Effects of Operating Conditions on Calcium Carbonate Fouling in a Plate Heat Exchanger

World Academy of Science, Engineering and Technology 53 2009
Effects of Operating Conditions on Calcium
Carbonate Fouling in a Plate Heat Exchanger
K. Pana-Suppamassadu, P. Jeimrittiwong, P. Narataruksa, and S. Tungkamani
crystallization or precipitation fouling, chemical reaction
Abstract The aim of this work is to investigate on the internal-
fouling, corrosion fouling, biological fouling or biofouling,
flow patterns in a plate heat exchanger channel, which affect the
freezing fouling and particulate fouling or sedimentation
rate of sedimentation fouling on the heat transfer surface of the
fouling. Particulate fouling has a major problem when sea
plate heat exchanger. The research methodologies were the water and under ground water that contain high amount of
computer simulation using Computational Fluid Dynamics (CFD)
calcium ion were used [2]-[5].
and the experimental works. COMSOL MULTIPHYSICS™
Version 3.3 was used to simulate the velocity flow fields to verify
Flow characteristics of fluid (angle of fluid flow) in plate
the low and high flow regions. The results from the CFD technique
heat exchanger are the main parameters that case of fouling
were then compared with the images obtained from the form on the surface of plate heat exchanger. Adjustment
experiments in which the fouling test rig was set up with a single-
flow characteristics of fluid by using simulation program
channel plate heat exchanger to monitor the fouling of calcium
can predict fouling of plate heat exchanger. Information of
carbonate. Two parameters were varied i.e., the crossing angle of
flow characteristics from simulation results can help design
the two plate: 55/55, 10/10, and 55/10 degree, and the fluid flow
team to design heat exchanger which reduces fouling form.
rate at the inlet: 0.0566, 0.1132 and 0.1698 m/s. The type of plate
For the experiments, fouling in plate heat exchanger
“GX-12” (the surface area 0.12 m2, the depth 2.9 mm, the width of
(calcium sulfate as a cold fluid) studied by [6] was shown
fluid flow 215 mm and the thickness of stainless plate of 0.5 mm)
was used in this study. The results indicated that the velocity
that heat transfer rate decrease 20 %, pressure drop
distribution for the case of 55/55 degree seems to be very well
increased 300-400 %, rate of fouling increase at high
organized when compared with the others. Also, an increase in the
temperature and second order was the reaction control.
inlet velocity resulted in the reduction of fouling rate on the surface
Kinetic model was defined by [7] that shown three steps of
of plate heat exchangers.
fouling, first, initial step that heat transfer rate increase due
to roughness of surface area, second step was the fouling
Keywords Computational fluid dynamics, crossing angles, form and last step was the constant of thermal conductivity.
finite element method, plate heat exchanger.
Fouling was been off by shear force that equal rate of
fouling.
I. INTRODUCTION
For the simulation models, Fouling in plate heat
OWADAYS, plate heat exchangers are used exchanger was studies by computational fluid dynamic that
Nworldwide in all industries such as petroleum industry, represented one dimensional and shown flow pattern in
food industry, pharmaceutical and material industry by radius direction [8]-[11] and two dimensional. Two and
replacing shell and tube heat exchangers. Advantages of
three dimensional models represented higher accuracy of
plate heat exchangers are higher thermal conductivity, flow pattern, temperature distribution and fouling than one
higher accuracy of temperature control, lower of contact
dimensional model [12], [13]. Three dimensional models
time, lower construction area, and easier for cleaning than
shown point of the highest of fouling that occurred at the
shell and tube heaters [1].
highest temperature and velocity of fluid flow had effect at
However, disadvantage of plate heat exchangers is initial of two or three of corrugate plate [13].
fouling (formed by CaCO
In the present work, fluid velocity profile in plate heat
3 at high operating temperature)
that cased by flow pattern in plate heat exchanger that exchanger was investigated by using the experiment and
fouling was defined as the accumulation of unwanted
computational fluid dynamic technique (CFD). The
materials on the surface of heat exchanger. It has a lot of
objective was to find the optimum angle of plate heat that
problems in design and operation such as the fouling layer
low fouling of CaCO3 occurred.
has a low thermal conductivity and as deposition occurs, the
cross-sectional area was reduced, that causes an increase in
II. METHODOLOGY
pressure drop. Fouling can be separated in six types such as
A. Experimental
K. P. is with the Department of Chemical Engineering, King Mongkut’s
Experimental equipments of plate heat exchanger were
University of Technology North Bangkok,1518 Piboonsongkhram Rd.,
designed for studying flow pattern and fluid distribution at
Bangsue, Bangkok 10800, Thailand (e-mail: mhc@kmutnb.ac.th,
monpilai@gmail.com).
via angle of plat heat which was shown in Fig. 1. The
P. J. is with the Department of Chemical Engineering, King Mongkut’s
descriptions of equipments and parts were explained in
University of Technology North Bangkok, 1518 Piboonsongkhram Rd.,
Table I. The plates were set at different angle i.e., 10/10,
Bangsue, Bangkok 10800, Thailand (e-
55/10 and 55/55, respectively, and plate properties were
mail:karanp@kmutnb.ac.th).
P. N. is with the Department of Chemical Engineering, King Mongkut’s
shown in Table II. In experimental work, a corn powder
University of Technology North Bangkok, 1518 Piboonsongkhram Rd.,
solution (50% by weight) was used as a replacement for
Bangsue, Bangkok 10800, Thailand (e-mail:phn@kmutnb.ac.th).
CaCO3 at the beginning of fouling. Feed flow rate of corn
S. T. is with the Department of Industrial Chemistry, King Mongkut’s
powder solution was set at 0.0566, 0.1132 and 0.1698 m/s,
University of Technology North Bangkok,1518 Piboonsongkhram Rd.,
Bangsue, Bangkok 10800, Thailand (e-mail: mhc@kmutnb.ac.th, respectively.
monpilai@gmail.com).
1204

---

World Academy of Science, Engineering and Technology 53 2009
Fig. 1 Schematic diagram of plate-and-frame (PHE) experimental set up for fouling investigation
B. Simulation
375642, and 312737 elements, respectively.
Simulation was obtained by using COMSOL
TABLE I
MULTIPHYSICS program version 3.3. Characteristics
EQUIPMENTS AND PARTS IN THE SET UP
of flow pattern and fluid distribution in plat heat
Code Equipments/Parts
exchanger were investigated via angle of plate heat as
T-1 Stainless
Steel
Tank AISI 304 (100 liter)
shown in Figure 2. Calculation of flow properties in plate
T-2 Stainless
Steel
Tank AISI 304 (100 liter)
heat exchanger was based on mathematical models for the
P-1 Centrifugal
Pump
conservation of momentum and mass by the used of
P-2 Centrifugal
Pump
Incompressible Navier-Stokes Equations.
M-H Motor
V-1 Globe
Valve
T

u
u
.
u
u
p
0 (1)
V-2 Globe
Valve
V-3 Globe
Valve
V-4 Globe
Valve
where
is fluid viscosity (kg/m s)
F-1 Rotameter
u is fluid velocity (m/s)
F-2 Rotameter
is fluid density (kg/m3)
p is pressure (Pa)
TABLE II
3D application mode is used to model some parts of plate
PLATE SPECIFICATIONS
heat exchanger. The corresponding computational
Material Stainless
Steel
ASTM
316
domains of plate configurations were exhibited in Fig. 3.
Type GX-12
Density (994 kg/m3) and viscosity (7.2×10-4 kg/m s) of
Surface Area (m2) 0.12
solution was set in Sub-domain setting. In flow/out flow
Depth (mm)
2.9
velocity (feed inlet), neutral (side wall) and no slip were
Width (mm)
215
set in Boundary setting as shown in Fig. 4(a). The
Angle H/L (H-pl0)
10/10, 55/55, 55/10
computational mesh fineness was set at Free Mesh
Thickness (mm) 0.5
Parameter, and Tetrahedral Mesh was adopted as shown
in Fig. 4(b). A number mesh element used for 55/55,
10/10, and 55/10 plate configurations were 348034,
1205

---

World Academy of Science, Engineering and Technology 53 2009
(a)
(b)
(c)
Fig. 3 Computational domain of flow between plates of various configurations: (a) 55/55 degree, (b) 10/10 degree, and
(c) 55/10 degree
Neutral
Neutral
Neutral
No slip
Vz, in (Flow Inlet)
(a)
(b)
Fig. 4 Computational set up: (a) boundary conditions, (b) mesh in the FE scheme
1206

---

World Academy of Science, Engineering and Technology 53 2009
III. RESULTS AND DISCUSSIONS
with the periodicity of the flow field in the z-x, and y-z
planes.
Fig. 4(a)-4(c) illustrated the comparison between the
Well-defined spots of fouling was noticed in the 10/10
simulation and experimental results of flow patterns and
degree configuration as well as shown in Fig 4(b). Even
velocity distributions within the plate heat exchanger with
though the fouling formation occurred in a well-defined
configurations of 55/55, 10,10 and 55/10 degree, fashion, only small amount of fouling formed because the
respectively, and the flow velocity at inlet was set at 0.1132
flow velocity was relatively higher than the previous case of
m/s. From plan view or x-y plane, the simulated flow the 55/55 degree plate. The maximum velocity in this
patterns agreed with the experimental results. The results
configuration was 0.7 m/s, and found in the z-x plane. On
shown in Fig. 4(a) were for the 55/55 degree configuration
the contrary, there was almost no flow velocity in a certain
of the plate heat exchanger. Point 1 in Fig. 4(a) was the
y-z plane, thus the fouling tended to occur in this section
contact point where the upper and lower plate were in close to the contact and the backside points. The plate heat
contact, and point 2 was the location right behind the contact
with the same degree of rotation of the upper and lower
point. The velocity was null at the contact point and plates but in the opposite sense exhibited symmetrical or
increased towards point 3 near the largest cross-sectional
skew-symmetrical patterns such like those found in the
area where the local maximum velocity was highest about
cases of 55/55 and 10/10 degree configuration. In the 55/10
0.3 m/s. The experimental image indicated that fouling degree plate heat exchanger, the fouling spots appeared on
occurred on the backside of the contact point i.e., at point 2
the back of contact points, which formed the diagonal
since a local flow and shear stress were low. On the other
patterns. The velocity distribution in the y-z plane indicated
hand, point 3 showed no fouling according to the high flow
that the high velocity zones were fouling free sites.
and shear. The pattern of fouling formation extended over
the whole plate with well-defined repeated pattern in accord
3
2
Flow Inlet
1
Contact Point
z-x plane

y-z plane
Fig. 4(a) Flow distributions on the z-x, and y-z planes in the case of inlet velocity equalled to 0.1132 m/s (at z=0), and
fouling spots on the plate with 55/55 degree configuration
1207

---

World Academy of Science, Engineering and Technology 53 2009
Flow Inlet
Contact Point
y-z
plane
z-x
plane
Fig. 4(b) Flow distributions on the z-x, and y-z planes in the case of inlet velocity equalled to 0.1132 m/s (at z=0), and
fouling spots on the plate with 10/10 degree configuration
1208

---

World Academy of Science, Engineering and Technology 53 2009


10 degree angle
55 degree angle
Contact Point
Flow Inlet
10 degree angle
55 degree angle
z-x plane
y-z plane
Fig. 4(c) Flow distributions on the z-x, and y-z planes in the case of inlet velocity equalled to 0.1132 m/s (at z=0), and
fouling spots on the plate with 55/10 degree configuration
In order to investigate the influence of flow velocity, the
10/10 of x direction that all points had low fluid velocity
inlet velocity was varied as 0.0566, 0.1132, and 0.1698 m/s in
due to fluid distribution occurred at different direction.
the 55/55 degree configuration. Fig. 5(a)-5(c) presented the
Fig. 6(c) presented 5 points at different fluid velocity of
flow field and matched the simulated flow field with the plate angle 55/10 of x direction that point 3 shown
images obtained from experiments. The overall pattern of minimum fluid velocity due to this angle break up fluid
fouling sites was similar among the three cases, but the flow.
amount of fouling decreased with the increased inlet velocity
Fig. 7(a) presented 5 points at different fluid velocity of
i.e., in Fig. 5(a) at the lowest flow of 0.0566, fouling occurred
plate angle 55/55 of y direction. The results were shown
more, and fouling became lesser at 0.1132, and 0.1698 m/s as
that low fluid distribution occurred at all point of this
seen in Fig. 5(b) and 5(c), respectively. At low flow rate, there
angle due to fluid distribution occurred at different
was insufficient shear to overcome a fouling process, whereas
direction. Fig. 7(b) presented 5 points at different fluid
a high shear available at high flow rate was sufficient to velocity of plate angle 10/10 of y direction that point 1,2,
prevent or lessen fouling formation.
and 3 represented low fluid distribution and point 4 and 5
Fig. 6(a) presented 5 points at different fluid velocity of
represented high fluid distribution because these points
plate angle 55/55 of x direction. The results were shown that
were close to contact point. Fig. 7(c) presented 5 points at
low fluid velocity was found at point 2, 4 and 5 (contact point
different fluid velocity of plate angle 55/10 of y direction
area) and high fluid velocity was found at point 1 and 3. Fig.
that all points had low fluid velocity due to fluid
6(b) presented 5 points at different fluid velocity of plate angle
distribution occurred at different direction.
1209

---

World Academy of Science, Engineering and Technology 53 2009
Fig. 8(a) presented 5 points at different fluid velocity of
distribution. Fig. 8(c) presented 5 points at different fluid
plate angle 55/55 of z direction. The results were shown that
velocity of plate angle 55/10 of z direction that point 1, 2
high fluid distribution occurred at all point of this angle. Fig.
and 3 shown low fluid distribution and point 4 and 5
8(b) presented 5 points at different fluid velocity of plate
shown high fluid distribution. Fig. 9(a)-9(c) illustrated the
angle 10/10 of z direction that point 1, 2 and 3 shown low
averaged velocity for the corresponding crossing angles.
fluid distribution and point 4 and 5 shown high fluid
Flow Inlet
(a)

0.0566 m/s
Flow Inlet
(b)
0. 1132 m/s
Flow Inlet
(c)
0. 1698 m/s
Fig. 5 Matching the simulated flow field and fouling formation in the 55/55 degree plate configuration at (a) 0.0566
m/s (b) 0.1132 m/s, and (c) 0.1698 m/s.
1210

---

World Academy of Science, Engineering and Technology 53 2009
Contact Point
Fouling
(a)
Contact Point
Fouling
(b)
Contact Point
Fouling
(c)
Fig. 6 Distribution of the velocity component in x-direction of each plate configuration: (a) 55/55,
(b) 10/10, and (c) 55/10
1211

---

World Academy of Science, Engineering and Technology 53 2009
Contact Point
Fouling
(a)
Contact Point
Fouling
(b)
Contact Point
Fouling
(c)
Fig. 7 Distribution of the velocity component in y-direction of each plate configuration: (a) 55/55,
(b) 10/10, and (c) 55/10
1212

---

World Academy of Science, Engineering and Technology 53 2009


Contact Point
Fouling
(a)
Contact Point
Fouling
(b)
Contact Point
Fouling
(c)
Fig. 8 Distribution of the velocity component in z-direction of each plate configuration: (a) 55/55,
(b) 10/10, and (c) 55/10
1213

---