You are on page 1of 12

Research Article

Advances in Mechanical Engineering


2017, Vol. 9(8) 1–12
Ó The Author(s) 2017
Numerical simulation on a novel DOI: 10.1177/1687814017717665
journals.sagepub.com/home/ade
shell-and-tube heat exchanger with
screw cinquefoil orifice baffles

Xuankai Zhang, Dong Han, Weifeng He, Chen Yue and Wenhao Pu

Abstract
A novel shell-and-tube heat exchanger with screw cinquefoil orifice baffles is designed to grasp the weakness of the
traditional shell-and-tube heat exchanger with cinquefoil orifice baffles. It specifically enhances the heat transfer coeffi-
cient in the area between adjacent baffles and enhances the shell-side fluid flushing ability on bundles. In the proposed
shell-and-tube heat exchanger with screw cinquefoil orifice baffles, screw-type cinquefoil orifice baffles are installed in
the shell side. Shell-and-tube heat exchanger with screw cinquefoil orifice baffle is compared with shell-and-tube heat
exchanger with cinquefoil orifice baffles and the traditional shell-and-tube heat exchanger with segmental baffles by
means of numerical simulations. The numerical result shows that the heat transfer coefficient and shell-side fluid flushing
ability in the shell-and-tube heat exchanger with screw cinquefoil orifice baffle is higher than that in the shell-and-tube
heat exchanger with cinquefoil orifice baffles and shell-and-tube heat exchanger with segmental baffles, for the shell-side
fluid that is urged to flow in approximately continuous helical flow. Under the same shell-side mass flow rate, heat trans-
fer coefficient of shell-and-tube heat exchanger with screw cinquefoil orifice baffle is about 9.2% higher than that of
shell-and-tube heat exchanger with cinquefoil orifice baffles and shell-and-tube heat exchanger with segmental baffles by
about 5.4% on average. The article presents a novel design thought when researchers design heat exchangers.

Keywords
Computational fluid dynamics, heat exchanger, heat and mass transfer, numerical analysis, cinquefoil orifice baffle

Date received: 6 November 2016; accepted: 2 June 2017

Academic Editor: Oronzio Manca

Introduction The STHX with orifice baffles was originally pro-


posed by Phillips Petroleum Company,5 which has been
Shell-and-tube heat exchangers (STHXs) have been widely used as heat exchanger for steam generator of
widely used in the chemical engineering, aviation indus- nuclear power plants.6,7 In the STHX with orifice baf-
try, and so on for their wide range of allowable design fles, orifice plates have been developed as a support to
pressures and temperatures.1 However, shell-and-tube tubes, and the shell-side fluid flows longitudinally
heat exchanger with segmental baffles (STHX-SG), the through the gaps between the orifice edges and tube
most widely used, has many disadvantages. For exam-
ple, the shell-side fluid in STHX-SG flows across tube
bundles in a zigzag manner, which could lead to a great College of Energy and Power Engineering, Nanjing University of
flow resistance and a high vibration level. Other issues Aeronautics and Astronautics, Nanjing, China
like the ‘‘dead’’ flow region and fouling, which cause
Corresponding author:
low heat transfer efficiency,2–4 also trouble such heat Dong Han, College of Energy and Power Engineering, Nanjing University
exchanger users. To change this situation, many novel of Aeronautics and Astronautics, Nanjing 210016, China.
STHXs have been proposed. Email: handong@nuaa.edu.cn

Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License
(http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without
further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/
open-access-at-sage).
2 Advances in Mechanical Engineering

walls. This heat exchanger not only has a lower pres- orifice baffles. The heat transfer performance in the
sure drop, but also being less liable to fouls, eliminates area between adjacent baffles is still not satisfactory.
stagnant recirculation zones and avoids flow-induced Although there appear various types of orifice baffles
vibration compared to the conventional STHX-SG.6,8 in recent years, design of new shapes of the orifice
Because of these advantages, the STHX with orifice attracts research interests. The aforementioned prob-
baffles attracted considerable attention in recent years. lem has not been resolved, which limits the develop-
A El Maakoul et al.1 using computational fluid ment of STHX-COB.
dynamics (CFD) to simulated three STHXs with the While the shell-and-tube heat exchanger with helical
recently developed trefoil-hole, helical baffles, and the baffles (STHX-HB) shows good thermo-hydraulic per-
conventional segmental baffles, respectively, at low formance and low vibration level.9 STHX-HB was first
shell-side flow rates. And the results indicated that heli- proposed by Lutcha and Nemcansky.10 They proposed
cal baffles result in higher thermo-hydraulic perfor- the helical baffles in STHX-HB are the most important
mance while trefoil-hole baffles have a higher heat factor influencing heat transfer, which directly leads to
transfer performance with large pressure drop com- the shell-side fluid to flow in approximately continuous
pared to segmental baffles. Y You et al.6 based on the helical flow. And they found that the helical baffles
experimental results found that trefoil-hole baffles could force the shell-side fluid to approach plug flow,
could generate high-speed flush, intensive recirculation which increased the average temperature driving force.
flow, and high turbulence intensity level, which lead the The flow patterns induced by the baffles also intensified
Nusselt number of the shell side is about 4.5 times than the shell-side heat transfer remarkably.11 Generally
without baffles. Y You et al.8 studied the effect of baf- speaking, there are two kinds of helical baffles, contin-
fle number and baffle distance on the small-size heat uous helical baffles and discontinuous helical baffles.
exchanger with trefoil-hole baffles. The convection heat Considering the difficulty in the manufacture of contin-
transfer coefficient on the shell side of the small-size uous helical baffles, the discontinuous helical baffles
heat exchanger monotonically drops while the shell side formed by overlapped fans or oval-shaped plates to
overall thermo-hydraulic performance monotonically replace it now. But the leakage by discontinuous helical
rises. There are many types of orifice baffles, such as baffle is relatively large due to the triangle zones, which
trefoil-hole baffles, quatrefoil-hole baffles, and cinque- will reduce the heat transfer performance.12 S Wang
foil orifice baffles, but very few related academic litera- et al.11 proposed folded helical baffles which can block
ture are available.6 Figure 1 depicts the sketch of some the triangle leakage zones between two adjacent plain
familiar orifice baffles. And the shell-and-tube heat baffles in STHX-HB. And the experimental results
exchanger with cinquefoil orifice baffles (STHX-COB) showed that the integrative performance of STHX-HB
was the choice for further study. is greatly enhanced by the improved folded baffles.
However, the enhanced heat transfer region in To solve above problems, through using STHX-
STHX-COB is focused on the vicinity of cinquefoil COB as the foundation, introducing the concept of the
helical flow in STHX-HB, a new type of heat exchan-
ger, shell-and-tube heat exchanger with screw cinque-
foil orifice baffles (STHX-SCOB), was proposed.
STHX-SCOB provides a new choice for industry. In
this article, STHX-SCOB was simulated by CFD
method. The shell-side flow patterns of STHX-SG,
STHX-SCOB, and STHX-COB were compared
numerically. Moreover, impacts from two significant
parameters, including the helical angle and baffle thick-
ness, on the heat transfer and pressure drop perfor-
mance of STHX-SCOB, were studied.

Mathematical modeling of the shell-side


flow field
Physical models
The physical model of STHX-SG is presented in
Figure 2(a) and (d). In STHX-SG, the baffle thickness
Figure 1. The sketch of some familiar orifice baffles: (a) trefoil- is 3 mm, and the baffle cut is 36%. The baffle number
hole baffle, (b) cinquefoil orifice baffle, (c) rectangular orifice is 4. And the baffle pitch is 117 mm. More details are
baffle, and (d) small round hole orifice baffle. listed in Table 1.
Zhang et al. 3

Table 1. Common geometry parameters for three models.

Item Dimensions

Baffle pitch, B (mm) 166


Effective tube length, L (mm) 600
Shell diameter, Di (mm) 95
Tube central distance, Dt (mm) 30
Tube outside diameter, dt (mm) 20

2. The leak flows between baffle and the shell are


neglected;
3. The tube wall temperatures are kept constant in
the whole shell side;
4. The natural convection induced by the fluid
density variation is neglected;
5. The heat exchanger is assumed well insulated,
and the heat loss to the environment is totally
neglected.8

Governing equations and boundary conditions


The renormalization group (RNG) k-e model of
Yakhot and Orszag is adopted in the simulation
because the model provides improved predictions of
near-wall flows and be chosen by many researchers.13
Figure 2. The shape of three heat exchangers: (a) graphical
The RNG k-e model was derived by a statistical tech-
model of segmental baffle in STHX-SG, (b) graphical model of
screw-type cinquefoil orifice baffle, (c) graphical model of
nique called renormalization method, which is widely
cinquefoil orifice baffle, (d) graphical model of original tube used in industrial flow and heat transfer because of its
bundle in STHX-SG, (e) two significant parameters which economy and accuracy.14 The governing equations for
influent heat transfers performance in STHX-SCOB, and continuity, momentum, energy, k, and e in the compu-
(f) graphical model of original tube bundle in STHX-COB. tational domain can be expressed as follows:
Continuity equation
In STHX-SCOB, screw-type cinquefoil orifice baffles
are substituted for plain baffles to be installed in the ∂
shell side. The original straight flow path was replaced ðrui Þ = 0 ð1Þ
∂xi
by helical flow path. The shapes of two kinds of baffles
are shown in Figure 2(b) and (c), respectively. Momentum equation
As shown in Figure 2(e), the helical angle a refers  
the angle between the central axis and the channel line ∂ ∂ ∂uk ∂p
ðrui uk Þ = m  ð2Þ
tangent. In this figure, a at 38° and the baffle thickness ∂xi ∂xi ∂ui ∂xk
with Db at 8 mm is designed. Meanwhile, the baffle
Energy equation
thickness is 3 mm in STHX-COB. The arrangement of
tube bundle is regular triangle in both heat exchangers.  
∂ ∂ k ∂t
The simulated STHX-SCOB has three baffles in the ðrui tÞ = ð3Þ
∂xi ∂xi CP ∂xi
shell-side direction with a total tube number of 7, and
the same situation is designed for the referenced STHX- Turbulent kinetic energy equation
COB. Cinquefoil orifice angle b is 40°, and outer dia-  
meter of orifice do is 28 mm in both STHX-SCOB and ∂ ∂ ∂ ∂k
ðrk Þ + ðrkui Þ = ak meff + Gk  re ð4Þ
STHX-COB. Other common geometry parameters of ∂t ∂xi ∂xj ∂xj
two models are listed in Table 1.
To simplify numerical simulation while still keep the Turbulent dissipation energy equation
basic characteristics of the process, following assump-  
tions are made: ∂ ∂ ∂ ∂e
ðreÞ + ðreui Þ = ae meff
∂t ∂xi ∂xj ∂xj
ð5Þ
1. The fluid-flow and heat transfer processes are e e2
+ Cle Gk  C2e r
turbulent and in steady state; k k
4 Advances in Mechanical Engineering

where rus de
2 Re = ð12Þ
k ms
meff = m + mt , mt = rCm ,
  e
h ð6Þ where ms is the dynamic viscosity is the shell side, and
h 1h de is the equivalent diameter. So, the expression used to
Cle = C1e 
0

1 + bh3 calculate Reynolds number in STHX-SG


12 k 1 2m
h = 2Eij  Eij , Eij =   ð7Þ Re = ð13Þ
e ∂ui
2 ∂xj +
∂uj ms ½ðNcl + 1ÞB + ðDi  Ncl dt Þ
∂xi
About STHX-SCOB and STHX-COB, the following
The empirical constants for the RNG k-e model are expression is used to calculate Reynolds number
assigned the following values
4m
Cm = 0:0845, C1e = 1:42, C2e = 1:68 ð8Þ Re = ð14Þ
ms pðDi + Ndt Þ
b = 0:012, h0 = 4:38, ak = ae = 1:39 ð9Þ N is the tube number.
The shell-side inlet pipe is set as velocity inlet, while
outlet pipe is set as pressure outlet. Water is used as the Heat transfer rate and overall heat transfer coefficient. Heat
working fluid, and its thermal properties depend on its transfer rate of the shell-side fluid is determined by
operation temperature. The temperature of inlet is
360 K and the pressure of outlet is 0 atm. The reference [ = mcp ðts, in  ts, out Þ ð15Þ
pressure is 1 atm. The wall temperatures of the tubes
In above equations, m is the shell-side mass flow rate,
are uniform and fixed to 300 K. Other shell walls are
cp is the fluid specific heat capacity, ts, in is the inlet tem-
non-slip, impermeable, and adiabatic. The computer
perature at shell side, and ts, out is the average outlet
code CFX is used to simulate the flow and heat trans-
temperature at shell side. And heat transfer coefficient
fer. The governing equations are iteratively solved by
is defined by
the finite-volume method with semi-implicit method for
pressure-linked equations (SIMPLE) pressure–velocity Dtmax = ts, in  tw ð16Þ
coupling algorithm. The advection scheme chose high
resolution for calculation. The convergence criterion is Dtmin = ts, out  tw ð17Þ
that the normalized residuals are less than 106 for the Dtmax  Dtmin
flow equations and 108 for the energy equation. The Dtm = ð18Þ
lnðDtmax =Dtmin Þ
computations were performed with a work station with
CPU frequency of 4.8 GHz, and it usually took approx- A = N pdt L ð19Þ
imately 24 h for every case. [
h= ð20Þ
ADtm
Data reduction Dtm is the log mean temperature difference, A means
Shell-side velocity and Reynolds number. The mean shell-side the heat exchange area based on the outer diameter of
fluid velocity is defined by the tubes, L is the effective length of the tubes, N is the
tubes number, and dt is the outer diameter of the tubes.
m
us = ð10Þ Nusselt number, friction coefficient, and comprehensive
rAc performance
hde
us is the shell-side average velocity. r is the fluid density Nu = ð21Þ
l
and Ac is the cross-flow area at the shell centerline.1 For
the segmental baffles15 f=
2de Dp
ð22Þ
rLu2i
Ac = ðDi  Ncl dt ÞB ð11Þ
In above equations, B is baffle pitch, Dp is the total
Here, Di is the internal diameter of the shell, dt is the pressure drop in shell side, l is the thermal conductiv-
external diameter of the tubes, B is the baffle spacing, ity, de is the equivalent diameter of shell side, and ui is
and Ncl is the number of tubes in the central row. the inlet average velocity, according to Yang and Liu.16
With the mean velocity value, the Reynolds number In order to evaluate the comprehensive performance,
for the shell side is determined by the parameter CP = Nu  f 1=3 14,17 was introduced.
Zhang et al. 5

Figure 3. Meshes of computational model: (a) meshes of


STHX-COB and (b) meshes of STHX-SCOB (a = 38º,
Db = 8 mm).

Model validation
Figure 4. STHX-SCOB results of different grid systems
Grid generation and independence. The three-dimensional (a = 38º, Db = 8 mm, m = 0.375 kg/s).
(3D) geometry was created in unigraphics next genera-
tion, which next meshed by tetrahedral and hexahedral
grids in the commercial code integrated computer engi- Figure 5 shows the simulation results by the method
neering and manufacturing code for computational fluid in this article compared with the results in Ozden and
dynamics (ICEM). Because ICEM has excellent merit on Tari.20 In addition, the results in Ozden and Tari20 are
managing very complex 3D geometries (see Figure 3). considered satisfactory by MG Yehia et al.21 The maxi-
According to Zhu et al.,18 the same method that the tet- mum relative deviation for heat transfer coefficient h is
rahedral mesh was used in the region near to the cinque- 5%, and the shell-side pressure drop Dp has the maxi-
foil orifice baffles and the hexahedral mesh is used in the mum relative deviation of 3%. Therefore, it can be con-
region outside of the cinquefoil orifice baffles is applied cluded that this model could gives a good prediction for
in the research, and the simulation results by this method heat transfer characteristics.
coincide well with the experimental results.
For make sure the accuracy of simulations, accord-
ing to Wang et al.12 and Yang et al.,19 the grid indepen- Results and discussion
dence of the numerical solutions was conducted on
STHX-SCOB, STHX-COB, and STHX-SG. Flow distributions in different heat exchangers
Four different grid systems are generated for the The flow behavior in STHX-SG is shown in Figure 6.
STHX-SCOB (a = 38°, Db = 8 mm, m = 0.375 kg/s) The streamline pattern in shell side is a zigzag pattern,
and their results are shown in Figure 4. Under the oper- which causes large dead zones. The region back of the
ation conditions of G3 and G4, differences of h=Dp and baffles has eddy formation and fluid recirculation, caus-
Dp were less than 2%. Considering both convergent ing a large amount of energy spent in this region.
time and solution precision, the grid quantities of The flow distributions in shell side of STHX-SCOB
STHX-SCOB, STHX-COB, and STHX-SG were chose and STHX-COB are totally different. Figure 7 shows
as 3,668,432, 3,086,673, and 3,269,825, respectively. the distribution of flow lines in shell-side velocity flow
field of these two heat exchangers when the shell-side
mass flow rate m is 0.375 kg/s. As shown in Figure 7,
Numerical investigations compared with other papers. The due to the sudden decrease in flow passage area at baf-
verified STHX-SG model geometry was made exactly fles, the shell fluid velocity increases, which results in
as Ozden and Tari20 and MG Yehia et al.21 for it is sim- jet effect and destruction to the boundary layer in two
ilar to the STHXs being studied in this article. The veri- heat exchangers. Besides, in Figure 7(c), the secondary
fied STHX-SG baffle number is 6, and the baffle cut flow can be found in local enlarging graph, which dis-
ratio is 36%, while other model geometrical parameters turbs the shell fluid and enhances the heat transfer.
are presented in Ozden and Tari20 and Yehia et al.21 While in Figure 7(b), due to the effect of the helical
According to the literature value, water is the working angle, the tube bundles are washed out by revolving
fluid. The shell inlet temperature is 300 K, and tube wall fluid. As shown in Figure 8(b), the shell flow velocity in
temperature is 450 K. The shell outlet is pressure outlet. STHX-SCOB is higher, especially in peripheral side
While the turbulence model is RNG k-e model, and tubes’ wall. The spiral motion brings about good mix-
pressure–velocity coupling is SIMPLE. The CFD pack- ing, which directly leads to improvement of heat trans-
age CFX is used to simulation. After grid independency fer. Moreover, a higher flow velocity enhances the
check, the number of cells is 2,841,973. shell-side fluid flushing ability. The new type of heat
6 Advances in Mechanical Engineering

Figure 5. The comparison between the result in Ozden and Tari20 and the simulation result by the proposed method: (a) variation
of h with tube side inlet flow rate m in comparison with Ozden and Tari and the results by the proposed method and (b) variation of
Dp with tube side inlet flow rate m in comparison with Ozden and Tari and the results by the proposed method.

Figure 6. (a) Flow distributions in STHX-SG (m = 0.375 kg/s) and (b) local magnification in STHX-SG.

exchanger is more like helical baffle heat exchanger in Pressure drop and comprehensive performance
some way. The relationship between the shell-side pressure drop
Dp and m is shown in Figure 10(a). The results of the
simulation indicate that a high Dp in STHX-SCOB is
Heat transfer coefficient obtained compared to that in STHX-COB and STHX-
Heat transfer coefficient h is a very significant para- SG. It is clear that the pressure drop of the STHX-
meter in design of industrial STHXs because it is related SCOB is about 15.4% on average higher than that of
to the area of land occupied by the equipment and the the STHX-SG and 36.0% on average higher than that
cost of production materials. Therefore, h is obtained of the STHX-COB. The reason for this phenomenon is
more attention in some cases. that helical flow path on screw-type cinquefoil orifice
Figure 9 illustrates relationship between h and m. baffles changes the direction of shell-side fluid flow and
In Figure 9, h of STHX-SCOB is obviously higher causes substantial dissipations of mechanical energy.
than that of STHX-COB by about 9.2% and STHX- But the pressure drop is not the only standard to
SG by about 5.4% on average. With m of 0.15 kg/s, h judge the performance of the heat exchanger.
in the STHX-SCOB is higher than that of the STHX- Comprehensive performance, CP = Nu  f 1=3 , of the
COB by 15.4%; but when m increases to 1.5 kg/s, the three heat exchangers is shown in Figure 10(b). It can
difference drops down to 6.9%. The reason for this be seen from Figure 10(b) that CP in STHX-SCOB is
phenomenon is that in STHX-SCOB turbulence is 3.0% higher than that of STHX-COB with Re at 2322,
stronger than STHX-SG and STHX-COB, which and the difference drops down to 0.3% with Re at
leads to a higher heat transfer coefficient, particularly 11,610. However, when Re increases to 23,220, CP in
at low m. STHX-SCOB is 4.0% lower than that in STHX-COB.
Zhang et al. 7

Figure 7. Flow distributions in different heat exchangers (m = 0.375 kg/s): (a) flow distributions in STHX-COB, (b) flow
distributions in STHX-SCOB (a = 38º, Db = 8 mm), (c) local magnification in STHX-COB, and (d) local magnification in STHX-SCOB
(a = 38º, Db = 8 mm).

Figure 8. Velocity contours in different heat exchangers (m = 0.375 kg/s, Z = 200 mm): (a) velocity contours in STHX-COB and
(b) velocity contours in STHX-SCOB.

It can be inferred from above that the heat transfer per- Effects of structural parameters on STHX-SCOB
formance in STHX-SCOB is higher than that in
The helical angle of helical flow paths. According to the
STHX-COB at low Re number. Moreover, CP of
design thought about helical baffle heat exchanger,22,23
STHX-SG is higher than that of STHX-SCOB and
the helical angle a impacts heat transfer and flow char-
STHX-COB. It should be emphasized that this result is
acteristics of the STHX-SCOB significantly. STHX-
under the same mass flow rate, the fluid velocity of
SCOB with different helical angles (a = 27°, 38°, and
STHX-SG is much lower than those of STHX-SCOB
46°) was studied. As shown in Figure 11, h and Dp
and STHX-COB, leading to significant increase in CP
increase as a is increased. From Figure 11(d), h in
of STHX-SG at the same Reynolds number.
8 Advances in Mechanical Engineering

STHX-SCOB with a at 46° is of 36.9% higher than


that of a at 38°. Similarly, from Figure 11(e), Dp in
STHX-SCOB with a at 38° is 31.6% higher than that
with a at 27° on average. Then, about comprehensive
performance, CP in STHX-SCOB with a at 46° is
4.6% lower than that with a at 38° on average. Then,
CP in STHX-SCOB with a at 38° is 2.3% lower than
that a at 27° on average.
Above phenomenon can be explained that the shell
fluid radial velocity increases after flowing through heli-
cal flow paths. The bigger the a gets, the bigger the shell
fluid radial velocity becomes. However, the axial velo-
city of shell fluid is small at high a. Therefore, it brings
about good mixing, which directly leads to the strong
heat transfer. However, the energy loss is big.
Figure 9. Heat transfer coefficient versus the shell-side mass
flow rate.
The baffle thickness. Due to the novel baffle changes the
shell fluid distribution, both h and Dp in STHX-SCOB
are higher than that in STHX-COB. In addition to a,
the baffle thickness Db is another significant parameter
in STHX-SCOB, and its effects on the heat transfer
and flow characteristics are analyzed in this part. The
conclusion can be obtained that the streamline is more
crowded in flow field when the baffle thickness
increases by comparing the pictures of Figure 12(a)–(c).
From Figure 12(d), h in the STHX-SCOB with Db
at 3 and 8 mm are 15.8% and 8.3% lower than that
with Db at 13 mm on average, respectively. Besides, Dp
in the STHX-SCOB with Db at 3 and 8 mm is 79.5%
and 35.7% lower than that with Db at 13 mm on aver-
age, respectively. Then, CP in the STHX-SCOB with
Db at 3 and 8 mm is 5.1% and 2.3% higher than that
with Db at 13 mm on average, respectively. In conclu-
sion, the thicker the Db , the more intense the distur-
bance of the shell-side fluid, leading to a greater energy
loss, and then more the helical flow paths of the fluid
velocity at the shell side.

Conclusion
In this article, a novel STHX-SCOB with helical flow
paths is proposed to improve heat transfer performance
at shell side. When the traditional STHX-COB and
STHX-SG are compared, considering the impacts from
the structure parameters of screw-type cinquefoil orifice
baffle, heat transfer and pressure drop performance at
shell side is investigated, and the most significant con-
clusions are summarized as follows:
Figure 10. Other heat transfer performances: (a) pressure
drop versus the shell-side mass flow rate and (b) comprehensive 1. As a novel heat exchanger, STHX-SCOB pro-
performance versus the shell-side Reynolds number. vides a new heat transfer enhancement choice.
Besides, the screw structure in the STHX-SCOB
STHX-SCOB with a at 46° is 6.2% higher than that of improves fluid flushing ability on the shell side.
a at 38° on average. Similarly, h in STHX-SCOB with The screw-type cinquefoil orifice baffles enhance
a at 38° is 7.1% higher than that of a at 27°. But Dp in the heat transfer considerably, and this
Zhang et al. 9

Figure 11. Heat transfer performances with different helical angles: (a) velocity flow distribution (a = 27º, m = 0.375 kg/s,
Db = 8 mm), (b) velocity flow distribution (a = 38º, m = 0.375 kg/s, Db = 8 mm), (c) velocity flow distribution (a = 46º, m = 0.375 kg/s,
Db = 8 mm), (d) heat transfer coefficient versus the shell-side mass flow rate, (e) pressure drop versus the shell-side mass flow rate,
and (f) comprehensive performance versus the shell-side Reynolds number.

enhancement is done at the expense of a large angle of a at 38° and baffle thickness of Db at
pressure drop. So next, other screw-type orifice 8 mm in STHX-SCOB.
baffle heat exchangers should be researched to 3. Under the same shell-side mass flow rate m and
obtain the optimal structure parameters for Db at 8 mm, the bigger the a, the higher the h
other screw-type orifice baffle heat exchanger. and Dp. The heat transfer coefficient h of
2. Under the same shell-side mass flow rate m, STHX-SCOB with a at 46° is 6.2% higher than
heat transfer coefficient h of STHX-SCOB is that with a at 38°, while the pressure drop Dp
about 9.2% higher than that of STHX-COB of STHX-SCOB with a at 46° is about 36.9%
and STHX-SG by about 5.4% on average, higher than that with a at 38°. Similarly, the
while the pressure drop Dp of STHX-SCOB is heat transfer coefficient h of STHX-SCOB with
about 36.0% higher than that of STHX-COB a at 38° is about 7.1% higher than that with a
and STHX-SG by about 15.4% with the helical at 27°, while the pressure drop Dp of STHX-
10 Advances in Mechanical Engineering

Figure 12. Heat transfer performances with different baffle thicknesses: (a) temperature flow distribution (Db = 3 mm,
m = 0.375 kg/s, a= 46º), (b) temperature flow distribution (Db = 8 mm, m = 0.375 kg/s, a= 46º), (c) temperature flow distribution
(Db = 13 mm, m = 0.375 kg/s, a= 46º), (d) heat transfer coefficient versus the shell-side mass flow rate, (e) pressure drop versus the
shell-side mass flow rate, and (f) comprehensive performance versus the shell-side Reynolds number.

SCOB with a at 38° is about 31.6% higher than Declaration of conflicting interests
that with a at 27°. The author(s) declared no potential conflicts of interest with
4. Under the same shell-side mass flow rate m and respect to the research, authorship, and/or publication of this
a at 46°, the bigger the Db , the higher h and Dp article.
are. The heat transfer coefficient h in the
STHX-SCOB with Db at 3 and 8 mm are 15.8%
Funding
and 8.3% lower than that with Db at 13 mm on
average, respectively. And the pressure drop Dp The author(s) disclosed receipt of the following financial sup-
in the STHX-SCOB with Db at 3 and 8 mm is port for the research, authorship, and/or publication of this
article: This work was supported by the Program of National
79.5% and 35.7% lower than that with Db at
Natural Science Foundation of China (No. 51506090).
13 mm on average, respectively.
Zhang et al. 11

References overlapped helical baffles and continuous baffles—part I:


1. El Maakoul A, Laknizi A, Saadeddine S, et al. Numeri- numerical model and results of whole heat exchanger
cal comparison of shell-side performance for shell and with middle-overlapped helical baffles. Int J Heat Mass
Tran 2009; 52: 5371–5380.
tube heat exchangers with trefoil-hole, helical and seg-
18. Zhu LY, Lang HF, Zhou G, et al. Numerical simulation
mental baffles. Appl Therm Eng 2016; 109: 175–185.
on shell side fluid flow and heat transfer in heat exchan-
2. Li HD and Kottke V. Effect of the leakage on pressure
ger with trefoil-baffles. CIESC J 2014; 65: 829–835 (in
drop and local heat transfer in shell-and-tube heat
Chinese).
exchangers for staggered tube arrangement. Int J Heat
19. Yang JF, Zeng M and Wang QW. Numerical investiga-
Mass Tran 1998; 41: 425–433.
tion on combined single shell-pass shell-and-tube heat
3. Gu X, Liu B, Wang Y, et al. Heat transfer and flow resis-
exchanger with two-layer continuous helical baffles. Int J
tance performance of shutter baffle heat exchanger with
Heat Mass Tran 2015; 84: 103–113.
triangle tube layout in shell side. Adv Mech Eng 2016; 8:
20. Ozden E and Tari I. Shell side CFD analysis of a small
1–8.
shell-and-tube heat exchanger. Energ Convers Manage
4. Guo J, Xu M and Cheng L. The application of field
2010; 51: 1004–1014.
synergy number in shell-and-tube heat exchanger optimi-
21. Yehia MG, Attia AA, Abdelatif OE, et al. Computa-
zation design. Appl Energ 2009; 86: 2079–2087.
tional investigations of thermal simulation of shell and
5. Gentry CC. Rod baffle heat exchanger technology. Chem
tube heat exchanger. In: Proceedings of the ASME 2014
Eng Prog 1990; 86: 48–56.
12th biennial conference on engineering systems design and
6. You Y, Fan A, Lai X, et al. Experimental and numerical
analysis, Copenhagen, 25–27 July 2014, p.V003T12A001.
investigations of shell-side thermo-hydraulic perfor-
New York: ASME.
mances for shell-and-tube heat exchanger with trefoil-
22. Yang JF, Lin YS, Ke HB, et al. Investigation on com-
hole baffles. Appl Therm Eng 2013; 50: 950–956.
bined multiple shell-pass shell-and-tube heat exchanger
7. Zhou GY, Xiao J, Zhu L, et al. A numerical study on
with continuous helical baffles. Energy 2016; 115:
the shell-side turbulent heat transfer enhancement of 1572–1579.
shell-and-tube heat exchanger with trefoil-hole baffles. 23. Lei YG, He YL, Li R, et al. Effects of baffle inclination
Enrgy Proced 2015; 75: 3174–3179. angle on flow and heat transfer of a heat exchanger with
8. You Y, Chen Y, Xie M, et al. Numerical simulation and helical baffles. Chem Eng Process 2008; 47: 2336–2345.
performance improvement for a small size shell-and-tube
heat exchanger with trefoil-hole baffles. Appl Therm Eng
2015; 89: 220–228. Appendix 1
9. Zhang JF, Li B, Huang WJ, et al. Experimental perfor-
mance comparison of shell-side heat transfer for shell- Notation
and-tube heat exchangers with middle-overlapped helical
A heat transfer area (m2 )
baffles and segmental baffles. Chem Eng Sci 2009; 64:
1643–1653.
Ac cross-flow area at the shell centerline (m2 )
10. Lutcha J and Nemcansky J. Performance improvement of B baffle pitch (m)
tubular heat exchangers by helical baffles. Chem Eng Res Cp specific heat capacity (J/(kg K))
Des 1990; 68: 263–270. CP comprehensive performance
11. Wang S, Wen J, Yang H, et al. Experimental investiga- de equivalent diameter (m)
tion on heat transfer enhancement of a heat exchanger do outer diameter of orifice (mm)
with helical baffles through blockage of triangle leakage dt tube outside diameter (mm)
zones. Appl Therm Eng 2014; 67: 122–130. Db baffle thickness (mm)
12. Wang Q, Chen Q, Chen G, et al. Numerical investigation Di internal shell diameter (mm)
on combined multiple shell-pass shell-and-tube heat Dt tube central distance (mm)
exchanger with continuous helical baffles. Int J Heat f friction coefficient
Mass Tran 2009; 52: 1214–1222.
h average heat transfer coefficient
13. Yakhot V and Orszag SA. Renormalization-group analy-
(W=m2 K)
sis of turbulence. Phys Rev Lett 1986; 57: 1722–1724.
14. Jian W, Huizhu Y, Wang S, et al. Numerical investigation k turbulent fluctuation kinetic energy
on baffle configuration improvement of the heat exchan- (m2 =s2 )
ger with helical baffles. Energ Convers Manage 2015; 89: L total effective tube length (mm)
438–448. m shell side mass flow rate (kg/s)
15. Eduardo C. Heat transfer in process engineering. New N number of tubes
York: McGraw-Hill Publishing Company, 2009. Ncl number of tubes in the central row
16. Yang J and Liu W. Numerical investigation on a novel Nu Nusselt number
shell-and-tube heat exchanger with plate baffles and Dp pressure drop (Pa)
experimental validation. Energ Convers Manage 2015; Re Reynolds number
101: 689–696. Dtm logarithmic mean temperature
17. Zhang JF, He YL and Tao WQ. 3D numerical simula-
difference (K)
tion on shell-and-tube heat exchangers with middle-
12 Advances in Mechanical Engineering

ts, in , temperature of inlet tube and outlet e turbulent kinetic energy dissipation rate
ts, out tube (K) (m2 =s3 )
tw tube wall temperature (K) m dynamic viscosity (kg/(m s))
u, v, w velocities in different directions (m/s) mt turbulent dynamic viscosity (kg/(m s))
ui inlet average velocity (m/s) l thermal conductivity (W/(m K))
us shell side average velocity (m/s) r fluid density (kg=m3 )
x, y, z Cartesian coordinate [ heat transfer quantity (W)
a helical angle of helical flow paths (°)
b cinquefoil orifice angle (°)