Numerical Analysis of Shell and Tube Heat Exchanger With Different Baffle Parameters Using CFD

Download as pdf or txt
Download as pdf or txt
You are on page 1of 47

Numerical Analysis of Shell and Tube Heat Exchanger with Different

Baffle Parameters using CFD

By

Md. Samiuzzaman Real


Roll No. 1605015

Khulna University of Engineering & Technology


Khulna-9203, Bangladesh
May, 2022
Numerical Analysis of Shell and Tube Heat Exchanger with Different
Baffle Parameters Using CFD

By

Md. Samiuzzaman Real

Roll: 1605015

A thesis report submitted to the Department of Mechanical Engineering of Khulna University


of Engineering & Technology in partial fulfillment of the requirements for the degree of
Bachelor of Science in Mechanical Engineering

Khulna University of Engineering & Technology


Khulna 9203, Bangladesh
May, 2022
Declaration

This is to certify that this thesis work entitled “Numerical Analysis of Shell and Tube Heat
Exchanger with Different Parameters using CFD” has been carried out by Md. Samiuzzaman
Real, Roll No. 1605015 in the Department of Mechanical Engineering, Khulna University of
Engineering & Technology, Khulna 9203, Bangladesh. Any part of this work has not been
submitted for the award of any degree or diploma

Signature of Supervisor Signature of Student

Dr. Md. Shahidul Islam Md Samiuzzaman Real


Professor & Head Roll: 1605015
Dept. of Mechanical Engineering
Khulna University of Engineering & Technology

i
Approval

This is to certify that the thesis work submitted by Md. Samiuzzaman Real (Roll No. 1605015)
entitled “Numerical Analysis of Shell and Tube Heat Exchanger with Different Baffle
Parameters using CFD” has been approved by the board of examiners for the partial
fulfillment of the requirements for the degree of B.Sc. in the department of Mechanical
Engineering, Khulna University of Engineering & Technology, Khulna, Bangladesh in May
2022

BOARD OF EXAMINERS

1. _____________________

Dr. Md. Shahidul Islam (Supervisor)


Professor
Department of Mechanical Engineering
Khulna University of Engineering & Technology
Khulna, 9203

2. _____________________

Dr. A.N.M. Mizanur Rahman (External Member)


Professor
Department of Mechanical Engineering
Khulna University of Engineering & Technology
Khulna, 9203

ii
Acknowledgement

First and foremost, I would like to thank Almighty Allah for His kindness. It would be quite
impossible for me to continue with this thesis without His kindness.

I would also like to thank to my honorable supervisor Dr. Md Shahidul Islam, Professor,
Department of Mechanical Engineering, Khulna University of Engineering & Technology for
his proper guidance and support. His helping hand has made my work quite easy.

I am extremely grateful to Dr. Md. Shahidul Islam, Head of the Department of Mechanical
Engineering, KUET to provide such a good opportunity to complete this study and providing
all laboratory facilities.

I also like to express my deepest gratitude to Prof. Dr. Quazi Sazzad Hossain, Vice Chancellor
of Khulna University of Engineering & Technology for financial help to perform this thesis
work successfully.

Thanks, are also extended to other teachers and friends of the Department of Mechanical
Engineering, who helped me to complete the thesis works.

‘Author’

iii
Abstract

In this present era, the extensive use of shell and tube heat exchanger is seen in industrial
application like power plant, large chemical process and oil refinery due to its high pressure
and temperature applications. Segmental baffle shell and tube heat exchanger (STHX) causes
high pressure drop and extensive tube vibration. To solve this problem helical baffle is used
instead of segmental baffle. The ultimate goal of this research work is to compare the
performance of shell and tube heat exchanger with segmental and helical baffle. First the
geometry was designed using SOLIDWORKS software tools. The geometry contains 5 copper
tubes, 3 baffles, each baffle are 50% cut and a cylindrical shell with single pass parallel flow.
Instead of using regular square arrangement of tube bundle circular arrangement is used with
an extra tube inserted at the middle. Then CFD analysis is carried out using ANSYS FLUENT
commercial code. The flow pattern and heat transfer phenomenon has been investigated with
0° and 20° baffle inclination. Effectiveness for segmental and helical baffle STHX is found
0.29 and 0.36 respectively. Overall heat transfer co-efficient is found 4486.86 and 5031.81
W/m2 K respectively for segmental and helical baffle STHX. From the streamline and contour
it is seen that flow is smoother in helical baffle compare with segmental baffle. The analysis
showed that helical baffle shows better performance than segmental baffle shell and tube heat
exchanger in terms of effectiveness and efficiency of heat exchanger.

iv
Contents

Page No.

Title Page i
Approval ii
BOARD OF EXAMINERS ii
Acknowledgement iii
Abstract iv
Contents v
List of Tables vii
List of Figures viii
Nomenclature ix

Chapter-I INTRODUCTION 1
1.1 Overview 1
1.2 Objectives of the study 2

Chapter-II THEORETICAL ASPECTS AND LITERATURE REVIEW 3


2.1 The Governing Equation 3
2.2 Calculation Procedure 6
2.3 Literature Review 8
2.4 Analysis of Model 10
2.4.1 Geometry of Model 10
2.4.2 Mesh of Model 12
2.4.3 Basic Assumptions 14
2.4.4 Boundary Conditions 14
2.4.5 Problem Set Up 14
2.4.6 Accuracy Verificaton 17
2.4.7 Streamline & Contour 19

v
Page No.
Chapter- III NUMERICAL RESULTS & DISCUSSION 21
4.1 Geometry of Design 21
4.2 Mesh of Design 24
4.3 Boundary Conditions 25
4.4 Results 25
4.5 Contour 28
4.6 Discussion 32

Chapter-V CONCLUSION AND RECOMMENDATION 33


5.1 Conclusion 33
5.2 Recommendations for future work 33

References 36

vi
LIST OF TABLES

Table No. Title of the Table Page No.


Table 3.1 Dimension from paper 10
Table 3.2 Boundary Conditions 14
Table 3.3 Properties of Fluid 15
Table 3.4 Residual for convergence 16
Table 3.5 Accuracy Verification 17
Table 3.6 Mesh Sensitivity 17
Table 4.1 Dimension of Design 21
Table 4.2 Boundary Conditions 24

vii
LIST OF FIGURES

Figure No. Description of the Figure Page No.

Fig. 1.1 Segmental baffle STHX 2

Fig. 1.2 Helical Baffle STHX 3

Fig. 3.1 Front view of the HX 11

Fig. 3.2 Three dimensional view of HX 12

Fig. 3.3 Tube side mesh of HX 12

Fig. 3.4 Mesh of shell 13

Fig. 3.5 Mesh dependency check 18

Fig. 3.6 Convegence of equation 16

Fig. 3.7 Temp streamline of cold and hot fluid 17

Fig. 3.8 Temperature contour of hot and cold fluid 18

Fig. 3.9 Pressure contour of cold and hot fluid 18

Fig. 3.10 Velocity contour of cold and hot fluid 19

Fig. 4.1 Isometric view of segmental baffle STHX 21

Fig. 4.2 Isometric view of helical baffle STHX 22

Fig. 4.3 Meshing of segmental baffle STHX 23

Fig. 4.4 Meshing of helical baffle STHX 24

Fig. 4.5 Velocity streamline for segmental baffle STHX 28

Fig. 4.6 Temp contour for segmental baffle STHX 28

Fig. 4.7 Pressure contour for segmental baffle STHX 29

Fig. 4.8 Velocity profile for helical baffle STHX 30

Fig. 4.9 Velocity streamline for helical baffle STHX 31

Fig. 4.10 Temp contour for helical baffle STHX 31

viii
Nomenclature

Symbol Description Unit


𝜌 Density Kg/m3
𝜇 Viscosity N.s/m2
𝜀 Effectiveness
𝜔 Velocity in z direction m/s
𝑈 Overall heat transfer coefficient W/m2k
𝑃 Pressure Pa
𝑃𝑟 Prandtl number m2/s
𝑞 Heat transfer rate W
𝐶𝑝 Specific heat J/kg.k
𝐾 Thermal conductivity W/m.k
𝑁𝑢 Nusself Number
𝑢 Velocity in x direction m/s
𝑣 Velocity in y direction m/s

ix
CHAPTER I
INTRODUCTION

1.1 Overview:
Heat exchangers are special kind of devices which ensures heat transfer between two fluids
knows as cold and hot fluids at two different temperatures and during the transfer of heat two
fluids never get mixed with each other. Heat exchangers are extensively employed in a lot of
applications, ranging from industrial work, home air conditioning and heating to power
production in large plants and heavy chemical applications.

In STHX conjugate heat transfer phenomena happens. Which means convection heat transfer
between hot and cold fluid through the tube wall which separates the two fluids. To analyze
the effectiveness and performance of heat exchanger we need to find out overall heat transfer
coefficient denoted by U which will help us to understand the performance of HX [1].

Three types of fluid flow are possible in STHX. Flow direction of cold fluid and hot fluid is
same in parallel flow but for counter flow, flow direction is opposite with respect to each other.
In cross flow the flow direction is right angle with each other (90 degree). Counter flow is the
most effective and efficient type of flow because it increases the LMTD of heat exchanger.

A shell and tube heat exchanger consists with a number of tubes surrounded by cylindrical
shell. Due to the high number of tube heat transfer area increases thus shell and tube heat
exchanger has high heat transfer rate than conventional tubular heat exchanger. The baffle is
used to support the tubes (prevent tube vibration) and also to ensure the zig-zag pattern of shell-
side flow. Due to this support of baffle shell and tube heat exchanger can withstand the high-
pressure application. Segmental and helical are two types of baffles most commonly used in
STHX to improve the performance.

There is some problem associated with segmental baffle like it causes high-pressure drop
because it hinders the flow of shell-side flow significantly thus the high-power pump is
required which increases pumping cost.

1
Fig. 1.1: Segmental baffle STHX [15]

From Fig. 1.1 we can see a typical segmental baffle which is mostly used in industrial
applications. To solve the problem associated with segmental baffle STHX nowadays helical
baffle is used instead of a segmental baffle which ensures the smooth flow of shell-side fluid
and also increase the heat transfer rate [1].

But the problem associated with helical baffle is that it is difficult to manufacture thus
increasing manufacturing cost. Fouling is a common phenomenon in heat transfer problems
due to fouling heat transfer rate decreasing with time for this we need to clean the surface with
time. But for the helical baffle, this isn’t feasible.

2
Fig. 1.2: Helical baffle STHX [15]

From Fig. 1.2 we can see a typical helical baffle STHX which is becoming popular nowadays
in modern and sophisticated industries to increase the efficiency and effectiveness.

A typical shell and tube heat exchanger can be used at 552 bar pressure in industrial
applications. According to the flow direction, it can be classified into parallel and counter flow
STHX. Counterflow is most desirable because of the high log mean temperature difference [2].

Shell and tube heat exchanger results in the formation of a recirculation zone close to the
baffles and high-pressure drops. Most of the research is going on helical baffles in recent times.
In the present situation, the shell and tube heat exchanger are the most prevalent type of heat
exchanger which is extensively used in industrial, power plant applications. It is widely used
in large chemical and oil refinery processes. Since it can withstand high-temperature
applications it uses in so many applications [2].

3
1.2 Objectives:

The main goal of this research work is to compare the performance of shell and tube heat
exchangers. Specific objectives of this research work are:

1) To analyze the performance of shell and tube heat exchanger with segmented baffle
2) To analyze the effectiveness of shell and tube heat exchanger with helical baffle
3) To compare the performance between segmental baffle and helical baffle STHX

4
CHAPTER II
THEORETICAL ASPECTS AND LITERATURE REVIEW

2.1 The Governing Equation:

Since the flow in the heat exchanger was turbulent RNG k-€ model was used in this CFD
simulation. It contains realizable models. On the other hand, it also provides properties and
makes it dependable and truer for the swirl type of flow. RNG model improves the accuracy of
the strained flow which is very rapid because it has an additional € equation which is don’t
possible by the standard k-€ model. RNG model also enhances the effect of turbulence on swirl
flow because the effect of swirl is included in the RNG model. All these great features make
the RNG model great for fluid flow problems rather than the standard k-€ model. Flow in the
heat exchanger was considered steady and gravity was considered in the negative y-direction.

The fluid flow on the heat exchanger was turbulent and continuity, energy, and momentum
equation was present in the heat exchanger.

Continuity Equation:

 (  u )  (  v )  (  w) 
+ + + =0 (1)
x y z t

For steady-state and incompressible flow, the continuity equation is as follows

u v w
+ + =0 (2)
x y z

Momentum equation:

  p     
xi
 =−  ( U iU j ) 

(3)
 xi  xi  x j 

Energy:

 (  c pT )  (  c pT )  2T U j U k
= −Uj +K −P +  kj (4)
t x j xi x j x j x j

5
Turbulent kinetic energy equation:

 ( pk )  ( pkU j )   k 
+ = ( k eff )  + Gk + Gb −  − YM + Sk (5)
t x j x j  x j 

Turbulent dissipation energy equation:

 (  )  ( U j )     *  2
+ = ( k eff )  + C ( Gk + C 3Gb ) − C 2  + S (6)
t x j x j  x j  k k

Here,

eff =  + t (7)
k2
t =  C (8)

 = ( 2 Eij .Eij )
2 k (9)

1  U U j  (10)
Eij =  i + 
2  x j xi 

On the tube wall, no-slip boundary conditions were applied to the heat exchanger. The shell
side velocity of flow increases from the tube wall and it decreases sharply near the shell wall
region. During CFD simulation standard wall function is implemented as it provides more
accurate results than other methods.

The expression of velocity at a near-wall grid point can be written as follows:

U * = y* . y*  11.25 (11)

ln ( Ey* ) y*  11.25
1
(12)
k

Here,

1 1
 C 4 k p 2 y p
y* = (13)

T *
=
(T
w − Tp )  C pU * (14)
q
1 1
(15)
U = C k p
* 4 2

6
Here, viscous heating is denoted by D, and the convective conductive part is denoted by 𝑇 ∗ 𝑐
can be expressed by the following forms

𝑃𝑟𝑦 ∗ . 𝑦 ∗ < 𝑦 ∗ 𝑇 (17)

1
𝑇 ∗ 𝑐 = 𝑃𝑟𝑡 [𝑘 ln(𝐸𝑦 ∗ ) + 𝑃] 𝑦 ∗ > 𝑦 ∗ 𝑇 (18)

Again

1
𝜌𝑈 ∗ . 2 𝑃𝑟𝑈 2 𝑝 𝑦 ∗ < 𝑦 ∗ 𝑇 (19)

1
𝜌𝑈 ∗ . 2 [𝑃𝑟𝑈 2 𝑝 + (Pr − 𝑃𝑟𝑡 ) 𝑈 2 𝑐 ] 𝑦 ∗ < 𝑦 ∗ 𝑇 (20)

To compute P, we have the following formula

Pr 1
𝑃 = 9.24 [( 𝑃𝑟 )4 − 1] [1 + 0.28𝑒 −0.007𝑃𝑟/𝑃𝑟𝑡 ] (21) [3]
𝑡

2.2 Calculation Procedure:

The efficiency of a heat exchanger mainly relays on the outlet temperature of the shell and tube
fluid. An essential term to calculate the performance of shell and tube heat exchangers is log
mean temperature difference. Which is defined as follows [4]:

∆𝑇𝑖 − 𝑇𝑜
∆𝑇𝐿𝑀𝑇𝐷 = 𝑇 (22)
𝑙𝑛 𝑖
𝑇𝑜

For parallel flow,

∆𝑇𝑖 = 𝑇ℎ𝑖 − 𝑇𝑐𝑖

∆𝑇𝑜 = 𝑇ℎ𝑜 − 𝑇𝑐𝑜

For counterflow,

∆𝑇𝑖 = 𝑇ℎ𝑖 − 𝑇𝑐𝑜

∆𝑇𝑜 = 𝑇ℎ𝑜 − 𝑇𝑐𝑖

7
In shell and tube exchanger cross flow happens so we have to find out correction factor F from
the chart based on the design of heat exchanger.

Overall heat transfer co-efficient

𝑄𝑎𝑣𝑒
𝑈= (23)
𝐴∆𝑇𝐿𝑀𝑇𝐷

Here,

𝑄𝑐 + 𝑄ℎ
𝑄𝑎𝑣𝑒 = (24)
2

𝑄𝑐 = Shell side cold fluid heat transfer

𝑄ℎ = Tube side hot fluid heat transfer

𝑄𝑐 = 𝑚𝑐 𝐶𝑝𝑐 (𝑇𝑐𝑖 − 𝑇𝑐𝑜 )

𝑄ℎ = 𝑚ℎ 𝐶𝑝ℎ (𝑇ℎ𝑖 − 𝑇ℎ𝑜 )

𝑄𝑚𝑎𝑥 = (𝑚𝐶𝑝 )𝑚𝑖𝑛 (𝑇ℎ𝑖 − 𝑇𝑐𝑖 )

𝑄
Effectiveness of heat exchanger, 𝜀 = (25)
𝑄𝑚𝑎𝑥

2.3 Literature Review:

The performance of the shell and tube heat exchanger solely depends on the arrangement and
design of the tube bundle and baffle. Extensive research has been going on to improve the
effectiveness and efficiency so that performance of shell and tube heat exchangers can be
improved.

Kiran et al. investigated the effect of baffle spacing baffle spacing in STHX using CFD to
understand the characteristics of heat transfer on shell side fluid flow. They found that
increasing the number of baffles means decreasing baffle spacing heat transfer rate and pressure
drop both increases [5].

They also found that pressure drop is more sensitive than heat transfer coefficient on baffle
spacing. As a heat exchanger designer, one must have to design the heat exchanger according
to the outlet temperature and associated pressure drop.

8
Due to extensive pressure drop associated with segmental baffle, Lutch and Nemcansky were
the first who made significant changes. Instead of using a segmental baffle, a helical baffle was
recommended to improve the performance of tubular heat exchanger. This diminishes the
problem associated with segmental baffle thus fluid is less staggered means reducing shell side
pressure drop and improving the overall performance of STHX [6].

Another research B. Peng and his team made an experiment on STHX with a continuous helical
baffle and they found that heat transfer rate increases significantly with the rate of pressure
drop in STHX. A continuous helical baffle also reduces the flow-induced tube vibration and
effect of fouling [7].

Taher investigated the impact of helical baffle spacing on STHX using commercial Fluent
software code and found that with the increase of baffle spacing heat transfer rate decreases.
They also noticed that with the increase of helical baffle spacing pressure gradient gradually
decreases [8].

STHX works based on a couple of co-relation among them kern method and the bell Delaware
method are the most commonly used method. Kern method can be used to calculate both shell-
side and tube side heat transfer coefficients [9].

Bell Delaware method can only be applicable for shell side heat transfer problems and it is a
more accurate methodology than kern method to mathematically calculate the heat transfer rate
in shell side fluid [10].

Roktupal Borah and R.K Chitharthan analyzed the helical baffle STHX using numerical
simulations and found the optimum helix angle. They investigated 30, 35, 40, 45, and 50-degree
baffle inclination and found 45 degrees is the best baffle inclination which increases heat
transfer rate more than other baffle inclination angles [11].

Andrews and Prithiviraj numerically simulated the three-dimensional STHX and got a better
idea about the flow characteristics of STHX and compared these results with the existing model
to predict the flow pattern in STHX [12].

Andrews with Master also did the three-dimensional numerical analysis using computational
fluid dynamics on helical baffle STHX.

9
To get more accurate results Jian Wen proposed an improved algorithm for the modeling of
computer response. The algorithm is known as MOGA means multi objective genetic
algorithm. This algorithm was proposed to optimize the STHX with helical baffles [13].

The ultimate goal of this research work is to compare the performance of shell and tube heat
exchangers with different baffle parameters like a segmental and helical baffle. And find out
the best possible solution also to discuss which baffle will be more effective and viable.

First, the geometry was designed using SOLIDWORKS software tools. The geometry contains
5 copper tubes, 3 baffles, each baffle is 50% cut, and a cylindrical shell with the single-pass
parallel flow. Then CFD analysis was done using ANSYS FLUENT commercial code. And
the flow pattern and heat transfer phenomenon were investigated for 0- and 20-degree baffle
inclination angles.

Computational Fluid Mechanics also known as CFD is now a well-known branch of fluid
mechanics that deals with complicated fluid flow problems based on numerical methods and
algorithms. CFD is also a very helpful tool for studying static structural problems, chemical
reactions inside IC engines and heat transfer by solving complicated mathematical equations
using numerical analysis.

A lot of commercial software is now available like Ansys, OpenFOAM, and researcher and
scientists often use them to simplify their work. Computational fluid dynamics works on a very
simple principle. At first, it resolves the entire system into grids or small cells and then applies
governing equations on these elements and finds the result in the form of temperature gradients
or pressure distribution [14].

CFD tools work on three basic steps


Pre-Processor:

▪ Identify the process to be evaluated


▪ Drawing the geometry using CAD tools
▪ Creating the domain and name selection tool
▪ Creating the mesh in mesh in the flow domain

10
Solver:

• Apply boundary conditions on the domain


• Solve the governing equations using computational fluid dynamics

Post-Processor:

▪ Interpret the results


▪ To highlight the findings post-process the solution [14]

2.4 Analysis of Model:

The entire research work has been carried out using commercial code Ansys Fluent. To verify
the methodology “Three-Dimensional Study of Baffle Effects by Latih Habeeb” had been
simulated.

2.4.1 Geometry of Model:

CFD analysis of heat exchangers depends on the geometry. To draw the geometry properly
we need the necessary dimension of the heat exchanger which I got from the paper. In Table
3.1 all the necessary dimension of the model for validation purpose are clearly depicted.

Table 3.1: Dimension from Paper

Tubes length = 610 mm Shell length = 610 mm


Tube diameter = 16 mm Shell diameter = 160 mm
No of tubes = 21 Tube thickness = 2 mm
Tube arrangement = square Tube pitch = 25.5 mm

To make the CFD process simple extra length of the shell due to inlet and outlet clearance was
neglected since it doesn’t make much effect on heat transfer rate.

As a heat exchanger designer while designing the heat exchanger one must have to comply
with tubular exchanger manufacturer guidelines. These data follow their manufacturing
guidelines.

11
In a tubular heat exchanger, conjugate heat transfer takes place means natural convection and
conduction. To make sure of this all the bodies were combined and made a new part.

The efficiency and effectiveness of STHX depend on the design of the tube bundle and the
arrangement of the baffle. There are two popular methods of tube arrangement

• Square arrangement
• Triangular arrangement

Here square arrangement of tube was used. The distance between tube to tube is known as tube
pitch. In Fig. 3.1, from the front view of the heat exchanger, the arrangement of the tube bundle
can be seen. From Fig. 3.2, a three-dimensional view of the heat exchanger can be seen.

Fig. 3.1 Front view of the heat exchanger

From Fig. 3.1, we can see the number of tubes used in the design. We can also see that there
are 21 tubes present, with a square arrangement of tube bundles.

12
Fig. 3.2: Three-dimensional view of heat exchanger

From Fig. 3.2, we can see the three-dimensional view of a heat exchanger with 21 tubes inside
a cylindrical shell for the validation purpose of our CFD methodology.

2.4.2 Mesh of Model:

Since this heat exchanger contains 21 tubes, each tube has its thickness and one outer
cylindrical shell. It is almost impossible to do fine m5`eshing (i.e., structured mesh) using a
personal laptop. For simplification, ANSYS FLUENT default mesh was used (fine mesh
instead of coarse mesh) And the simulation was let to run with proper boundary conditions for
validation purposes.

In Fig. 3.3, the tube side mesh of the heat exchanger is shown. From Fig. 3.4, the mesh of the
shell can be seen properly.

13
Fig. 3.3: Tube side mesh of the heat exchanger

From Fig. 3.3, we can see a fine mesh of the tubes which ensures the accuracy of our
methodology to minimize numerical error below one percent.

Fig. 3.4: Mesh of shell

Fig. 3.4, is the meshing of shell side cold fluid. This shell contains 21 tubes with square
arrangement of tube bundles inside for hot fluid flow.

14
2.4.3 Basic Assumptions:

To make our CFD simulation easier we need to take some basic assumption for simplification
purpose.

• Conjugate heat transfer was considered between two fluids


• Due to the simplification purpose, the outer wall thickness of the tube surface was
neglected
• Steady-state heat transfer was considered
• Fluid properties were considered as an incompressible ideal gas
• Natural radiation and convection were neglected

2.4.4 Boundary Conditions:

Cold fluid flows inside the shell side on the other hand-hot fluid flows inside the tube side.
Shell has its inlet and outlet and tube has its own also. Due to fouling cold fluid is let to flow
through the shell so that the tubes can be easily cleaned. To solve the problem, we need to
assign boundary conditions. In Table 3.2, all of the necessary boundary conditions are given
here. In Table 3.3, properties of fluids are given here. To solve the equation, we need to set the
convergence criteria. From Table 3.4, the residual for convergence can be taken.

Table 3.2: Boundary Conditions

Boundary The rate of Turbulent kinetic Turbulent rate of Temperature


Conditions mass flow energy dissipation
Shell inlet Mass flow 0.02 kg/s 1×10−2 m2 /s 1×10−2 m2 /s 3 300 K
Tube inlet Mass flow 0.01 kg/s 1×10−2 m2 /s 1×10−2 m2 /s 3 350 K

2.4.5 Problem Set-Up:

The energy equation is enabled first. Since the flow is turbulent RNG k - € model was used for
this CFD analysis. Aluminum was used as a material for shells and tubes due to its high heat
transfer characteristics.

15
Shell domain and tube domain were set to fluid. Boundary condition of inlet is assigned as
mass flow inlet and boundary condition of outlet is assigned as mass flow outlet. From the
definition of a heat exchanger shell, the outer surface acts as an adiabatic wall means no heat
can be transferred to surroundings.

Solution Models:

• Simple scheme
• Standard pressure
• Least square cell
• Momentum – second-order upwind
• Turbulent kinetic energy – second-order reversing
• Turbulent rate of dissipation – second-order reversing

Table 3.3: Properties of Fluid (Water)

Variable Unit
Density 1 kg/m3
Momentum 0.70 kg.m/s
Body forces 1 kg.m2 /s2
Pressure 0.30 Pa
Turbulent kinetic 0.8 m2 /s2

Convergence Criteria:

Before initializing the program, we must have to set up the convergence criteria which means
how much accuracy we want. Since in the case of heat transfer solving the energy equation is
the main goal as accurate as possible that’s why convergence criteria for the energy equation
were set far below any other equation.

16
Table 3.4 Residual for Convergence

Variable of residual Residual


X component of velocity. 0.1×10−5
Y component of velocity. 0.1×10−5
Z component of velocity. 0.1×10−5
Energy 0.1×10−8
Continuity 0.1×10−5
Turbulent kinetic 0.1×10−4
Dissipation energy 0.1×10−4

Run Calculation:

Hybrid initialization was used and the simulation was set to run for 4000 iterations. At 560
iterations the equation is converged and the shell side cold fluid outlet temperature was found
315.9 K and the tube side hot fluid outlet temperature was found 316 K. Convergence of the
equation can be seen from Fig. 3.7 for 560 iterations.

Fig. 3.5: Convergence of equation

17
2.4.6: Mesh Sensitivity:

A series of simulations were performed to find out the independent solutions. Mesh
containing 1.9 million was selected for this analysis.

Table 3.5: Mesh dependency check

Number of elements Hot outlet (K)

319514 309.7

445956 313.1

761354 314.9

1119659 315.97

1930654 315.99

Hot Fluid
317

316

315
Hot Outlet (K)

314

313

312

311

310

309
0 500000 1000000 1500000 2000000 2500000
Num of Elements

Fig. 3.6: Mesh dependency check

From Fig. 3.6, we can see that 1.5 million to 1.9 million mesh, result doesn’t change
significantly. So, 1.9 million elements were selected for this numerical analysis.

18
2.4.7 Accuracy Verification:

The desired shell side outlet temperature was 315 K and the tube side outlet temperature was
316 K. From Table 3.5, the result of our numerical analysis can be seen, and the tables show
that our methodology is validated.

Table 3.5: Accuracy Verification

Outlet Paper Work My Work % Error


Cold Outlet 317 K 315.9 K 0.35 %
Hot Outlet 316 K 315.99 K 0.003 %

2.4.8 Streamline & Contour:

Fig. 3.7: Temperature streamline of cold and hot fluid

From Fig. 3.7, it is seen that the temperature of cold fluid is increasing and the temperature of
hot fluid is decreasing as the flow advances towards the outlet due to the transfer of heat
between cold fluid and hot fluid.

19
Fig. 3.8: Temperature contour of cold and hot fluid

From Fig. 3.8, the temperature contour shows that temperature of the cold fluid increases and
the temperature of the hot fluid decreases as the flow advances.

Fig. 3.9: Pressure contour of cold and hot fluid

20
From Fig. 3.9, the pressure contour clearly depicts that the pressure of cold fluid and hot fluid
decreases as the flow advances and it is zero at the outlet, which it should be. Because of the
boundary condition, there is a requirement for zero pressure at the outlet.

Fig. 3.10: Velocity contour of hot and cold fluid

From Fig. 3.10, the velocity profile shows that hot fluid velocity is higher than cold fluid
velocity, which validates our criteria from the boundary conditions.

21
Chapter IV
Numerical Results & Discussion

The ultimate objectives of the research work is to compare the performance of shell and tube
heat exchangers with different baffle parameters and find out the most efficient and effective
types of baffles.

Two types of shell and tube heat exchanger are available according to the flow direction.

• Parallel flow STHX


• Counterflow STHX

The research work has been carried out for parallel flow STHX

As previously stated earlier the core performance of shell and tube heat exchanger depends on
the tube bundle design and baffle arrangement.

There are two popular methods of tube arrangement

• Square tube arrangement


• Triangular tube arrangement

4.1 Geometry of Design:

A typical industrial heat exchanger contains 500 tubes and approx. 100 baffles. Simulating this
kind of heat exchanger using a personal computer can’t imagine. To make the simulation
process easier the STHX was designed using five tubes and three baffles.

Every heat exchanger designer must comply with the tubular manufacturer association
(TEMA) guidelines during the designing of a heat exchanger

During the design of this shell and tube heat exchanger, tubular exchanger manufacturers
association (TEMA) guidelines were completely followed.

The tube bundle was redesigned with a rotated square arrangement with an extra tube inserted
in the middle to enhance the heat transfer area a little bit more of the shell and tube heat
exchanger.

22
The pumping power required of a shell and tube heat exchanger mainly depends on the quantity
of baffles and the percentage of baffle cut. To reduce the pumping cost, the heat exchanger was
designed using three baffles with 140 mm baffle spacing. A 50 percent baffle cut is used to
ensure the smooth flow of shell-side cold fluid to reduce the pressure drop and pumping cost.
In helical baffle heat, exchanger baffles are arranged at a 20o inclination angle to make sure the
smooth flow of shell side cold fluid.

Aluminum is used as a material for shells and tubes due to its high heat transfer capability. In
Table 4.1, all the necessary dimensions of the design are presented. The isometric view of
segmental baffle STHX can be seen from Fig. 4.1, and from Fig. 4.2, the baffle and tube
position in segmental baffle and helical baffle STHX can be understood more clearly.

Table 4.1: Dimension of Design

Diameter of shell = 120 mm Length of shell = 520 mm


Number of baffles = 3 Number of tubes = 5
Tube outer diameter = 15 mm Baffle cut = 50%
Tube arrangement: rotated square Baffle inclination = 20 degree

Fig. 4.1: Isometric view of segmental baffle STHX

From Fig. 4.1, we can see a typical segmental baffle STHX with 3 baffles, 5 tubes with square
arrangement of tube bundle with 0o baffle inclination.

23
Fig. 4.2: Isometric view of helical baffle STHX

From the three-dimensional view of helical baffle STHX of Fig. 4.2, we can see a typical helical
baffle STHX with 3 baffles at 20o inclination, 5 tubes with square arrangement of tube bundle
and the inlet and outlet of fluids.

4.2 Mesh of Design:

Since a shell and tube heat exchanger contains several tubes, each tube has its tube thickness.
Baffles are also present. Due to its complex geometry, it is very difficult to do fine mesh using
a personal laptop. During this simulation fine mesh was used instead of coarse mesh. This mesh
has 1 million elements which take almost 10 to 12 hours to complete. The tube side mesh of
the segmental baffle can be seen from Fig. 4.3, and from Fig. 4.4, the meshing of the helical
baffle STHX can be seen.

24
Fig. 4.3: Meshing of segmental baffle STHX

From Fig. 4.3, we can see the meshing of segmental baffle STHX to make the CFD analysis
easier with less time mesh refinement wasn’t done in this analysis. Because with baffles it
creates almost 10 million of elements which is impossible to solve using personal computer.

Fig. 4.4: Meshing of helical baffle STHX

25
From Fig. 4.4, we can see the meshing of helical baffle STHX to make the CFD analysis easier
with less time mesh refinement wasn’t done in this analysis. Because with baffles it creates
almost 10 million of elements which is impossible to solve using personal computer.

4.3 Boundary Conditions:


Before initializing the problem we need to assign the boundary condition of hot and cold
inlet. Means we have to assign hot and cold inlet temperature and velocity of flow. Necessary
boundary conditions can be taken from Table 4.2.

Table 4.2: Boundary Conditions

Hot inlet velocity = 4 m/s Hot inlet temperature = 400 K


Cold inlet velocity = 2 m/s Cold inlet temperature = 273 K

4.4 Results:
Segmental Baffle:

Cold fluid outlet temperature = 282.71 K

The outlet temperature of hot fluid = 374.11 K

Rate of mass flow for cold fluid = 1.26 kg/s

Rate of mass flow for hot fluid = 1.77 kg/s

From the velocity of fluid by using the following formula, m = 𝜌. 𝐴. 𝑉

Heat rejected by hot fluid

qhot = mhot C ph (Thi − The )

Calculating we get,

𝑞ℎ𝑜𝑡 = 191641.4 𝑤

To calculate maximum heat transfer rate, we need to use the following formula
26
qmax = ( mC p ) (Thi − Tci )
min

Calculating we get

𝑞𝑚𝑎𝑥 = 669203.64 𝑤

Now effectiveness for segmental baffle STHX

q
=
qmax

We get, effectiveness for segmental baffle STHX = 0.29

Helical Baffle:

Hot fluid outlet temperature = 367.12 K

Cold fluid outlet temperature = 277.9 K

Het rejected by hot fluid

Calculating,

qhot = mhot C ph ( Thi − The )

𝑞ℎ𝑜𝑡 = 243382.36 𝑤

Similarly

𝑞𝑚𝑎𝑥 = 669203.64 𝑤

Effectiveness for helical baffle STHX = 0.36

So, from this numerical result, we can say that helical baffle STHX is more efficient than
segmental baffle STHX

LMTD Calculation:

Ti − To
TLMTD =
T 
ln  i 
 To 

27
Segmental Baffle:

∆𝑇𝑖 = 127 K

𝑇𝑜 = 91.4 K

Calculating we get,

∆𝑇𝐿𝑀𝑇𝐷 = 108.23 K

Helical Baffle:

∆𝑇𝑖 = 127 K

𝑇𝑜 = 89.22 K

Calculating we get,

∆𝑇𝐿𝑀𝑇𝐷 = 107 K

Overall Heat Transfer Co-efficient:

To calculate the overall heat transfer co-efficient first we have to find out the average heat
transfer rate and heat transfer area

A = 𝑁. 2. 𝜋. 𝑅. 𝐿

Calculating heat transfer area = 0.25 𝑚2

qavg
U=
A TLMTD

qhot + qcold
qavg =
2

Segmental Baffle:

𝑞𝑎𝑣𝑔 = 121403.25 W

By calculating the overall heat transfer co-efficient = 4486.86 W/ 𝑚2 𝑘

28
Helical Baffle:

𝑞𝑎𝑣𝑔 = 134601.01 W

By calculating overall heat transfer co-efficient = 5031.81 W/ m2 k

From this calculation, we can say that helical baffle STHX is far far better than segmental
baffle STHX

4.5 Contour:

From the contour and streamline we can easily understand the temperature, pressure, and
velocity of fluid flow. This research work has been done for parallel flow.

Segmental Baffle:

To understand the fluid flow and heat transfer phenomena in segmental baffle STHX we need
to take a look at streamline and contour.

Fig. 4.5: Velocity streamline for segmental baffle STHX

29
From Fig. 4.5, the velocity streamline depicts that the segmental baffle ensures the zig-zag
flow of shell side cold fluid flow, thus increasing the heat transfer rate

Fig. 4.6: Temperature contour for segmental baffle STHX

From the temperature contour of Fig. 4.6, it is seen that the cold fluid temperature is increasing
with the flow advances. We can also visualize the flow direction from the temperature contour.

30
Fig. 4.7: Pressure contour for segmental baffle STHX

From the pressure contour of Fig. 4.7, it is seen that as the flow advances, pressure is
decreasing. The reason for this pressure outlet being set to 0 Pa from the definition of boundary
conditions.

Helical Baffle:

To understand the fluid flow and heat transfer phenomena in helical baffle STHX we need to
take a look at streamline and contour.

Fig. 4.8: Velocity profile for helical baffle STHX

From the velocity profile of Fig. 4.8, it can be clearly said that in helical baffle shell and tube
heat exchanger fluid flow on the shell side is less staggered, which means flow is smoother in
helical baffle than in segmental baffle.

31
Fig. 4.9: Velocity streamline for helical baffle STHX

From the velocity streamline of Fig. 4.9, it can be seen that in helical baffle STHX flow is
smoother than segmental baffle STHX which solves the problem associated with segmental
baffle STHX.

Fig. 4.10: Temperature contour of helical baffle STHX

32
From the temperature contour of Fig. 4.10, it is seen that the temperature of shell-side fluid
decreases rapidly at the later stage of the heat exchanger than at the first stage.

The ultimate objective of this research work is to find the best possible solution for STHX.
From the calculation, it is clear that the overall heat transfer coefficient and effectiveness of
helical baffle STHX is far better than segmental baffle STHX. Although helical baffle solves
the pressure drop and flow pattern problems associated with segmental baffle STHX, helical
baffle STHX also has some problems, like it is difficult to manufacture. And the most important
problem associated with STHX is fouling. Due to fouling, the heat transfer rate decreases with
time. And for this reason, tubes need to be cleared regularly.

It is very easy to remove the tube in the segmental baffle shell and tube heat exchanger, but
since helical baffle STHX is complex in shape, it is not too easy to remove the tube and again
assemble them.

Cleaning of tubes is a difficult task in helical baffle STHX since the assembling of tube bundles
is a time-consuming task. In terms of manufacturing cost, helical baffle STHX is also more
expensive than segmental baffle STHX.

33
CHAPTER V
CONCLUSION AND RECOMMENDATIONS

5.1 Conclusions:

In this research work effect of the segmental and helical baffle on STHX was analyzed using
computational fluid dynamics. Comparison of performance of segmental baffle and helical
baffle STHX was also discussed here. The results lead to the following conclusion

• Segmental baffle causes higher pressure drop and vorticity than helical baffle STHX
• Flow is smoother in helical baffle STHX which reduces pumping cost
• Helical baffle is more effective and shows better performance than segmental baffle
STHX

5.2 Recommendations for future work:

This research work has been carried out for a parallel flow shell and tube heat exchanger with
segmental and helical baffle. By changing the flow direction from parallel to counter flow,
another research work can be done, which will give better results. Performance can be further
improved by making a little bit of modification in baffle design, like instead of using segmental
and helical baffles with inclination, continuous helical baffle can be used, which will make the
shell side cold fluid flow smoother than segmental baffles and will also increase the
effectiveness and efficiency of the heat exchanger further.

34
REFERENCES:

[1] D. De, T. K. Pal, and S. Bandyopadhyay, “Helical baffle design in shell and tube type
heat exchanger with CFD analysis,” Int. J. Heat Technol., vol. 35, no. 2, pp. 378–383,
2017, doi: 10.18280/ijht.350221.

[2] A. Joshi, “Numerical Analysis of Shell-Tube Heat Exchanger with Baffle At Inclined
Angle,” vol. 13, no. 11, pp. 40–46, 2017.

[3] K. Kiran, M. A M, and J. Manoj and Umesh C, “Investigation of baffle spacing effect
on shell side heat transfer characteristics in shell and tube heat exchanger using
computational fluid dynamics,” Elixir Therm. Engg 26022-26026, vol. 73, pp. 26022–
26026, 2014.

[4] J. Lutcha and J. Nemcansky, “Performance improvement of tubular heat exchangers by


helical baffles,” Chem. Eng. Res. Des., vol. 68, no. A3, 1990.

[5] B. Peng et al., “An experimental study of shell-and-tube heat exchangers with
continuous helical baffles,” J. Heat Transfer, vol. 129, no. 10, pp. 1425–1431, 2007,
doi: 10.1115/1.2754878.

[6] F. Nemati Taher, S. Zeyninejad Movassag, K. Razmi, and R. Tasouji Azar, “Baffle
space impact on the performance of helical baffle shell and tube heat exchangers,”
Appl. Therm. Eng., vol. 44, pp. 143–149, 2012, doi:
10.1016/j.applthermaleng.2012.03.042.

[7] S. Shinde and H. Pancha, “Comparative Thermal Performance Analysis Of Segmental


Baffle Heat Exchanger with Continuous Helical Baffle Heat Exchanger using Kern
method,” Mustansir Hatim Pancha / Int. J. Eng. Res. Appl., vol. 2, no. August, pp.
2264–2271, 2012, [Online]. Available: www.ijera.com

[8] M. Serna and A. Jiménez, “A compact formulation of the Bell-Delaware method for
heat exchanger design and optimization,” Chem. Eng. Res. Des., vol. 83, no. 5 A, pp.
539–550, 2005, doi: 10.1205/cherd.03192.

[9] I. Journal et al., “Analysis of Helical Baffle Heat Exchanger for Optimum Helix,”
Anal. Helical Baffle Heat Exch. Optim. Helix Angle Through Numer. Simulations, vol.
4, no. 4, pp. 44–50, 2015.

[10] M. J. Andrews and B. I. Master, “Three-dimensional modeling of a Helixchanger®


35
heat exchanger using CFD,” Heat Transf. Eng., vol. 26, no. 6, pp. 22–31, 2005.

[11] J. Wen, H. Yang, G. Jian, X. Tong, K. Li, and S. Wang, “Energy and cost optimization
of shell and tube heat exchanger with helical baffles using Kriging metamodel based
on MOGA,” Int. J. Heat Mass Transf., vol. 98, pp. 29–39, 2016, doi:
10.1016/j.ijheatmasstransfer.2016.02.084.

[12] P. Abbas Kiyasudeen and T. Dharmaprabhakaran, “Analysis of Effect of Baffle


Inclination Angle on Shell and Tube Heat Exchanger Using Computational Fluid
Dynamics,” vol. 3, no. 1, pp. 48–59, 2017, [Online]. Available:
www.irjes.psyec.edu.in

[13] S. Hossain and S. Islam, “Numerical Investigation on Shell-and-Tube Heat Exchanger


with Segmental and Helical Baffles,” vol. 4, no. 2, pp. 1–16, 2018.

[14] Y. Cengel and T. M. Heat, A practical approach. New York, NY, USA: McGraw-Hill,
2003.

[15] M. S. Kassim, A. Oleiwi, D. S. Khudhur, and L. J. Habeeb, “Three Dimensional Study


of Baffles Effect on Heat Transfer in Shell and Tube Heat Exchanger,” J. Mech. Eng.
Res. Dev., vol. 43, no. 5, pp. 332–345, 2020.

36

You might also like