-
longitudinal vortices are generated and the heat transfer is signifi-
sented in Ref. 7. The predicted results are compared with the
cantly enhanced. The enhancement of heat transfer is notable even
experimental results from Ref. 7. The overall pressure loss pen-
far downstream of RWPs because the longitudinal vortices persist
alty P and the air-side heat transfer coefficient hair are shown in
for several winglet chords. Thus, the temperature behind the
Figs. 5a and 5b, respectively. The average discrepancy be-
RWPs for the inline-3RWP case is distinctly lower than that in the
tween the predicted pressure loss and the experimental values is
corresponding region for the baseline case. The results indicate
less than 11%, and the average discrepancy between the predicted
that the RWPs can significantly enhance the heat transfer with
air-side heat transfer coefficient and the experimental values is
moderate pressure loss penalty.
less than 5%. The good agreement between the predicted and
Figure 9 shows the longitudinal distribution of line-weighted
experimental results indicates that the numerical model is reliable
average Nusselt number on the RWP-mounted fins for the base-
to predict heat transfer characteristics and flow structure in com-
line and the three enhanced configurations at Re= 850. For both
pact heat exchangers.
the baseline and enhanced configurations, the variations in the
4 Results and Discussion Nusselt number are almost identical up to the location of the first
tube. The Nusselt number is very high in the vicinity of the inlet
4.1 Flow Pattern and Heat Transfer. In order to study the region and decreases gradually along the flow channel until the
influence of RWPs on the heat transfer characteristics and flow first fin-and-tube junction. The extremely high Nusselt number at
structure for fin-and-tube heat exchangers, a comparative investi-
gation for fin-and-tube heat exchangers with and without RWPs is
performed. The RWPs are symmetrically mounted adjacent to the
tubes. The angle of attack is set at 15 deg. The Re based on the
hydraulic diameter ranges from 500 to 880. Figure 6 shows dif-
ferent configurations for fin-and-tube heat exchangers with and
without RWPs.
Figure 7 presents the local velocity distribution on the middle
cross section parallel to the x y plane for the baseline case and
the inline-3RWP case at Re= 850. From Fig. 7a it can be seen
that a recirculation region behind the tube is formed. The recircu-
lating flow is separated from the main stream, and the fluid in this
region is thermally isolated. Owing to this, a poor heat transfer
zone is developed. Comparing Fig. 7a with Fig. 7b we can see
that the velocity distribution and the wake structure for the inline-
3RWP case are totally different from those for baseline case. The Fig. 7 Local velocity distributions on the middle cross section
RWPs for the inline-3RWP case develop longitudinal vortices, for the baseline case and the inline-3RWP case at Re= 850

091903-4 / Vol. 131, SEPTEMBER 2009 Transactions of the ASME

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 Fig. 8 Local temperature distributions on the middle cross section for the baseline case
and the inline-3RWP case at Re= 850

the inlet region can be primarily attributed to the large tempera- the size of the wake region will be reduced and the heat transfer is
ture gradient and the thin boundary layer at the entrance region. augmented. Comparing the curves for the inline-1RWP case, the
With the growth of velocity and thermal boundary along the chan- inline-3RWP case, and the inline-7RWP case, it can be found that
nel, the Nusselt number decreases. For the baseline case, the curve there are seven pairs of peaks around tube 1 the locations for the
shows seven peaks the locations for the peaks are X = 25.3 mm, peak pair are X = 25.3 mm and 30.7 mm, tube 2 X = 46.9 mm
54.1 mm, 79.2 mm, 104.4 mm, 129.5 mm, 154.6 mm, and 181.6 and 54.1 mm, tube 3 X = 72.0 mm and 79.2 mm, tube 4 X
mm at the same intervals along the longitudinal direction, corre- = 98.9 mm and 106.2 mm, tube 5 X = 124.1 mm and 131.3
sponding to the junctions of fin and the seven tubes. This is mm, tube 6 X = 149.3 mm and 156.4 mm, and tube 7 X
mainly due to the formation of horseshoe vortices at the fin-and- = 169.0 mm and 179.7 mm for the inline-7RWP case, and there
tube junction. After the first tube, for the baseline configuration, are three pairs of peaks around the tube 1 X = 25.3 mm and 30.7
the curve decreases abruptly until it meets the second tube in the mm, tube 3 X = 73.8 mm and 81.0 mm, and tube 5 X
channel. However, for the enhanced configurations, the curves = 124.1 mm and 131.3 mm for the inline-3RWP case. However,
decrease a little and rise to a new peak located at X = 30.7 mm there is one pair of peaks for the inline-1RWP case around tube 1
before the second tube. This is mainly due to the formation of X = 25.3 mm and 30.7 mm. But for the baseline configuration,
longitudinal vortices around the RWPs in the vicinity of the first there is only one peak in the vicinity of the tubes. It is interesting
tube. It can be seen clearly from the figure that the average Nus- to note that the location of the pair of peaks is corresponding to
selt number in the wake region of the tube for the baseline case is the location of the tubes and RWPs. The presence of the peak pair
apparently lower than that for the enhanced case. For the inline- around the tube may be due to the formation of horseshoe vortices
7RWP case, the curve in the wake region of the tube is slightly by the fin-and-tube junction and the formation of LVs by RWPs.
higher than that for the inline-1RWP case and the inline-3RWP After the fifth tube, the heat performance of the baseline configu-
case. This can be attributed to the generation of the strong swirl- ration gradually approaches that of the enhanced configurations.
ing flow by RWPs. The strong swirling flow can drag the fluid in Figure 10a presents the variation in the air-side heat transfer
the wake region to the mainstream and enhance the mixing. The coefficient hair versus the Re number. It can be seen from Fig.
wake will be suppressed by the high momentum swirling flow and 10a that the heat transfer coefficients hair for both the baseline
case and the enhanced cases increase with increasing Reynolds
number. With a higher value of Reynolds number, the thermal

Fig. 9 Longitudinal distributions of line-weighted average

Nusselt number on the RWP-mounted fins for the baseline case Fig. 10 Variations in the air-side heat transfer coefficient hair
and three enhanced cases at Re= 850 and the pressure drop P versus the Re number

Journal of Heat Transfer SEPTEMBER 2009, Vol. 131 / 091903-5

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 inline-7RWP case. The incremental pressure drop is mainly due to
the additional formed drag induced by the winglet vortex genera-
tors.
4.2 Influence of the Tube Bank Arrangement. In order to
study the influence of the tube bank arrangement on the heat
transfer characteristics and the flow structure for fin-and-tube heat
exchangers with RWPs, a comparative investigation for fin-and-
tube heat exchangers with inline arrangement and staggered ar-
rangement was performed. The angle of attack is set at 15 deg.
The Re based on the hydraulic diameter ranges from 500 to 880.
Figure 11 shows different configurations for fin-and-tube heat ex-
changers with staggered arrangement.
The thermal hydraulic performances of the heat exchangers
with staggered tube arrangement are compared with the heat ex-
changers with inline tube arrangement in Fig. 12. For the 1-RWP
Fig. 11 Different configurations for fin-and-tube heat exchang-
case, the heat transfer coefficient of the staggered-1RWP case is
ers with staggered arrangement
increased by 29.135.1% compared with that of the inline-1RWP
case. For the 3-RWP case, the staggered-3RWP case improves the
heat transfer coefficient from 19.1% to 24.2% over the inline-
boundary layer thickness decreases and the degree of fluid mixing 3RWP case. For the 7-RWP case, the staggered-7RWP case causes
increases. Consequently, a global augmentation in heat transfer is an increase of 17.420.1% in the heat transfer coefficient over the
observed with the increase in Reynolds number. Comparing with inline-7RWP case. Comparing with the inline tube arrangement,
the baseline case, the heat transfer coefficient for the inline-1RWP the heat transfer enhancement for staggered tube arrangement is
case is improved by 28.143.9% over the range of the Reynolds mainly due to the horseshoe vortices generated in the fin-tube
number considered. The heat transfer coefficient for the inline- junction. For the inline tube arrangement, the second and the sub-
3RWP case is increased from 71.3% at Re= 500 to 87.6% at Re sequent tubes are in the tube wake of their upstream tubes. The
= 880 compared with the baseline case. The inline-7RWP case heat transfer in the tube wake is very poor, and therefore the tube
improves the heat transfer coefficient hair from 98.9% to 131% wake greatly reduces the high heat transfer potential of the tube.
over the baseline case. These numerical results demonstrate that However, for the staggered tube arrangement, the formation of
the winglet vortex generators can significantly improve the heat horseshoe vortices and the large temperature gradient around the
transfer performance of the fin-and-tube heat exchangers. Figure tube can play the biggest role in heat transfer augmentation.
10b shows the relation of the pressure drop P and the Re The heat transfer enhancement is accompanied by additional
number. The heat transfer augmentation is usually accompanied pressure losses. The results presented in Fig. 12b indicate the
by additional pressure drop penalty. Comparing with the baseline pressure drop associated with the inline and staggered tube ar-
case, the inline-1RWP case increase in P by 11.325.1% over rangements at different Reynolds numbers. In comparison with
the range of the Reynolds number considered, the inline-3RWP the inline tube arrangement, the staggered-1RWP case increases
case increase in P by 54.472%, and 88.8121.4% for the the pressure drop from 46.9% to 54.1% over the inline-1RWP

Fig. 12 Variations in the air-side heat transfer coefficient hair, the pressure drop P, and
the overall performance j / f versus the Re number

091903-6 / Vol. 131, SEPTEMBER 2009 Transactions of the ASME

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 case, the staggered-3RWP case improves the pressure drop from
22.1% to 31.9% over the inline-3RWP case, and the pressure drop
of the staggered-7RWP case is increased by 19.625.8% com-
pared with that of the inline-7RWP case. The additional pressure
drop penalty is primarily due to the form drag of staggered tubes.
The overall performance of the fin-and-tube heat exchangers
with different tube arrangements are evaluated using the criterion
of the area goodness factor j / f. The staggered cases are compared
with the inline cases in Fig. 12c. For the Reynolds number rang-
ing from 500 to 880, all the inline configurations have a better
performance than the corresponding staggered configurations. For
the case of 1-RWP, the inline-1RWP case increases the j / f ratio
from 9.3% to 19.3% over the staggered-1RWP case. For the case
of 3-RWP, the inline-3RWP case improves the j / f ratio from 2.6%
to 7.2% compared with the staggered-3RWP case. For the case of
7-RWP, the inline-7RWP case causes an increase of 2.37.1% in
the j / f ratio over the staggered-7RWP case. It is interesting to
note that the influence of the tube arrangement on the overall
performance becomes progressively small with the increasing Fig. 13 The average intersection angles for baseline configu-
number of the RWPs. From the viewpoint of area goodness, the ration and enhanced configurations
inline-1RWP case will require a smaller frontal area than the other
cases.
plane for baseline configuration and enhanced configurations at
Re= 850. From the figures we can see that in the vicinity of the
4.3 Discussion on Fundamental Mechanism for Heat
inlet regions, the isotherms are almost perpendicular to the local
Transfer Enhancement. As mentioned above, from the tradi-
streamlines, which indicates that the temperature gradient are al-
tional perspective, vortex generators can enhance convective heat
most in the same direction as the velocity vector. This implies an
transfer because they can disrupt the boundary layer, intensify the
excellent synergy between the velocity field and the temperature
mixing, and increase the disturbance in the channel. All these
field. For the baseline case, it can be seen clearly in Fig. 14a that
functions serve ultimately to bring about heat transfer augmenta-
the recirculation regions appear in the rear of the tubes, which
tion. However, this is insufficient to reveal the fundamental
indicates poor heat transfer there. Furthermore, in most part of the
mechanism for heat transfer enhancement. According to the field
channel for the baseline case, especially in the rear of the tubes,
synergy principle FSP proposed by Guo et al. 10, it was proved
the isotherms are almost parallel to the local streamlines, which
by Tao and co-workers 11,12 that the existing single phase con-
indicates a bad synergy between the velocity field and the tem-
vective heat transfer enhancement methods can be unified by FSP.
perature field. It appears that the poor heat transfer regions are the
For simplicity, the details about FSP can be found in relevant
same regions of bad synergy. This is in accordance with the field
references 1013 and will not be restated here.
synergy principle. Due to the presence of RWPs for the inline-
The local intersection angle between the velocity vector and the
1RWP case, the velocity field and the temperature field are modi-
temperature gradient at a grid node is defined as
fied. From Figs. 14c and 14d it can be seen clearly that the

T T T synergy between the velocity field and the temperature field for

u +v +w the inline-1RWP case is significantly improved, especially in the
U T x y z
= cos1 = cos1 7 downstream region of the RWPs. However, for the inline-1RWP
UT Ugrad T case, in the downstream region of the channel, the strength of
The average intersection angle of the computation domain can be longitudinal vortices decays, and the synergy between the velocity
obtained by using numerical integration and temperature gradient is not improved a lot in comparison with
the baseline configuration. For the staggered-1RWP case, we can
i,j,kxiy j zk see from Fig. 14e and 14f that the synergy between the veloc-
i,j,k ity and the temperature gradient is further improved in compari-
m = 8
x y z
i,j,k
i j k
son with the inline-1RWP case. The better synergy is mainly due
to the staggered arrangement of the tubes. For the inline-1RWP
case, the second and the subsequent tubes are in the wake region
where the subscripts i, j, and k refer to the control volume of the of their upstream tubes. The wake region of the tube is associated
fluid. with the low heat transfer coefficient. Therefore the second and all
The average intersection angles for the baseline configuration the subsequent tubes are in poor heat transfer region, and it does
and enhanced configurations are presented in Fig. 13. From the nothing good to heat transfer enhancement. However, for the
figure we can see that the average intersection angle increases staggered-1RWP case, the upstream fluid will directly encounter
with the increasing Reynolds number for the three configurations, with the front part of each tube in the channel, and horseshoe
which indicates that the synergy between the velocity and the vortices will be generated to augment the heat transfer. Conse-
temperature gradient becomes worse with the increasing Reynolds quently, the heat transfer of the staggered-1RWP case is better
number. However, at the same Reynolds number the average in- than that of the inline-1RWP case.
tersection angle m for the enhanced configurations is smaller than
that for the baseline configuration, which indicates that the syn- 5 Conclusions
ergy for the enhanced configurations is better than that for the In this paper, three-dimensional numerical simulations are em-
baseline configuration. For the same Reynolds number, the aver- ployed to investigate the heat transfer characteristics and flow
age intersection angle of the baseline configuration is the largest, structure in full scale fin-and-tube heat exchangers with RWPs.
while that of the staggered-1RWP case is the smallest, which is in The major conclusions are drawn as follows.
accordance with their corresponding heat transfer performance.
In order to have a more intuitive understanding of the field 1 Comparing with the baseline case, the heat transfer coeffi-
synergy principle, Fig. 14 presents the streamlines and the isother- cient of the fin-and-tube heat exchanger is improved by
mal distributions on the middle cross section parallel to the x y 28.143.9%, 71.387.6%, and 98.9131% for the inline-

Journal of Heat Transfer SEPTEMBER 2009, Vol. 131 / 091903-7

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 Fig. 14 Distributions of isothermals and streamlines for baseline case and
enhanced cases

1RWP case, the inline-3RWP case, and the inline-7RWP f friction factor
case, respectively. The corresponding pressure drop penalty Fp fin pitch m
is also increased. The enhancement in heat transfer coeffi- Ft fin thickness m
cient outweighs the additional pressure drop penalty gener- h heat transfer coefficient W / m2 K
ated by RWPs. H channel height m
2 The staggered arrangement of tube bank provides a better j Colburn factor
heat transfer coefficient for the fin-and-tube heat exchanger L flow length m
compared with the inline tube arrangement. The staggered N the number of control volume or point
tube arrangement avoided the second and all subsequent Nu Nusselt number
tubes being in the tube wake of their upstream tubes, which P pressure Pa
reduce the bad influence of tube wake on downstream P air-side pressure drop Pa
tubes, and hence, augment the heat transfer in the flow Pr Prandtl number
channel. The pressure loss penalty for the staggered ar- Q heat transfer capacity W
rangement is also increased due to the extra form drag in- Re Reynolds number
duced by the staggered tubes. T temperature K
3 For the range of Reynolds number considered, the average
Tin inlet temperature K
intersection angles m decreases in the order of the baseline
Tw wall temperature K
case, the inline-1RWP case, and the inline-3RWP case;
however, the heat transfer performance increases in the or- T bulk average temperature K
der of the inline-3RWP case, the inline-1RWP case, and the p bulk average pressure Pa
baseline case. The smaller intersection angle m means the u velocity in x-direction
better synergy between the velocity and the temperature uin frontal velocity m/s
gradient. The better synergy causes higher heat transfer per- U velocity vector m/s
formance for the heat exchanger. v velocity in the y-direction
Vm mean velocity at Amin m/s
Acknowledgment w velocity in the z-direction
The present work is supported by the Key Project of National
Natural Science Foundation of China Contract No. 50736005 Greek Symbols
and the National Basic Research Program of China 973 Program angle of attack deg
Contract No. 2007CB206902. dynamic viscosity Pa s
density kg/ m3
Nomenclature thermal conductivity W / m K
AT total heat transfer surface area m2 the local intersection angle deg
Amin minimum flow area m2 computational domain
diffusion coefficient
b span of rectangular winglet m
winglet aspect ratio
B channel width m
c rectangular winglet chord length m Subscripts
cp specific heat J/kg K in inlet parameter
D outer tube diameter m m mean value
Dh hydraulic diameter, Dh = 4AminL / AT out outlet parameter

091903-8 / Vol. 131, SEPTEMBER 2009 Transactions of the ASME

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms

 w wall ers, Int. J. Refrig., 31, pp. 8797.
8 Joardar, A., and Jacobi, A. M., 2007, A Numerical Study of Flow and Heat
Transfer Enhancement Using an Array of Delta-Winglet Vortex Generators in a
References Fin-and-Tube Heat Exchanger, ASME J. Heat Transfer, 129, pp. 11561167.
1 Fiebig, M., 1995, Embedded Vortices in Internal Flow: Heat Transfer and 9 OBrien, J. E., Sohal, M. S., and Wallstedt, P. C., 2004, Local Heat Transfer
Pressure Loss Enhancement, Int. J. Heat Fluid Flow, 16, pp. 376388. and Pressure Drop for Finned-Tube Heat Exchangers Using Oval Tubes and
2 Deb, P., and Biswas, G., 1995, Heat Transfer and Flow Structure in Laminar Vortex Generators, ASME J. Heat Transfer, 126, pp. 826835.
and Turbulent Flows in a Rectangular Channel With Longitudinal Vortices, 10 Guo, Z. Y., Li, D. Y., and Wang, B. X., 1998, A Novel Concept for Convec-
Int. J. Heat Mass Transfer, 38, pp. 24272444. tive Heat Transfer Enhancement, Int. J. Heat Mass Transfer, 41, pp. 2221
3 Jacobi, A. M., and Shah, R. K., 1995, Heat Transfer Surface Enhancement 2225.
Through the Use of Longitudinal Vortices: A Review of Recent Progress, 11 Tao, W. Q., Guo, Z. Y., and Wang, B. X., 2002, Field Synergy Principle for
Exp. Therm. Fluid Sci., 11, pp. 295309. Enhancing Convective Heat TransferIts Extension and Numerical Verifica-
4 Biswas, G., Torii, K., Fujii, D., and Nishino, K., 1996, Numerical and Ex- tions, Int. J. Heat Mass Transfer, 45, pp. 38493856.
perimental Determination of Flow Structure and Heat Transfer Effects of Lon- 12 Tao, W. Q., He, Y. L., Wang, Q. W., Qu, Z. G., and Song, F. Q., 2002, A
gitudinal Vortices in a Channel Flow, Int. J. Heat Mass Transfer, 39, pp. Unified Analysis on Enhancing Single Phase Convective Heat Transfer With
34413451. Field Synergy Principle, Int. J. Heat Mass Transfer, 45, pp. 48714879.
5 Liou, T. M., Chen, C. C., and Tsai, T. W., 2000, Heat Transfer and Fluid Flow 13 Qu, Z. G., Tao, W. Q., and He, Y. L., 2004, Three-Dimensional Numerical
in a Square Duct With 12 Different Shaped Vortex Generators, ASME J. Heat Simulation on Laminar Heat Transfer and Fluid Flow Characteristics of Strip
Transfer, 122, pp. 327335. Fin Surface With X-Arrangement of Strips, ASME J. Heat Transfer, 126, pp.
6 Gentry, M. C., and Jacobi, A. M., 2002, Heat Transfer Enhancement by 697707.
Delta-Wing-Generated Tip Vortices in Flat-Plate and Developing Channel 14 Ferrouillat, S., Tochon, P., Garnier, C., and Peerhossaini, H., 2006, Intensifi-
Flows, ASME J. Heat Transfer, 124, pp. 11581168. cation of Heat-Transfer and Mixing in Multifunctional Heat Exchangers by
7 Joardar, A., and Jacobi, A. M., 2008, Heat Transfer Enhancement by Winglet- Artificially Generated Streamwise Vorticity, Appl. Therm. Eng., 26, pp.
Type Vortex Generator Arrays in Compact Plain-Fin-and-Tube Heat Exchang- 18201829.

Journal of Heat Transfer SEPTEMBER 2009, Vol. 131 / 091903-9

Downloaded From: http://heattransfer.asmedigitalcollection.asme.org/ on 03/23/2013 Terms of Use: http://asme.org/terms