2011 IEEE Symposium on Industrial Electronics and Applications (ISIEA2011), September 25-28, 2011, Langkawi, Malaysia
Computational Fluid Dynamics Analysis of Shell-
and-double Concentric-tube Heat Exchanger
Tan Wee Choon, Mohd Shahril Shariff, Matthew Law
School of Mechatronic Engineering
Universiti Malaysia Perlis
Perlis, Malaysia
tweechoon@unimap.edu.my
Abstract— A shell-and-double concentric-tube heat exchanger is a
new invention in heat transfer devices that is used for transfer of
internal thermal energy between three fluids at different
temperatures. The structure of the heat exchanger for this project is
made of a shell which encloses sixty-six double concentric tubes.
There are three inlets and three outlets for all the fluids. The typical
shell-and-tube heat exchanger is not ideal in terms of its size. This
contributes to an increased cost of manufacturing and installation,
and on top of that, consumes a lot of space. This project is
conducted using Computational Fluid Dynamics software package
of EFD.Lab. At the end, the result in terms of the length of the heat
exchanger required to achieve the desired outlet fluid temperature
is compared with Kern method. It is found that all results have
close agreement with each other, with acceptable range of
percentage differences and errors.
Keywords – Heat Exchanger; Shell-and-double Concentric-
tube; Compuational Fluid Dynamics
I. INTRODUCTION
Over the years, the importance of heat exchangers has
increased immensely from the viewpoint of energy
conservation, conversion, recovery, and successful
implementation of new energy sources [1]. Presently, heat
exchangers are used for many purposes, ranging from radiators
in vehicles to the cooling of dairy products.
Heat exchangers have been categorised based on transfer
process, compactness, type of construction, and flow directions
[2]. The most common type of heat exchanger is the shell-and-
tube heat exchanger which is suitable to be used in high
pressure applications [3,4].
Oversized heat exchangers are largely avoided to minimise
space wastage. Therefore, the requirement for the heat
exchangers to be as small as possible is significant. Shell-and-
tube and double-tube heat exchangers existed for several
decades. These classic heat exchangers possess a major
drawback, in terms of the size of the tubes [5,6]. Normally, the
smaller and the lighter the heat exchanger, the better it is [7].
Therefore, there must be a large enough space to accommodate
these devices.
Other types of heat exchangers generally do not support
situations where high fluid flow rate and large temperature
Universiti Malaysia Perlis (UniMAP), Short Term Grant 9001-00197
978-1-4577-1417-7/11/$26.00 ©2011 IEEE
Lim Eng Aik
Engineering Mathematics Institute
Universiti Malaysia Perlis
Perlis, Malaysia
difference are involved. When these conditions are required,
the shell-and-tube heat exchanger can be used [8].
This concern is believed to have been overcome by
introducing double-concentric tubes instead of a single tube.
The difference is that the simple tubes are now replaced by
double concentric-tubes [5].
Kara and Guraras [9] developed a computer program for
the design of shell-and-tube heat exchangers. Design
parameters which are investigated are shell diameter, baffle
spacing, number of tube-side pass, and pressure drop. The
restriction of the program generally covers single-segmental
baffle having 25% baffle cut, and fixed tube sheet heat
exchangers with one-pass and two-pass for tube-side flow. The
tubes arrangement is restricted to triangular pitch layout.
Kumar et al. [10] did a Computer Fluid Dynamics (CFD)
analysis of a tube heat exchanger. It is found that the results
obtained from CFD simulation have close agreement with
experiments. In addition Ozden and Tari [11] performed a CFD
analysis on the shell-side of a shell-and-tube heat exchanger.
From the analysis performed, it is found that the results
calculated using the Kern method yielded very high percentage
of error.
Barman and Ghosal [12] designed a program in MATLAB
that analysed a shell-and-tube heat exchanger consisting of
longitudinally finned tubes. Their research concerns are
parameters such as optimum fin height, diameter for both
triangular and square pitch tube, corresponding pressure drop
and etc.
Recently there are few research activities being conducted
into possibility of involving up to three difference medium of
working fluid [13 – 15]. It is found that different tube materials
resist heat conduction differently in the transverse direction,
thus allowing the accumulation heat during transient process.
Besides that, tube walls also conduct heat longitudinally,
accounting for heat loss during the process.
II. WORKING PRINCIPLE
Consider the steady state flow of hot fluid in inlets 1 and 3,
and the steady state counter flow of cold fluid in inlet 2 as
shown in Figure 1. Heat transfer takes place between fluid 1 in
535
 2011 IEEE Symposium on Industrial Electronics and Applications (ISIEA2011), September 25-28, 2011, Langkawi, Malaysia

the inner tube and fluid 2 in the annulus passage, as well as TABLE III. ANNULUS PASSAGE PROPERTIES FOR SHELL-AND-DOUBLE
CONCENTRIC-TUBE HEAT EXCHANGER
between fluid 2 and fluid 3 in the shell side.
Properties Unit Value
Using three fluids in the heat exchanger provides better Hydraulic diameter mm 12.000
heat transfer efficiencies compared to shell-and-tube heat Rate of heat transfer W 338173
exchangers [10]. Principally, the third tube improves the heat Mass flow rate kg/s 8.090
transfer through an additional flow passage and a larger heat Cross section area mm2 263.894
transfer area per unit exchanger length [5, 16]. Fluid velocity m/s 0.466
Reynolds number - 6258
Prandtl number - 6.136
Friction factor - 0.009
Nusselt number - 53.129
Heat transfer coefficient W/(m2·°C) 2687.442
Pressure drop Pa 144.191

TABLE IV. TUBE-SIDE PROPERTIES FOR SHELL-AND-DOUBLE

CONCENTRIC-TUBE HEAT EXCHANGER

Properties Unit Value

Fluid velocity m/s 5.024
Reynolds number - 61568
Prandtl number - 1.962
Figure 1. Schematic diagram of shell-and-double concentric-tube heat Nusselt number - 212.798
exchanger Heat transfer coefficient W/(m2·°C) 35909.663
Pressure drop Pa 27475.849

III. COMPUTATIONAL FLUID DYNAMICS ANALYSIS

At the initialisation phase, the general system of units to be
used is the SI units and type of analysis will be internal flow
with heat conduction in solids. The default fluid to be used is
water and the default solid material of the heat exchanger will
be stainless steel. The properties of working fluid at inlet is
given in Table I, while properties of shell-side, annulus passage
and tube-side based on the research done by Bougriou and
Baadache [5] are shown in Table II to Table IV respectively.
TABLE I. WORKING CONDITIONS
With all the properties mentioned from Table II to Table
IV, the minimum length of shell-and-double concentric-tube
heat exchanger that is required to achieve the goal is 0.453 m
length.
Based on all the parameters listed, a study by computational
fluid dynamics software is conducted. The results of simulation
are shown in Figure 2 to Figure 4. The results from
computational fluid dynamics software, the working fluid
properties at the outlet of heat exchanger are given in Table V.

Value
Fluid properties Unit
Shell Annulus Tube
Cooling
Fluid type - Hot water Hot water
water
Volume flow rate m3/h 15 - 15
Mass flow rate kg/s - 8.090 -
Inlet temperature °C 100 20 100
Outlet pressure Pa 101325 101325 101325

TABLE II. SHELL-SIDE PROPERTIES FOR SHELL-AND-DOUBLE

CONCENTRIC-TUBE HEAT EXCHANGER

Properties Unit Value

Hydraulic diameter mm 17.062 Figure 2. Flow trajectories in the shell
Tube clearance mm 6.000
Bundle cross flow area mm2 337.000
Mass velocity kg/(m2·s) 1193.200
Reynolds number - 64630
Prandtl number - 1.965
Nusselt number - 199.448
Heat transfer coefficient W/(m2·°C) 7890.482
Friction factor - 0.217
Pressure drop Pa 15800.532

536
 2011 IEEE Symposium on Industrial Electronics and Applications (ISIEA2011), September 25-28, 2011, Langkawi, Malaysia

be constant. In reality, fluid properties are dependant on

temperature.

TABLE V. OUTLET TEMPERTURE OF WORKING FLUIDS

Kern Different
Properties CFD
Method (%)
Shell-side outlet temperature
93.28 80.00 ±16.60%
(oC)
Annulus passage outlet
26.82 30.00 ±10.60%
temperature (oC)
Tube-side outlet temperature
91.43 80.00 ±14.29%
(oC)

Figure 3. Flow trajectories in the annulus passage

It is observed that the Kern method under-predicts the outlet
temperature of the heat exchanger. According to Ozden and
Tari [11], the Kern method is a conservative approach;
therefore, the result is expected. The authors found that in
general, the error increases by increasing mass flow rate. Both
errors are accepted because it is within the permitted range, and
the mass flow rate can be considered as low. By comparing
both approach (Kern method and CFD simulation), the results
is shown in Table V.

Figure 4. Flow trajectories in the tube side
In Figure 2, baffles create turbulence in the flow by
constantly mixing up the fluid. The baffles are there to direct
the shell fluid pass the tubes in such a way that heat transfer is
enhanced [8]. Baffles induces fluid flow at right angles to the
tube axis, thus causes higher heat transfer.
There is an abnormality on the water temperature occurring
at the bottom flow channels of the annulus in Figure 3. A check
with the flow trajectory animation confirms that this
irregularity is actually caused by the reflow of water in the
annulus. This is due to a design flaw in the heat exchanger
model.
Figure 4 shows the flow in the tubes. There is nothing
wrong with the flow, except that the surface area of the front
and rear covers of the heat exchanger can be further decreased
so that the time of flow can be reduced.
Based on Kern method [17], the total length of the shell-
and-double concentric-tube heat exchanger required to perform
the cooling operation based on the working conditions given in
Table I is L = 0.453 m.
Based on the total length required for the shell-and-double
concentric-tube heat exchanger, the outlet temperatures for
working fluid at three locations: shell-side outlet, annulus
passage outlet and tube-side outlet which obtained by both
CFD simulation and Kern method are shown in Table V.
From Table V, it is clearly seen that both methods do not
obtain the same outlet temperature value. One of the reasons
for this is with the assumption of fluid properties is assuming to
IV. CONCLUSION
This project carried out a performance analysis of a shell-
and-double concentric-tube heat exchanger using
computational fluid dynamics approach. A model of the heat
exchanger is created in SolidWorks and analysed using
EFD.Lab.
By using Kern method, the length of the shell-and-double
concentric-tube heat exchanger needed to cool the hot water
from 100°C to 80°C is Lsdct = 0.453 m. Whereas the the outlets
of the heat exchanger are found to be at 80°C at the shell-side
outlet, 30°C at the annulus outlet, and averaging at 80°C at the
tube-side outlet.
While with the same length of shell-and-double concentric-
tube heat exchanger, the fluid temperatures at the outlets of the
heat exchanger are measured and are found to be at
approximately 93°C at the shell-side outlet, 27°C at the annulus
outlet, and averaging at 91°C at the tube-side outlet using
computational fluid dynamics analysis.
By comparing both methods, the maximum error of 16.6%
is observed to the outlet temperature of heat exchanger. This is
due to the assumption made within Kern method. In additional,
the error will be greater by increasing mass flow rate of
working fluid.
The performance of the shell-and-double concentric-tube
heat exchanger is strongly dependent upon its tube diameters.
Optimising this parameter will bring about a considerable
amount of savings in space, material, and cost when compared
with a shell-and-tube heat exchanger.
ACKNOWLEDGMENT
Authors are grateful for the financial support from
Universiti Malaysia Perlis (UniMAP) to conduct this research
work.

537
 2011 IEEE Symposium on Industrial Electronics and Applications (ISIEA2011), September 25-28, 2011, Langkawi, Malaysia

REFERENCES [9] Kara, Y.A. and Guraras, O., (2004). A Computer Program for Designing
of Shell-and-Tube Heat Exchangers, Applied Thermal Engineering, 24:
[1] Shah, R.K. and Sekulić, D.P., (2003). Fundamentals of Heat Exchanger p. 1797-1805.
Design, John Wiley & Sons, New Jersey.
[10] Kumar, V., et al., (2003). Cfd Analysis of Cross Flow Air to Air Tube
[2] Jarallah, S.S. and Quadir, G.A., (2009). Design and Fabrication of Type Heat Exchanger, PHOENICS 10th International User Conference.
Triple Concentric Pipe Heat Exchanger, Proceedings of International
Conference on Applications and Design in Mechanical Engineering [11] Ozden, E. and Tari, I., (2010). Shell Side Cfd Analysis of a Small Shell-
(iCADME), Penang, Malaysia. and-Tube Heat Exchanger, Energy Conversion and Management, 51: p.
1004-1014.
[3] Wikipedia.org, (2010). Shell and Tube Heat Exchanger,
http://en.wikipedia.org/wiki/Shell_and_tube_heat_exchanger, March 12, [12] Barman, J. and Ghoshal, A.K., (2007). Performance Analysis of Finned
2011. Tube and Unbaffled Shell-and-Tube Heat Exchangers, International
Journal of Thermal Sciences, 46: p. 1311-1317.
[4] Ennis, C.J., Botros, K.K., and Patel, C., (2011). Dynamic model for a
[13] García-Valladares, O., (2004). Numerical Simulation of Triple
heat exchanger tube rupture discharging a high-pressure flashing liquid
Concentric-Tube Heat Exchangers, International Journal of Thermal
into a low-pressure liquid-filled shell, Journal of Loss Prevention in the
Sciences, 43: p. 979-991.
Process Industries, 24: p. 111-121.
[5] Bougriou, C. and Baadache, K., (2010). Shell-and-Double Concentric- [14] Malinowski, L. and Bielski, S., (2009). Transient Temperature Field in
Tube Heat Exchangers, Heat Mass Transfer, 46: p. 315-322. a Parallel-Flow Three-Fluid Heat Exchanger with the Thermal
Capacitance of the Walls and the Longitudinal Walls Conduction,
[6] Abazar, V. A., Majid, A., (2011). Economic optimization of shell and Applied Thermal Engineering, 29: p. 877-883.
tube heat exchanger based on constructal theory, Energy, 36: p. 1087-
1096. [15] Unal, A., (1998). Theoretical Analysis of Triple Concentric-Tube Heat
Exchangers, Int. Comm. Heat Mass Transfer, 25(7): p. 949-958.
[7] Çengal, Y.A., (2006). Heat and Mass Transfer: A Practical Approach,
McGraw Hill, New York. [16] Jarallah, S.S. and Quadir, G.A., (2009). Analysis of a Three-Fluid
Cocurrent and Countercurrent Parallel Flow Heat Exchanger,
[8] Janna, W.S., (2000). Engineering Heat Transfer. 2nd edition, CRC Press Proceedings of International Conference on Applications and Design in
LLC, Florida. Mechanical Engineering (iCADME), Penang, Malaysia.
[17] Kern, D.Q., (1950). Process Heat Transfer, McGraw Hill, New York.

538