Académique Documents
Professionnel Documents
Culture Documents
= ] uJA
A
The mean overall heat transfer coefficient is defined as
U
m
=
1
A
] uJA
A
Similarly we can define a mean temperatue difference (MTD) AT
m
as
1
ATm
=
1
]
d
A1
42
Where Q is the total heat transferred from one fluid to the other in the heat exchanger.
By eliminating ]
d
A1
we get,
Q = U
m
AAT
m
4.1.3. Capacity Ratio
Capacity Ratio is an important parameter in a heat exchanger. It is the ratio of heat
capacity of cold fluid to hot fluid or vice versa. Accordingly
R
1
=
mcCpc
mhCph
,
R
2
=
mhCph
mcCpc
,
R
1
= 1 / R
2
Where m is the mass flow rate and C
p
is the specific heat rate. Suffixes h and c indicate
hot or cold fluid respectively
4.1.4. Temperature Effectiveness (P)
Temperature effectiveness tells about the performance of a heat exchanger with respect
to temperature alone. It is the ratio of temperature difference that one fluid undergoes to
the maximum temperature prevailing across the heat exchanger. Accordingly,
P
1
= (T
c,out
T
c,in
) / (T
h,in
- T
c,in
)
P
1
= (T
h,in
- T
h,out
)/ (T
h,in
- T
c,in
)
43
4.1.5. Effectiveness of Heat Exchanger ()
The maximum amount of heat transfer (Q
max
) that can occur between two streams in a
countercurrent heat exchanger is that for which the outlet temperature of the stream
with the lowest mC
p
reaches the inlet temperature of the other stream. This case is
illustrated schematically in the diagram. If the cold stream has a value of mC
p
greater
than the hot stream then the maximum heat transfer occurs when the hot stream is
cooled to the inlet temperature of the cold stream. When mC
p
of the hot stream is
higher, the cold fluid will be heated to the inlet temperature of the hot stream. This is
because of the fact that heat balance should be maintained under such condition which
can be given by,
Q
max
= (mC
p
)
min
AT
max
where (mC
p
)
min
is the lower of the two for the respective streams and AT
max
is the
difference between the stream inlet temperature.
AT
max
= (T
h,in
T
c,in
)
To achieve maximum heat transfer, an infinite surface area for the heat exchanger will
be required because the temperature difference approaches zero at the end of the heat
exchanger at which the end temperatures become equal.Effectiveness of a heat
exchanger is defined as the raio of actual to maximum heat transfer rates.
c =
mux
Because Q= U
m
AAT
m
, we see that c is given by
44
c =
UmAATm
(mCpmIn )ATmax
It follows that
c =
mhCph(1h,n-1h,out)
(mCpmIn )(1h,n-1c,n)
=
mcCpc(1c,out-1c,n)
(mCpmIn )(1h,n-1c,n)
4.1.6. Number Of Transfer Units
It is often convenient to use the number of transfer units as the basis of design. The
NTU values for the respective streams are
NTU
h
=
A0
mhCph
NTU
c
=
A0
mcCpc
where A is the total heat exchanger area and U is the overall heat transfer
coefficient.We may also define NTU
min
as the NTU of the stream having minimum
mC
p
.
NTU
min
=
A0
(mCp)mn
4.2. HEAT EXCHANGER BASIC ANALYSIS METHODS
Depending on the nature of the problem and available data, a method of analysis of the
heat exchanger can be selected.
43
The following are some of the commonly used analysis methods:
The LMTD-F method
The c-NTU method
The P-NTU method
The 0-P method
F- 0-P-NTU method
P-R Combination method
As only the inlet temperatures of the streams and their specific heats are known ,the
analysis is carried out using Effectiveness NTU method.
4.3. THE -NTU METHOD
The LMTD approach to heat exchanger analysis is useful when the inlet and outlet
temperatures are known or easily deermined. The LMTD is then easily determined and
the heat flow can thus be obtained. When the inlet or exit temperatures are to be
evaluated, the analysis frequently involves an iterative procedure because of the
logarithmic function in LMTD. In these cases , the analysis is performed more easily
by utilizing a method based on the effectiveness of the heat exchanger.The
effectiveness method also offers many advantages for analyss of problems in which a
comparison between various types of heat exchangers is to be made.
The heat exchanger effectiveness is defined as
Effectiveness= c =
ActuuI hcut tuns]c
Muxmum possbIc hcut tuns]c
46
The maximum possible heat transfer rate could be achieved in a counter flw heat
exchanger of infinita length.In such an exchange , one of the fluids would experience
the maximum possible temperature difference, Ib, in -Ic, in
If m
c
C
pc
<m
h
C
ph
, thecold fluid would experience a larger temperature change and as the
heat exchanger length is infinite, it would be heated to the inlet temperature of the hot
fluid, ie, T
c,out
=T
h,in
Hence, q
max
=m
c
C
pc
(T
hi
-T
ci
)
Similaly if m
h
C
ph
< m
c
C
pc
, the hot fluid would experience the larger temperature
change and would be cooled to the inlet temperature of the cold fluid in which case we
have,
q
max
=m
h
C
ph
(T
hi
-T
ci
)
Therefore we can write,
c =
mhCph(1h,n-1h,out)
(mCpmIn )(1h,n-1c,n)
=
mcCpc(1c,out-1c,n)
(mCpmIn )(1h,n-1c,n)
The NTU values for the respective streams are
NTU
h
=
A0
mhCph
NTU
c
=
A0
mcCpc
We may also define NTU
min
as the NTU value corresponding to the stream having the
minimum mC
p.
47
NTU
min
=
A0
(mCp)mn
Now , c =
QmIn
=
A0ATm
(mCp)mnATmax
= NTU
min
ATm
ATmax
Usually, NTU
min
is defined as NTU.
For a Counter Flow Heat Exchanger, the effectiveness in terms of NTU is obtained as,
c =
j1-cxp _-N10]1-(
Cmin
Cmcx
)_[
j1-(
Cmin
Cmcx
)cxp_-N10]1-(
Cmin
Cmcx
)_[
The value of this expression becomes maximum when NTU is infinity for the given
values of C
min
and C
max
.
4.4. FINS OR EXTENDED SURFACES
In the study of heat transfer, a fin is a surface that extends from an object to increase
the rate of heat transfer to or from the environment by increasing convection. The
amount of conduction, convection, or radiation of an object determines the amount of
heat it transfers. Increasing the temperature difference between the object and the
environment, increasing the convection heat transfer coefficient, or increasing the
surface area of the object increases the heat transfer. Sometimes it is not economical or
it is not feasible to change the first two options. Adding a fin to an object, however,
increases the surface area and can sometimes be an economical solution to heat transfer
problems.
48
The knowledge of temperature distribution along the fin is necessary for the proper
design of fins. The mathematical analysis for finding out the temperature distribution
and heat flow is discussed below-
4.4.1. Simplified Case
To create a simplified equation for the heat transfer of a fin, many assumptions need to
be made.
Assume:
1. Steady state
2. Constant material properties (independent of temperature)
3. No heat transfer
4. No internal heat generation
5. One-dimensional conduction
6. Uniform cross-sectional area
7. Uniform convection across the surface area
The fin analysis can be carried out using the basic Fourier conduction equation.
Fouriers law states that
Q
x
= - kA
c
d1
dx
49
where A
c
is the cross-sectional area of the differential element. Therefore the
conduction rate at x+dx can be expressed as
Hence, it can also be expressed as
.
Since the equation for heat flux is
then dq
conv
is equal to
where A
s
is the surface area of the differential element.
By substitution it is found that
This is the general equation for convection from extended surfaces. Applying certain
boundary conditions will allow this equation to simplify.
The ab
where
convec
The so
0(x) =
where
and
The co
four c
The bo
length
above equatio
re P is the pe
vection from e
solution to the
= C
1
e
mx
+ C
2
e
re
constants C
1
cases have th
boundary con
th of the fin.
tion will simpl
perimeter of
extended surf
.
the simplified
e
mx
.
and C
2
can b
the boundary
ondition at x
plify because
f the cross-se
urfaces with co
.
d equation is
n be found by
ry condition T
= L, howeve
30
se the area is c
sectional area
constant cros
y applying th
T(x = 0) = T
ver, is differe
s constant and
rea. Thus, the
oss-sectional a
the proper bou
T
b
for the tem
rent for all of
nd
he general eq
l area simplifi
oundary cond
temperature a
of them, wher
equation for
ifies to
nditions. All
at the base.
here L is the
31
1ab|e 4.1. 1emperature D|str|but|on and neat 1ransfer kate for f|ns of Un|form Cross Sect|ona| Area
4.4.2. Fin Performance
Fin performance can be described in different ways.
Fin effectiveness-It is the ratio of the fin heat transfer rate to the heat transfer
rate of the object if it had no fin. c=
Hcut Iost wth ]n
Hcut Iost wthout ]n
c=
Kmhj1+][
mK
h
tanhmL[
][
mK
h
+tanhmL
where , m= PbKAc
2
= bKo
2
, (for rectangular fin)
L= Height of fin
2o= Thickness of fin
32
P=Fin perimeter
In this case, a non-dimensional number named as Biot Number is defined.
Biot number B
i
=
h6
K
=
IntcnuI csstuncc o] ]n mutcuI
LxtcnuI csstuncc o] ]Iud on ]n su]ucc
where the thermal conductivity K refers to the conducting body.
The value of Biot number directly affects the fin effectiveness.
1. If B
i
=1
Then, c =1. So there is no use of putting the fins.
2. If B
i
>1
Then, c <1. So there is an adverse effect on heat flow and the fin works as an
insulator.
3. If B
i
<1
Then, c >1. This is the most desirable case.This is possible by using the high
conductivity material like Cu or Al.
Fin Efficiency- This is the ratio of the fin heat transfer rate to the heat transfer
rate of the fin if the entire fin were at the base temperature.It should be noted
that because of the thermal resistance of the fin material, the temperature of the
fin surface away from the base decreases.Hence the lateral area of the fin near
the base is more effectively used than near its tip.If the fin is made of a material
of high thermal conductivity , the value of B
i
is very small and the whole fin
area is nearly at the base temperature and the fin s said to be fully effective.
33
q
f
=
Hcut Iost ]om thc ]n
Hcut Iost ]om ]n ] thc whoIc ]n su]ucc s ut thc oot tcmpcutuc
If the heat lost from the end surfaces and edges of the fin is neglected , then,
q
f
=
tanhmL
mL
, where L= Height of fin.
Fin Uses
Fins are most commonly used in heat exchanging devices such as radiators in
cars and heat exchangers in power plants. They are also used in newer
technology such as hydrogen fuel cells. Nature has also taken advantage of the
phenomena of fins. The ears of jackrabbits act as fins to release heat from the
blood that flows through them.
4.5. HEAT TRANSFER COEFFICIENT
The heat transfer coefficient is used in calculating the convection heat transfer
between a moving fluid and a solid in thermodynamics. The heat transfer coefficient is
often calculated from the Nusselt numberThere are different heat transfer relations for
different liquids, flow regimes, and thermodynamic conditions. A common example
pertinent to many of the necessary power plant efficiency and thermal hydraulic
calculations is the Dittus-Boelter heat transfer correlation, valid for water in a circular
pipe with Reynolds numbers between 10 000 and 120 000 (in the turbulent pipe flow
range) and Prandtl numbers between 0.7 and 120.
4.5.1.
The N
transfe
if just
transfe
directi
where
Nusselt Nu
Nusselt num
sfer from a su
st conduction
sfer when conv
ction
re
L = charac
Area of the
umber
mber is a dim
surface that oc
on occurred. T
onvection take
racteristic leng
the body (usef
imensionless n
occurs in a 're
. Typically it
kes place.
ngth, which i
eful for more
34
s number that
'real' situation
it is used to m
in
is simply Vo
re complex sha
at measures the
on, compared
o measure the
in perpendi
Volume of the
hapes)
the enhanceme
d to the heat t
he enhanceme
dicular to
the body divid
ment of heat
t transferred
ent of heat
the flow
vided by the
33
k
f
= thermal conductivity of the "fluid"
h = convection heat transfer coefficient
4.5.2. Prandtl Number
The Prandtl number is a dimensionless number approximating the ratio of momentum
diffusivity (viscosity) and thermal diffusivity. It is named after Ludwig Prandtl.
It is defined as:
where:
v is the kinematic viscosity, = / .
u is the thermal diffusivity, = k / ( c
p
).
Typical values for Pr are:
around 0.7 for air and many other gases,
around 7 for water
around 710
21
for Earth's mantle
between 100 and 40,000 for engine oil,
between 4 and 5 for R-12 refrigerant
around 0.015 for mercury
36
4.5.3. Reynolds Number
In fluid mechanics, the Reynolds number is the ratio of inertial forces (vsp) to viscous
forces (/L) and consequently it quantifies the relative importance of these two types of
forces for given flow conditions. Thus, it is used to identify different flow regimes,
such as laminar or turbulent flow.
It is one of the most important dimensionless numbers in fluid dynamics and is used,
usually along with other dimensionless numbers, to provide a criterion for determining
dynamic similitude. When two geometrically similar flow patterns, in perhaps different
fluids with possibly different flowrates, have the same values for the relevant
dimensionless numbers, they are said to be dynamically similar.
It is named after Osborne Reynolds (18421912), who proposed it in 1883. Typically it
is given as follows:
where:
* vs - mean fluid velocity,
* L - characteristic length,
* - (absolute) dynamic fluid viscosity,
* v - kinematic fluid viscosity: v = / p,
* p - fluid density.
37
For flow in pipes for instance, the characteristic length is the pipe diameter, if the cross
section is circular, or the hydraulic diameter, for a non-circular cross section.
Laminar flow occurs at low Reynolds numbers, where viscous forces are dominant, and
is characterized by smooth, constant fluid motion, while turbulent flow, on the other
hand, occurs at high Reynolds numbers and is dominated by inertial forces, producing
random eddies, vortices and other flow fluctuations.
The transition between laminar and turbulent flow is often indicated by a critical
Reynolds number (Recrit), which depends on the exact flow configuration and must be
determined experimentally. Within a certain range around this point there is a region of
gradual transition where the flow is neither fully laminar nor fully turbulent, and
predictions of fluid behaviour can be difficult. For example, within circular pipes the
critical Reynolds number is generally accepted to be 2300, where the Reynolds number
is based on the pipe diameter and the mean velocity vs within the pipe, but engineers
will avoid any pipe configuration that falls within the range of Reynolds numbers from
about 2000 to 3000 to ensure that the flow is either laminar or turbulent.
For flow over a flat plate, the characteristic length is the length of the plate and the
characteristic velocity is the free stream velocity. In a boundary layer over a flat plate
the local regime of the flow is determined by the Reynolds number based on the
distance measured from the leading edge of the plate. In this case, the transition to
turbulent flow occurs at a Reynolds number of the order of 105 or 106.
38
4.6. DOUBLE PIPE HEAT EXCHANGER
The double pipe heat exchanger is one of heat exchangers. This type of heat exchanger
is also called concentric tube/pipe, jacketed-tube or hair pin: heat exchanger. The
areas where double pipe heat exchangers are used in smaller duty, usually less than
5000kW where the heat transfer surface requirement is not so high as well the mass
flow rate of the shell fluid is low.
In a variety of applications the double pipe heat exchangers are found to be
advantageous. The advantages of double pipes heat exchangers are many folded such as
:
Counter current operation: The counter flow operation of double pipe heat
exchanger makes it most efficient heat exchanger for a given NTU and R.
Elimination of cocurency in multipass operation: it offers a unique design of No
of tube passes = No of shell passes.
Simplicity of construction and manufacturing: The double pipe heat exchanger
is one of the simplest constructions in heat exchanger family. Is manufacturing
is rather simple because it is simply joining together the standard available pats
such as tubes, flauges, valves.
Application to high temperature and pressure: when compared to the other
nearly counterflow heat exchangers it gives a very high pressure and
temperature range.
Ease of Maintenance: Due to simple construction and flanged joint it is easier to
access the heat exchangers and mechanical cleaning.
39
Enhancement : Heat transfer on the shell side can be enhanced by the finned
construction particularly for gas liquid application.
Cost and Flexibility: Due to simplicity it is cheaper and can be down or up sized
for different applications.
4.7. FOULING IN HEAT EXCHANGERS
Fouling is generally defined as the accumulation of unwanted materials on the surfaces
of processing equipment. It has been recognized as a nearly universal problem in design
and operation and affects the operation of equipment in two ways:
The fouling layer has a low thermal conductivity. This increases the resistance
to heat transfer and reduces the effectiveness of heat exchangers increasing
temperature
As deposition occurs, the cross-sectional area is reduced, which causes an
increase in pressure drop across the apparatus
60
CHAPTER 5: THERMAL DESIGN PROCEDURE
5.1. ANALYSIS OF DOUBLE PIPE HEAT EXCHANGERS
5.1.1. Simple Flow Configurations
The double pipe heat exchanger selected is a counterflow heat exchanger. Hence, the
correlations for LMTD and c for counterflow heat exchangers derived holds true. As
per the definition the mean temperature difference can be given by LMTD, thus
AT
m
= AT
lm
Therefore the heat transfer equation reduces to
Q = UA AT
LM
Where U is the overall heat transfer coefficient and A is the area on which heat transfer
coefficient is defined. In terms of the inlet and outlet temperatures the heat transfer Q can be
written as
Q = UA
{(Th-Tco)-(Tho-Tc)]
In
Thi-Tcc
Thc-Tci
Where suffixes h and c stand for hot and cold fluids and i and o indicate inlet and outlet
conditions respectively.
But in our problem the outlet conditions are unknown, so the above method cannot be
directly applied. So we use the c NTU method as described in the previous chapter.
In the c NTU method, the effectiveness of double pipe heat exchanger is given by
61
c =
1-cxp |-N10(1-R)]
1-Rcxp|-N10(1-R)]
In recent times the P-R method has become very popular because it does not require
specification of the fluid with minimum heat capacity a priori. But we decided to go
with the usual c NTU method.
5.1.2. Heat Transfer Characteristics
In the above section, the overall heat transfer coefficient U was used without any
specific reference to its evaluation. Now we take up this task. So as per the definition of
overall heat transfer coefficient we can write
1
0oAo
=
1
hoAo
+
Ro
Ao
+
Rw
A
+
R
A
+
1
hA
where,
A
o
= outside area of the (plain) inner tube
A
i
= inside area of the inner tube
h
o
,h
i
= the outside and inside film heat transfer coefficient of the inside tube
defined on the
basis of A
o
and A
i
respectively
R
o
,R
i
= outside and inside fouling resistance for inner tube
R
w
= tube wall resistance =
d In (do d / )
2kw
k
w
= tube wall thermal conductivity
62
The tube resistance term can be determined for the case of steady state conduction
through the walls of a symmetric cylinder. The overall heat transfer coefficient U
o
is
defined on the outside surface area of the (plain) pipe A
o
. For double pipe configuration
heat transfer area can be put as
A = nJI
where L is the length of the tube surface and d is the corresponding diameter (d
o
for
outer and d
i
for inner surface giving A
o
and A
i
respectively). This reduces the Eqn.
(5.5)
1
0o
=
1
ho
+ Ro +
d In (do d / )
2kw
[
do
d
+
Rdo
d
+
1
h
[
do
d
Thus, the task reduces to determination of film transfer coefficient h
i
and h
o
. For tube
side coefficient well known correlation such as Dittus-Boelter equation can be used
(Chapter 4). However, a more frequently used correlation is the Sieder-Tate equation.
But we used the former for simplicity in calculations. The important difficulty in using
these correlations is the determination of fluid property because the equations
mentioned above suggest to use the fluid properties at average film temperature which
is the mean between bulk mean temperature of the fluid and the wall temperature. In
our problem the fluid temperature vary from inlet to outlet and the wall temperature is
not known. Hence, two approaches are possible.
First, calculations are carried out with properties at the mean bulk temperature
on each side where
T
meanbulk
=
1buIk,n + 1buIk,out
2
63
On the basis of result obtained the wall temperature at each end can be
calculated by temperature drop from the hot stream and temperature rise in the
cold stream. At the hot fluid inlet wall temperature may be approximately
calculated as
1wuII-1c,out
RcoId
=
1h,n-1wuII
Rhot
And at the cold fluid inlet the wall temperature is calculated as
1wuII-1c,n
RcoId
=
1h,out-1wuII
Rhot
Here the wall resistance is divided into two (may be equal) parts and added to
film and fouling resistance of each side to get R
cold
and R
hot
. With these wall
temperatures on both the ends the mean wall temperature can be calculated and
then the properties can evaluated at the mean film temperature given by
T
f,mean
=
1mcunbuIk+1mcunwuII
2
This has to be iterated and within two to three iterations and a good converged
result can be obtained.
If iteration is to be avoided we can assume that both U and AT vary linearly
within the heat exchanger. This gives an average heat transfer coefficient U
m
as
U
m
=
A(02AT1-U1AT2)
In[
U2AT1
U1AT2
Here the suffixes 1 and 2 refer to the ends of the heat exchanger.
64
5.1.3. Heat Transfer Coefficient of the Annular Side
For shell side heat transfer coefficient the equation for heat transfer coefficient in
annulus has to be used. For turbulent flow the same correlation can be used for tube
flow only the diameter should be replaced by equivalent diameter, d
e
.
Here, one important distinction has to be made between thermal and hydraulic
performance. The fluid friction for the annular space takes place at both the inner wall
of the outer tube (shell) and outer wall of the inner tube, whereas, heat transfer takes
place at the outer surface of the inner tube. Thus, for evaluating the heat transfer
coefficient of the annular side, the equivalent diameter is calculated as
d
e
=
4 PIow Acu
Pcmctc o] Hcut 1uns]c
=
4
n
4
(ds
2
-do
2
)
do
=
(ds
2
-do
2
)
do
Where d
s
= the inside diameter of the shell
d
o
= the outside diameter of the tube
For fluid flow and the definition of the Reynolds Number, the hydraulic diameter
should be used which is given by
d
h
=
4 PIow Acu
cttcd Pcmctc
=
4
n
4
(ds
2
-do
2
)
(ds+do)
= d
s
d
o
or annular gap width.
Thus, it must be kept in mind that
Re =
dh
63
where V = fluid mean velocity =
m
NPA
= kinematic viscosity
but
Nu
o
=
ho dc
ko
where, o indicates outer surface of inner tube and k
o
is the thermal conductivity of the
fluid in the annulus.
However, the above quantities are for unfinned units only. For finned construction the
details are given in design section later.
5.2. DESIGN OF LONGITUDINALLY FINNED DOUBLE PIPE
HEAT EXCHANGERS
In the last section we saw the analysis of simple unfinned tube heat exchanger for
understanding the process of heat transfer and important issues such as evaluation of
properties and heat transfer in annulus. This section describes design approach used in
the case of a finned construction as shown in Fig. (3.1).
Now let us define the previously defined quantities in the light of finned construction.
Hydraulic Mean Diameter, d
h
=
4 NPA
p
Thermal Equivalent Diameter, d
e
=
4 NPA
p- ds
66
where, NFA is the Net Flow Area for fluid flow in the annulus and W
p
is the Wetted
Perimeter. These parameters are given by
NFA =
ds
2
4
j
do
2
4
o[
W
p
= ad
s
+ ad
o
+ 2 N
f
H
f
N
f
o
f
where, the new quantities are
N
f
= number of fins
H
f
= fin height
o
f
= fin thickness
A typical finned tube with the geometrical parameters is shown in Fig. ().
1ab|e S.1. 1herma| Des|gn Data 1ab|e
Tube Outside Diameter (d
o
) in mm No. of Fins
25.4 20
25.4 20
48.3 36
48.3 36
60.3 40
67
48.3 36
60.3 40
73.0 48
With the assumption of absence of contact resistance between the tube and the fins, a
constant heat transfer coefficient over the entire finned length and fin Biot Number
along thickness small enough to consider it one dimentional, the fin efficiency can be
calculated as
q
f
=
tanhmH]
mH]
Kern (1965) calculated the entire heat transfer coefficient based on inner tube area but
Guy (1983) used total outside finned area which is more popular. Here we use the latter
method. Thus, the total area A
tot
is given by
A
tot
= A
f
+ A
b
where, A
f
= total fin surface area = 2 N
f
L
f
H
f
A
b
= unfinned bare tube area = L
f
(ad
o
-N
f
o
f
)
The total fin efficiency of the finned surface neglecting heat transfer from the fin tip is
given by
68
q
f
=
A]q]+Ab
Atot
= q
f
[
A]
Atot
+
[1 -
A]
Atot
Since the total efficiency affects the fin surface as well as the fouling surface, the
fouled surface heat transfer coefficient can be given by
1
0]s
= R
o
+
1
ho
where, R
o
is the fouling resistance on the fins.
The fin perimeter P
f
can be approximated as 2L
f
since L
f
>>o
f
. This reduces the value of
m to
m =
2 0]s
kw 6]
where, m
2
=
hP]
k]A]c
h = heat transfer coefficient based on finned area
P
f
= fin perimeter = 2(o
f
+L
f
)
L
f
= length of pipe/fin
A
fc
= cross-sectional area of the fin = L
f
o
f
k
f
= thermal conductivity of fin matrial
Thus, the total finned surface efficiency acts as a correlation factor to U
fs
based on the
total area. Thus, the overall heat transfer coefficient can be calculated as
69
1
0o
=
1
0]s
+ R
w
[
Atot
Ao
+ [Ri
1
h
Atot
A
where,
U
o
= overall heat transfer coefficient
R
w
=
do In (do d / )
2kw
= wall thermal resistance
R
i
= tube side fouling factor
Now the design of a double pipe unit can be carried out based on the above equations
and basic heat exchanger analysis methods explained in the previous chapter. The only
difference between the finned and unfinned construction being, in case of unfinned A
tot
A
o
and q
f
1. Hence, the procedure we used in the design of the longitudinally
finned double pipe heat exchanger is listed in step by step format in the next section.
The design we did was only for the rating or performance evaluation.
For rating purpose the information available are
The pipe dimensions (tube and shell)
The fin geometry and numbers (Fin Height was varied from 0 mm to maximum
possible for the given shell size)
The material data for pipes and fins
The inlet temperatures and property data tables for both the fluids
All the thermal data were determined at unfinned condition as well as at all possible fin
heights in 1mm steps. In the next chapter the computer program, developed using C++,
70
to perform this design will be explained. The method used at a particular fin height is
given below.
5.3. STEPS INVOLVED IN THE THERMAL DESIGN OF A
LONGITUDINALLY FINNED DOUBLE PIPE HEAT
EXCHANGER
Step 1: Calculation of Fin Geometrical Parameters:
Calculation of Net Flow Area (NFA):
NFA =
ds
2
4
j
do
2
4
o[
Calculation of Wetted Perimeter (W
p
):
W
p
= ad
s
+ ad
o
+ 2 N
f
H
f
N
f
o
f
Total Fin Surface Area (A
f
):
A
f
= 2 x N
f
L
f
H
f
Unfinned Bare Tube Area (A
b
):
A
b
= L
f
(nd
o
N
f
x o
f
)
Total Area (A
tot
)
A
tot
= A
f
+ A
b
Step 2: Calculation of d
e
and d
h
from Fin Geometrical Parameters:
Thermal Equivalent Diameter (d
e
):
d
e
=
4 NPA
p- ds
71
Hydraulic Mean Diameter (d
h
):
d
h
=
4 NPA
p
Step 3: Calculation Of Reynolds Number And Nusselt Number:
In the computation of the non dimensional parameters we need the property
values of fluids. Due to the fact that we do not know the value of the mean temperature,
the fluid properties are found out at the inlet temperature.
Fluid Mean velocity (V
o
):
V
o
=
m c
c NPA
Reynolds Number (Re):
Re
o
=
o dh
Nusslet Number (Nu):
Nu
o
= 0.023 x (Re
o
)
0.8
x (Pr
o
)
0.4
Step 4 : Calculation Of Heat Transfer Coefficient In The Shell Side:
h
o
=
Nuo k
dc
Step 5 : Calculation Of Heat Transfer Coefficient In The Tube Side:
Repeat all the procedures from Step 1 to Step 4 to determine the heat transfer
coefficient in the tube side (h
i
).
72
Step 6: Calculation of Fin Efficiency:
Fouled Heat Transfer Coefficient (U
fs
):
1
0]s
= R
o
+
1
ho
m =
2 0]s
kw 6]
Fin Efficiency (q
f
)
q
f
=
tanhmH]
mH]
Step 7: Calculation Of Overall Heat Transfer Coefficient:
The total fin efficiency of the finned surface neglecting heat transfer from the
fin tip (q
f
):
q
f
=
A]q]+Ab
Atot
= q
f
[
A]
Atot
+
[1 -
A]
Atot
Overall Heat Transfer Coefficient (U
o
):
1
0o
=
1
0]s
+ R
w
[
Atot
Ao
+ [Ri
1
h
Atot
A
Step 8: Calculation of NTU and R:
Number of Transfer Units (NTU):
NTU =
0o Atot
(mC)mn
73
Capacity Ratio (R):
(mC)
min
/(mC)
max
Step 9 : Calculation Of Effecetiveness Of Double Pipe Heat Exchanger:
Effectiveness of double pipe heat exchanger (c):
c =
1-cxp |-N10(1-R)]
1-Rcxp|-N10(1-R)]
Step 10 : Calculation of Net Heat Transfer:
Net Heat Transfer (Q):
Q = Effectiveness x Maximum Possible Heat Transfer
Q
max
= (mC)
min
x AT
max
Q = c x (mC)
min
x (T
hi
T
ci
)
4. Step 11: Outlet Temperatures of Both Fluids:
These can be obtained from the Heat Balance Equation.
Q = m
h
C
ph
x (T
hin
T
ho)
=
m
c
C
pc
x (T
co
T
ci)
Cold Fluid Outlet Temperature:
T
co
= T
ci
+
mCp
Hot Fluid Outlet Temperature:
T
ho
= T
hi
-
mhCph
This completes the performance evaluation of thermal aspect.
74
CHAPTER 6: OPTIMIZATION USING COMPUTER
PROGRAM
In the previous chapter we have seen the complete design procedure of a longitudinally
finned double pipe heat exchanger. Based on this procedure a computer program is
written in Turbo C++. Program code is given in Appendix A. The program is able to
perform the performance evaluation of any longitudinally finned double pipe heat
exchanger if the required geometrical data and fluid properties at inlet are provided.
I|gure 6.1. Data Input Mode Se|ect|on MLNU
6.1. DATA INPUT
The first section of the program deals with data input. In this part all the required
variables are defined. The program first displays the assumptions used in the design
procedure. Then a MENU is displayed which will allow the user to select the mode of
data input. Option 1 will allow the user to skip the manual data entering procedure and
go with a sample problem. Option 2 will permit the user to input the geometrical data
of the heat exchanger and the fluid and material properties. This can be entered by
referring to a Heat Transfer Data Book. Option 3 will exit the program. When the
73
sample problem is selected the program displays all the property values and
geometrical data assumed. This completes the data input stage.
I|gure 6.2. Samp|e rob|em Screen 1
I|gure 6.3. Samp|e rob|em Screen 2
76
I|gure 6.4. Samp|e rob|em Screen 3
I|gure 6.S. Manua| Data Input Screen 1
I|gure 6.6. Manua| Data Input Screen 2
77
6.2. THERMAL DESIGN
After all the required data about the heat exchanger are obtained the program starts to
perform the performance evaluation to obtain the heat transfer, outlet temperatures,
effectiveness etc. This procedure is performed first in the case of a unfinned
construction. Now program performs design for finned construction with the fin height
varying from 1mm to the maximum possible ie the annular spacing in steps of 1mm.
All the performance data are displayed at each step. After the steps are all complete the
output results are displayed as a table. The design procedure at every fin height consists
of two parts:
I|gure 6.7. Unf|nned neLx Ca|cu|ated kesu|ts
78
6.2.1. Fluid Calculations
In this the NFA is calculated first and then the W
p
. Using this the hydraulic mean
diameter and thermal equivalent diameter are determined. The flow velocity is
determined using the mass flow rate, density and NFA. Then the Reynolds Number is
determined and Dittus-Boelter Eqn. is used to determine the Nusselt Number. From the
Nusselt Number the film heat transfer coefficient is determined. This method is used to
calculate the heat transfer coefficients of the outer and inner surfaces. This will be
repeated at all fin heights.
6.2.2. Thermal Calculations
After determining the heat transfer coefficient, the fin efficiency is calculated and total
finned surface efficiency is determined. The heat transfer coefficients are combined
with the fouling resistances to get the fouled heat transfer coefficient. The wall thermal
resistance is then calculated. Then from inner and outer fouled heat transfer coefficients
and the wall thermal resistance the overall heat transfer coefficient is determined. This
U
o
with the total surface area of the finned pipe gives the NTU. From NTU, the Heat
Exchanger effectiveness is determined by the formula for effectiveness-NTU method.
The obtained effectiveness when multiplied with the maximum possible heat transfer
gives the net heat transfer. From this heat transfer value, the outlet temperature of both
hot and cold fluid is determined. This will conclude the thermal design for the
particular fin height. The process is repeated for all fin heights.
79
I|gure 6.8. Ca|cu|at|on kesu|ts at I|n ne|ght = 1Smm
6.3. RESULTS AND DISCUSSIONS
The U
o
, Heat Transfer, NTU, effectiveness and outlet temperatures of hot and cold
fluids at unfinned stage, and at all possible fin heights are displayed as a table. From
the table we can see that the overall heat transfer coefficient is continuously decreasing
since the surface area is increasing even though there is an increase in the heat transfer
coefficient. But the effectiveness and heat transfer values are found increasing. This is
because the small drop in U
o
is compensated by a large increase in area, A
tot
. The
effectiveness will not stop increasing within our size limits. So to find an optimum fin
height, we cant just take the point of maximum effectiveness.
80
1ab|e 6.9. 1he 1herma| Des|gn kesu|ts at var|ous f|n he|ghts
To solve this problem a graph is plotted with the fin height on the x-axis and
effectiveness on the y-axis. The graph shows that beyond a particular point the graph
becomes horizontal, ie. Effectiveness becomes almost constant. So this point is
assumed to be the optimum fin height. To confirm this assumption another graph is
plotted with the percentage increase in volume of the fin and pipe material and
percentage increase in heat transfer at each step along the y-axis and fin height along
the x-axis. The graph shows that the percentage increase in volume is almost constant
at all fin heights. But the percentage increase in heat transfer falls to almost zero at a
particular fin height and the percentage increase in volume is much higher than zero. So
beyond this point increasing fin height results more in wastage of material and thus
more cost than in increase of heat transfer. Therefore this point fixed as the optimum
fin height.
81
The final results obtained are given below:
I|gure 6.10. 1he I|na| kesu|t of the program
82
CHAPTER 7: THERMAL ANALYSIS USING ANSYS
7.1. FINITE ELEMENT METHOD
The Finite element method (FEM) is used for finding approximate solution of partial
differential equations (PDE) as well as of integral equations such as the heat transport
equation. The solution approach is based either on eliminating the differential equation
completely (steady state problems), or rendering the PDE into an equivalent ordinary
differential equation, which is then solved using standard techniques such as finite
differences, etc.
In solving partial differential equations, the primary challenge is to create an equation
that approximates the equation to be studied, but is numerically stable, meaning that
errors in the input data and intermediate calculations do not accumulate and cause the
resulting output to be meaningless. There are many ways of doing this, all with
advantages and disadvantages. The Finite Element Method is a good choice for solving
partial differential equations over complex domains (like cars and oil pipelines), when
the domain changes (as during a solid state reaction with a moving boundary), or when
the desired precision varies over the entire domain. For instance, in simulating the
weather pattern on Earth, it is more important to have accurate predictions over land
than over the wide-open sea, a demand that is achievable using the finite element
method.
83
7.2. ANSYS
ANSYS is the original (and commonly used) name for ANSYS Mechanical or ANSYS
Multiphysics, general-purpose finite element analysis software. ANSYS, Inc. actually
develops a complete range of CAE products, but is perhaps best known for ANSYS
Mechanical & ANSYS Multiphysics.
ANSYS Mechanical and ANSYS Multiphysics are self contained analysis tools
incorporating pre-processing (geometry creation, meshing), solver and post processing
modules in a unified graphical user interface. ANSYS is a general purpose finite
element modeling package for numerically solving a wide variety of mechanical
problems. These problems include: static/dynamic structural analysis (both linear and
non-linear), heat transfer and fluid problems, as well as acoustic and electro-magnetic
problems.
The software is used to analyze a broad range of applications. ANSYS Mechanical
incorporates both structural and material non-linearities. ANSYS Multiphysics includes
solvers for thermal, structural, CFD, electromagnetics, acoustics and can couple these
separate physics together in order to address multi-disciplinary applications. ANSYS
software is also used in Civil Engineering, Electrical Engineering, Physics and
Chemistry.
7.2.3. Analyzing Thermal Phenomena in ANSYS
A thermal analysis calculates the temperature distribution and related thermal
quantities in a system or component. Typical thermal quantities of interest are:
The temperature distributions
84
The amount of heat lost or gained
Thermal gradients
Thermal fluxes.
The basis for thermal analysis in ANSYS is a heat balance equation obtained from the
principle of conservation of energy. The finite element solution you perform via
ANSYS calculates nodal temperatures, then uses the nodal temperatures to obtain other
thermal quantities.The ANSYS program handles all three primary modes of heat
transfer: conduction, convection, and radiation.
7.2.4. Convection
You specify convection as a surface load on conducting solid elements or shell
elements. You specify the convection film coefficient and the bulk fluid temperature at
a surface; ANSYS then calculates the appropriate heat transfer across that surface. If
the film coefficient depends upon temperature, you specify a table of temperatures
along with the corresponding values of film coefficient at each temperature.
7.2.5. Types of Thermal Analysis
ANSYS supports two types of thermal analysis:
1. A steady-state thermal analysis determines the temperature distribution and
other thermal quantities under steady-state loading conditions. A steady-state
loading condition is a situation where heat storage effects varying over a period
of time can be ignored.
83
2. A transient thermal analysis determines the temperature distribution and other
thermal quantities under conditions that vary over a period of time.
In the following analysis, we are performing a steady state thermal analysis.
The results obtained from the C++ program has been used for modeling the heat
exchanger in Ansys. The thermal solid element 20 node 90 has been used for all the
models.
7.2.6. SOLID90 Element Description
SOLID90 is a higher order version of the 3-D eight node thermal element (SOLID70).
The element has 20 nodes with a single degree of freedom, temperature, at each node.
The 20-node elements have compatible temperature shapes and are well suited to model
curved boundaries. The 20-node thermal element is applicable to a 3-D, steady-state or
transient thermal analysis. If the model containing this element is also to be analyzed
structurally, the element should be replaced by the equivalent structural element (such
as SOLID95). Convection or heat flux (but not both) and radiation may be input as
surface loads at the element faces as shown by the circled numbers on figure. Heat
generation rates may be input as element body loads at the nodes
I|gure 7.1. So||d 90 e|ement
86
7.2.7. SOLID90 Input Summary
Nodes
I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z, A, B
Degrees of Freedom
TEMP
Real Constants
None
Material Properties
KXX, KYY, KZZ, DENS, C, ENTH
Surface Loads
Convection or Heat Flux (but not both) and Radiation (using Lab = RDSF) --
face 1 (J-I-L-K), face 2 (I-J-N-M), face 3 (J-K-O-N),
face 4 (K-L-P-O), face 5 (L-I-M-P), face 6 (M-N-O-P)
Body Loads
Heat Generations --
HG(I), HG(J), HG(K), HG(L), HG(M), HG(N), HG(O), HG(P), HG(Q), HG(R),
HG(S), HG(T), HG(U), HG(V), HG(W), HG(X), HG(Y), HG(Z), HG(A), G(B)
87
7.2.8. SOLID90 Output Data
Table 7.1. SOLID90 Element Output Definitions
Label Definition
EL Element Number
NODES Nodes - I, J, K, L, M, N, O, P
MAT Material number
VOLU: Volume
XC, YC, ZC Location where results are reported
HGEN Heat generations HG(I), HG(J), HG(K), HG(L), HG(M),
HG(N), HG(O), HG(P), HG(Q), ..., HG(Z), HG(A), HG(B)
TG:X, Y, Z,
SUM
Thermal gradient components and vector sum at centroid
TF:X, Y, Z,
SUM
Thermal flux (heat flow rate/cross-sectional area) components
and vector sum at centroid
FACE Face label
NODES Corner nodes on this face
AREA Face area
HFILM Film coefficient
TAVG Average face temperature
TBULK Fluid bulk temperature
88
Label Definition
HEAT RATE Heat flow rate across face by convection
HEAT
RATE/AREA
Heat flow rate per unit area across face by convection
HFLUX Heat flux at each node of face
HFAVG Average film coefficient of the face
TBAVG Average face bulk temperature
HFLXAVG Heat flow rate per unit area across face caused by input heat
flux
7.3. BUILDING THE MODEL
To build the model, you specify the jobname and a title for your analysis. Then, you
use the ANSYS preprocessor (PREP7) to define the element types, element real
constants, material properties, and the model geometry
1. To specify element types, you use the following:
GUI: Main Menu> Preprocessor> Element Type> Add/Edit/Delete
2. To define constant material properties, use either of the following:
GUI:
Main Menu> Preprocessor> Material Props> Material Models>
Thermal
89
7.3.1. Creating Model Geometry
I|gure 7.2. Cross sect|on
I|gure 7.3. Lxtruded mode|
The model is created either from primitives or can be directly modeled in Ansys or can
even be imported from other softwares. Here, the cross section is first modelled and
then extruded to get the required geomety.
7.3.2. Defining the Analysis Type
During this phase of the analysis, you must first define the analysis type:
90
In the GUI, choose menu path Main Menu Solution> Analysis Type> New Analysis>
Steady-state (static).
The geometry is then created using the preprocessor
7.4. MESHING
The model is then meshed using the mesh tool or using smart size option. A free mesh
will form an unstructured grid whereas a mapped mesh will form a structured grid.In
mapped meshing ,the element size is to be specified whereas in free mesh , ANSYS
automatically decides the mesh size.
I|gure 7.4. Meshed mode|
91
7.5. APPLYING LOADS
You can apply loads either on the solid model (keypoints, lines, and areas) or on the
finite element model (nodes and elements). You can specify loads using the
conventional method of applying a single load individually to the appropriate entity, or
you can apply complex boundary conditions as tabular boundary conditions
1. Convections (CONV)
Convections are surface loads applied on exterior surfaces of the model to account for
heat lost to (or gained from) a surrounding fluid medium. They are available only for
solids and shells. In line-element models, you can specify convections through the
convection link element
Main Menu> Solution> Define Loads> Apply> Thermal> Convection
2. neat I|uxes (nILUk)
Heat fluxes are also surface loads. Use them when the amount of heat transfer across a
surface (heat flow rate per area) is known, or is calculated through a FLOTRAN CFD
analysis. A positive value of heat flux indicates heat flowing into the element. Heat flux
is used only with solids and shells. An element face may have either CONV or HFLUX
(but not both) specified as a surface load. If you specify both on the same element face,
ANSYS uses what was specified last
Main Menu> Solution> Define Loads> Apply> Thermal> Heat Flux
92
7.6. SOLUTION
Once meshing is done and the loads are applied, the fem model can be solved using the
solve command.
To start the solution, use the following:
GUI: Main Menu> Solution> Solve> Current LS
7.7. POSTPROCESSING
After the finite element model is solved results are retrieved using the postprocessor of
Ansys.
To display your results, use the following menu paths.
For contour displays:
GUI:
Main Menu> General Postproc> Plot Results> Contour Plot> Element Solu
Main Menu> General Postproc> Plot Results> Contour Plot> Elem Table
Main Menu> General Postproc> Plot Results> Contour Plot> Nodal Solu
Contour plots show the variation of the selected parameter on the geometry itself .In
addition to contour plots, vector plots can also be obtained. Graphs can be plotted by
defining paths and mapping the required item onto the path.
The results can be listed using the following GUI path:
For table listings:
93
GUI:
Main Menu> General Postproc> List Results> Element Solution
Main Menu> General Postproc> List Results> Nodal Solution
7.7.1. Reviewing Analysis Results
ANSYS writes the results from a thermal analysis to the thermal results file,
Jobname.RTH. Results contain the following data:
Primary data
Nodal temperatures (TEMP, TBOT, TE2, TE3, . . . TTOP)
Derived data
Nodal and element thermal fluxes (TFX, TFY, TFZ, TFSUM)
Nodal and element thermal gradients (TGX, TGY, TGZ, TGSUM)
Element heat flow rates
Nodal reaction heat flow rates
7.8. PROBLEM DESCRIPTION
7.8.1. Pipe Without Fins
The inner tube of a tube-in-tube counter flow heat exchanger without fins is
analyzed.
Specifications
Pipe length 5cm
Inside diameter 6cm
94
Outside diameter 6.9cm
Fin type No fin
Loading
Convection on the inside surface with Film Coefficient=8334.7W/m
2
K
Fluid bulk temperature (inside) = 95
0
Celsius
Convection on the outside surface with Film Coefficient = 93.35 W/m
2
K
Fluid bulk temperature(outside) = 40
0
Celsius
Pipe is insulated at both ends, ie, Heat flux=0
7.8.2. Results
I|gure 7.S. Contour p|ot of noda| temperature- Lnd v|ew
93
I|gure 7.6. Contour p|ot of Noda| 1emperature
I|gure 7.7. Contour p|ot of Noda| heat f|ux
96
I|gure 7.8. Vector p|ot of 1herma| f|ux
I|gure 7.9. Vector p|ot of therma| grad|ent
97
PRINT ALONG PATH DEFINED BY LPATH COMMAND. DSYS= 0
***** PATH VARIABLE SUMMARY *****
S TEMP FLUX HEATFLOW
0.0000 94.306 5805.9 0.33825E-01
0.22500E-03 94.285 5766.2 0.22339E-01
0.45000E-03 94.263 5726.4 0.11674E-01
0.67500E-03 94.242 5686.6 0.18316E-02
0.90000E-03 94.221 5646.9 -0.71892E-02
0.11250E-02 94.200 5607.1 -0.15388E-01
0.13500E-02 94.179 5567.4 -0.22766E-01
0.15750E-02 94.158 5527.6 -0.29321E-01
0.18000E-02 94.137 5487.8 -0.35055E-01
0.20250E-02 94.117 5448.1 -0.39967E-01
0.22500E-02 94.096 5408.3 -0.44057E-01
0.24750E-02 94.076 5368.6 -0.47326E-01
0.27000E-02 94.056 5328.8 -0.49772E-01
0.29250E-02 94.036 5289.0 -0.51397E-01
0.31500E-02 94.017 5249.3 -0.52200E-01
98
0.33750E-02 93.997 5209.5 -0.52182E-01
0.36000E-02 93.978 5169.8 -0.51342E-01
0.38250E-02 93.958 5130.0 -0.49679E-01
0.40500E-02 93.939 5090.2 -0.47196E-01
0.42750E-02 93.920 5050.5 -0.43890E-01
0.45000E-02 93.901 5010.7 -0.39763E-01
The above values are plotted in the following graphs-
I|gure 7.10. I|ux vs kad|a| d|stance
99
I|gure 7.11. neat f|ow vs kad|a| d|stance
The following items are included in the element table:
Heat1-Heat flow rate across face1 of the element by convection
Heat2- Heat flow rate across face2 of the element by convection
Heat3- Heat flow rate across face3 of the element by convection
Heat4- Heat flow rate across face4 of the element by convection
Heat5- Heat flow rate across face5 of the element by convection
Heat6- Heat flow rate across face6 of the element by convection
Heat flow- Sum of heat flow by convection across all the faces
Volume- Volume of each element.
100
After defining the element table, sum of each item is calculated over all elements
and the results are listed below.
SUM ALL THE ACTIVE ENTRIES IN THE ELEMENT TABLE
TABLE LABEL TOTAL
HEAT1 23.2804
HEAT2 29.3213
HEAT3 34.9446
HEAT4 0.00000
HEAT5 21.5256
HEAT6 0.00000
HEATFLOW 109.072
VOLUME 0.455926E-04
All units are in SI system
7.8.3. Pipe with rectangular longitudinal fins
The inner tube of a tube-in tube counter flow heat exchanger with longitudinal fins of
rectangular cross section is analyzed.
Specifications
Pipe length 5cm
101
Inside diameter 6cm
Outside diameter 6.9cm
Fin type Longitudinal fins, rectangular cross section
Number of fins-40
Fin height-15mm
Fin thickness-0.9mm
Loading
Convection on the inside surface with Film Coefficient=8334.7W/m
2
K
Fluid bulk temperature (inside) = 95
0
Celsius
Convection on the outside surface with Film Coefficient = 268.13
W/m
2
K
Fluid bulk temperature(outside) = 40
0
Celsius
Pipe is insulated at both ends, ie, Heat flux=0
102
7.8.4. Results
I|gure 7.12. contour p|ot of Noda| 1emperature
I|gure 7.13. Contour p|ot of Noda| 1emperature
103
I|gure 7.14. Contour p|ot of 1herma| f|ux
I|gure 7.1S. Vector p|ot of 1herma| f|ux
104
PRINT ALONG PATH DEFINED BY LPATH COMMAND. DSYS= 0
***** PATH VARIABLE SUMMARY *****
S TEMP FLUX HEATFLOW
0.0000 88.232 56851. 0.53365E-01
0.97500E-03 87.316 55539. 0.43654E-01
0.19500E-02 86.405 54280. 0.22573E-01
0.29250E-02 85.509 68839. -0.18220E-02
0.39000E-02 84.178 0.13639E+06-0.12090E-02
0.48750E-02 81.616 0.18793E+06-0.26461E-02
0.58500E-02 78.166 0.20087E+06-0.52274E-02
0.68250E-02 75.059 0.18054E+06-0.62020E-02
0.78000E-02 72.284 0.16132E+06-0.65923E-02
0.87750E-02 69.811 0.14330E+06-0.54885E-02
0.97500E-02 67.620 0.12671E+06-0.20283E-02
0.10725E-01 65.688 0.11123E+06-0.34405E-02
0.11700E-01 63.998 96815. -0.46582E-02
0.12675E-01 62.534 83477. -0.19783E-02
0.13650E-01 61.283 70756. -0.19005E-02
103
0.14625E-01 60.232 58633. -0.30442E-02
0.15600E-01 59.372 47228. -0.27243E-02
0.16575E-01 58.695 36216. -0.12757E-02
0.17550E-01 58.194 25506. -0.29579E-02
0.18525E-01 57.864 15044. -0.20605E-02
0.19500E-01 57.704 4734.2 -0.96108E-03
The above values are plotted in the following graphs-
I|gure 7.7. 1emperature vs rad|a| d|stance
106
I|gure 7.8. neatf|ow vs kad|a| d|stance
I|gure 7.18. I|ux vs kad|a| d|stance
After defining the element table, sum of each item is calculated over all elements
107
and the results are listed below.
SUM ALL THE ACTIVE ENTRIES IN THE ELEMENT TABLE
TABLE LABEL TOTAL
HEAT1 106.352
HEAT2 124.293
HEAT3 745.511
HEAT4 0.00000
HEAT5 86.2822
HEAT6 0.00000
HEATFLOW 1062.44
VOLUME 0.725942E-04
7.8.5. Pipe With Triangular Longitudinal Fins
The inner tube of a tube-in tube counter flow heat exchanger with longitudinal fins of
triangular cross section is analyzed.
Specifications
Pipe length 5cm
Inside diameter 6cm
Outside diameter 6.9cm
Fin type Longitudinal fins, triangular cross section
108
Number of fins-40
Fin height-15mm
Fin thickness-0.9mm
Loading
Convection on the inside surface with Film Coefficient=8334.7W/m
2
K
Fluid bulk temperature (inside) = 95
0
Celsius
Convection on the outside surface with Film Coefficient = 268.13
W/m
2
K
Fluid bulk temperature(outside) = 40
0
Celsius
Pipe is insulated at both ends, ie, Heat flux=0
7.8.6. Results
I|gure 7.9. Contour p|ot of Noda| temperature
109
I|gure 7.10. Contour p|ot of 1herma| f|ux
I|gure 7.11. Vector p|ot of therma| f|ux
110
PRINT ALONG PATH DEFINED BY LPATH COMMAND. DSYS= 0
***** PATH VARIABLE SUMMARY *****
S TEMP FLUX HEATFLOW
0.0000 88.717 52774. 0.88081E-01
0.97500E-03 87.866 51601. 0.54954E-01
0.19500E-02 87.022 50430. 0.24177E-01
0.29250E-02 86.189 61966. 0.47707E-04
0.39000E-02 84.998 0.12340E+06-0.24159E-02
0.48750E-02 82.626 0.17467E+06-0.35495E-02
0.58500E-02 79.331 0.17290E+06-0.11868E-01
0.68250E-02 76.213 0.17856E+06-0.20279E-01
0.78000E-02 73.295 0.17452E+06-0.19466E-01
0.87750E-02 70.524 0.16539E+06-0.29327E-01
0.97500E-02 67.905 0.15633E+06-0.37053E-01
0.10725E-01 65.436 0.14727E+06-0.37315E-01
0.11700E-01 63.117 0.13808E+06-0.46430E-01
0.12675E-01 60.941 0.13012E+06-0.47067E-01
0.13650E-01 58.890 0.12248E+06-0.42218E-01
111
0.14625E-01 56.964 0.11484E+06-0.33039E-01
0.15600E-01 55.162 0.10723E+06-0.24990E-01
0.16575E-01 53.484 99754. -0.16915E-01
0.17550E-01 51.912 93384. -0.10818E-01
0.18525E-01 50.445 87015. -0.80580E-02
0.19500E-01 49.081 80646. -0.86362E-02
The above values are plotted in the following graphs-
I|gure 7.12. I|ux vs kad|a| d|stance
112
I|gure 7.13. neat f|ow vs kad|a| d|stance
I|gure 7.14. 1emperature vs kad|a| d|stance
After defining the element table, sum of each item is calculated over all elements
113
and the results are listed below.
SUM ALL THE ACTIVE ENTRIES IN THE ELEMENT TABLE
TABLE LABEL TOTAL
HEAT1 88.0925
HEAT2 102.742
HEAT3 696.007
HEAT4 0.00000
HEAT5 99.8752
HEAT6 0.00000
HEATFLOW 986.717
VOLUME 0.590911E-04
7.9. DISCUSSION
7.9.1. Pipe Without Fins
In the contour plot of nodal temperature , we find that as we proceed towads the
periphery of the pipe from the inside, there is a linear drop in temperature as is evident
from the temperature vs radial distance plot.The contour plot of Thermal flux as well as
Thermal gradient also show similar variation .The vector plot of thermal flux shows
that the thermal flux is directed radially outwards whereas thermal gradient vectors are
directed towards the inside.From the plot of Heatflow vs the radial distance , it is
114
found that heatflow is directed into the elements for the initial part of the radius, but as
we proceed outwards, the heat flow changes sign meaning that heat flow is directed
outwards.
7.9.2. Rectangular Finned Pipe
The contour plot of nodal temperature shows the variation of temperature in the finned
pipe.It is found that there is very little temperature drop in the pipe thickness and the
major drop is confined to the fins. Apart from that , the drop in temperature is
substantial as compared to an unfinned pipe of the same dimensions.The contour plot
of thermal flux is completely different from the temperature plot n this case. We find
that maximum thermal flux is found at the bottom parts of the fin . This can be
attributed to the sudden reduction in area of heat flow.The thermal gradient plot is
almost similar to the plot of flux.From the vecto plot of thermal flux, it is evident that
most of the flux lines are confined to the fins and very little heat flux is present in the
unfinned region between the fins.The thermal gradient vectors are directed towards the
inside as expected. The plot of temperature vs radial distance shows that the variation
in temperature is no longer linear but follows a curve. The plot of flux shows that the
maximum flux is present at the lower regions of the fins and it decreases as we move to
the top of the fin.Heat flow is found tobe positive initially and becomes negative as we
reach the fins because the heat starts to flow out of the elements and into the
surrounding fluid due of the presence of fins.
7.9.3. Pipe With Triangular Fins
The variation in temperature follows a curve similar to that of a rectangular finned case.
The flux is found to remain almost constant in the fins owing to the reduction in cross
113
sectional area towards the periphery. The vector plots in this case are similar to that of
the rectangular fins, but for the fact that they are closer together in the fins due to the
pointed shape.The heat flow is also identical to the rectangular case.
7.10. COMPARISON OF RESULTS
The sum of element table results for each case is given below-
1ab|e 7.2. Compar|son Var|ous Mode| Ana|ys|s kesu|ts
Rectangular fins Pipe (no fins) Triangular fins
TABLE LABEL TOTAL TOTAL TOTAL
HEAT1 106.352 23.2804 88.0925
HEAT2 124.293 29.3213 102.742
HEAT3 745.511 34.9446 696.007
HEAT4 0.00000 0.00000 0.00000
HEAT5 86.2822 21.5256 99.8752
HEAT6 0.00000 0.00000 0.00000
HEATFLOW 1062.44 109.072 986.717
VOLUME 0.725942E-04 0.455926E-04 0.590911E-04
We find that maximum heat flow for a given length of pipe and a given fin height is
obtained in case of rectangular cross section fins. Heat flow in the case of triangular
fins is comparable to that of rectangular fins. Compared to a bare pipe, the heatflow is
116
found to increase many times .This can be attributed to the increase in surface area of
heat transfer.
It is found that the use of triangular fins reduces the volume of material used by a
substantial amount without appreciable change in the heat transferred. As a result a
triangular fin provides the added advantage of reduced material as compared to
rectangular fins without much change in heat transfer. This is due to the fact that as we
move towards the top of a rectangular fin , the heat flux is found to decrease. This
means that the fin cross section towards the tip of the fin is not effectively utilized as
compared to the bottom part of the fin. To account for this, the cross sectional area is
reduced towards the periphery to obtain a triangular fin .A fin of parabolic profile can
also be used, but triangular profile is preferred owing to the ease of manufacture and
the cost involved.
117
CHAPTER 8: CONCLUSION
With the help of the Optimization Program and ANSYS software we have been able to
conclude that as the Fin height increases the rate of increase of Heat Transfer
Decreases. Moreover increasing the fin height does not necessarily result in an increase
in heat transfer. There is an optimum value of fin height above which further increase
in height does not aid the heat transfer process considerably. With the help of the C++
program we have been able to successfully determine this value of optimum fin height
for particular input conditions and fin thickness. We also obtained substantial increase
in the values of heat transfer coefficients and effectiveness of a heat exchanger when
fins were provided. This also demonstrates the fact that Fins provide a thermodynamic
advantage. Thereby designing a heat exchanger at the optimum fin height can lead to
reducing capital costs and increasing savings. Also providing cheap materials for the
fin and expensive durable materials for thinner pipes can increase Heat Exchanger Life
span and save capital costs as well. If outlet conditions are provided and fins are also
created then by virtue of the fins we can decrease the length of the heat exchanger thus
save material. Using triangular fins we can save more material and obtain heat transfer
close to that of rectangular fins.
Use of parabolic fins in heat exchangers can be analyzed in the future. In this analysis
we have considered fluid properties at bulk mean temperature, modeling in FLUENT
can assist in the design and analysis in transient conditions which will provide a
bonafide report on the advantages of fins and design of heat exchangers.
118
APPENDIX A: C++ PROGRAM CODE FOR THE
OPTIMIZATION OF FIN HEIGHT
// FINAL YEAR PROJECT DONE BY ALPHIN C. TOM, ARJUN RAMANATHAN & ARUN
KRISHNAN
//*****************************************************************************//
// DEPARTMENT OF MECHANICAL ENGINEERING
// COLLEGE OF ENGINEERING, TRIVANDRUM
//-----------------------------------------------------------------------------//
// OPTIMISATION AND ANALYSIS OF
// LONGITUDINALLY FINNED DOUBLE PIPE HEAT EXCHANGER
//-----------------------------------------------------------------------------//
#include <iostream.h>
#include <stdio.h>
#include <process.h>
#include <stdlib.h>
#include <conio.h>
#include <graphics.h>
#include <math.h>
#define pi 3.14
void main()
{
clrscr();
long double
inID,inOD,outID,length,fin_h,de,dh,Rei,Reo,Nui,Nuo,Uo[500],Ufs,m,fin_t,NFA,WetP,NFAi,kw,Af,Ab,
Atot[500],Ro,ho,hi,Ri,Rw;
int noffins;
char ch;
long double mh,mc,cpc,cph;
long double etaf,etaf1;
119
long double dens_o,dens_i,ko,ki,Pri,Pro,iflmeanvel,ikinvisco,oflmeanvel,okinvisco;
long double
Ao,Vol[500],NTU[500],eff[500],Heat[500],R,mC_min,mC_max,Thi,Tci,Tho[500],Tco[500];
textcolor(2);
clrscr();
gotoxy(32,5);cout<<"FINAL YEAR PROJECT";
gotoxy(24,18);cout<<"OPTIMISATION AND ANALYSIS OF";
gotoxy(15,20);cout<<"LONGITUDINALLY FINNED DOUBLE PIPE HEAT EXCHANGER";
gotoxy(13,16);cout<<"----------------------------------------------------";
gotoxy(13,22);cout<<"----------------------------------------------------";
cout<<"\n\n\n\n\t\t\t\t Done by";
cout<<"\n\n\n\t\t\t Alphin C. Tom S8M1 - 28106";
cout<<"\n\n\t\t\t Arjun Ramanathan S8M1 - 28110";
cout<<"\n\n\t\t\t Arun Krishnan S8M2 - 28210";
cout<<"\n\n\n\t\t\t\t Guided by";
cout<<"\n\n\t\t\t\t Dilip D.";
getch();
clrscr();
cout<<"\n\n\t\t\t\t ASSUMPTIONS";
cout<<"\n\t\t\t\t ---------------";
cout<<"\n\n\n\t\t1. Flow type is Counterflow";
cout<<"\n\n\t\t2. Hot fluid flows in the pipe (Inner Tube)";
cout<<"\n\n\t\t3. Cold fluid flows in the annulus (Outer Tube)";
cout<<"\n\n\t\t4. Heat transfer at the tip of fin is neglected";
getch();
MENU:
clrscr();
cout<<"\n\n\n\t Select the Mode of DATA Input";
cout<<"\n\t ---------------------------------";
cout<<"\n\n\n 1. Go with a SAMPLE PROBLEM";
cout<<"\n\n 2. Enter your own DATA";
cout<<"\n\n 3. Exit Program";
120
cout<<"\n\n\n Enter your choice : ";
cin>>ch;
switch(ch)
{
case '1': clrscr();
cout<<"\n\n\n\t SAMPLE PROBLEM DEFINITION : ";
cout<<"\n\t ---------------------------------";
cout<<"\n\n Hot Fluid is Water in the pipe";
cout<<"\n\n Cold Fluid is Ethylene Glycol in conterflow in the annulus";
cout<<"\n\n Geometric Values are : ";
cout<<"\n\n Pipe Inner Diameter = 6cm";inID=.06;
cout<<"\n\n Pipe Outer Diameter = 6.9cm";inOD=.069;
cout<<"\n\n Shell Inner Diameter = 11.7cm";outID=.117;
cout<<"\n\n Pipe Length = 114m";length=114.0;
cout<<"\n\n Fin Thickness = 0.9mm";fin_t=.0009;
cout<<"\n\n No. of Fins = 40";noffins=40;
cout<<"\n\n Massflow Rate of Water = 5kg/s";mh=5.0;
cout<<"\n\n Massflow Rate of Ethylene Glycol = 2kg/s";mc=2.0;
cout<<"\n\n Specific Heat of Water = 4205.5 J/kgK";cph=4205.5;
cout<<"\n\n Specific Heat of Ethylene Glycol = 2474 J/kgK";cpc=2474.0;
cout<<"\n\n\n\n Press any key to continue....";
getch();
clrscr();
Thi=95.0;
Tci=40.0;
cout<<"\n\n\n\t PROPERTIES OF WATER AT 95 degree celcius";
cout<<"\n\n Density = 967.5kg/m^3";dens_i=967.5;
cout<<"\n\n Kinematic Viscosity = .3285 x 10^-6 m^2/s";ikinvisco=.0000003285;
cout<<"\n\n Prandtl Number = 1.98";Pri=1.98;
cout<<"\n\n Thermal Conductivity = 0.675W/mK";ki=0.675;
cout<<"\n\n\n\t PROPERTIES OF ETHYLENE GLYCOL AT 40 degree celcius";
121
cout<<"\n\n Density = 1101kg/m^3";dens_o=1101.0;
cout<<"\n\n Kinematic Viscosity = 8.686 x 10^-6 m^2/s";okinvisco=.000008686;
cout<<"\n\n Prandtl Number = 93.00";Pro=93.00;
cout<<"\n\n Thermal Conductivity = .2559W/mK";ko=.2559;
break;
case '2': textcolor(7);
clrscr();
cout<<"\n\n\t\t\t THERMAL DESIGN DATA INPUT";
cout<<"\n\t\t\t ------------------------";
cout<<"\n\n\n Tube Inner Diameter (in m) = ";cin>>inID;
cout<<" \n Tube Outer Diameter (in m) = ";cin>>inOD;
cout<<" \n Shell Inner Diameter (in m) = ";cin>>outID;
cout<<" \n Total Length of the Pipe (in m) = ";cin>>length;
cout<<" \n Thickness of Fin (as per std) = 0.9 mm";fin_t=0.0009;
cout<<" \n\n No. of Fins (From Data Table) = ";
if (inOD>=.07)
noffins=48;
else
if (inOD>=0.06)
noffins=40;
else
if (inOD>=0.045)
noffins=36;
else
noffins=20;
cout<<noffins;
getch();
cout<<"\n\n\n Mass Flow Rate of the Hot Fluid (kg/s) = ";cin>>mh;
cout<<" \n Mass Flow Rate of th Cold Fluid (kg/s) = ";cin>>mc;
cout<<" \n\n Inlet Temp. of Hot Fluid (degree C) = ";cin>>Thi;
cout<<" \n Inlet Temp. of Cold Fluid (degree C) = ";cin>>Tci;
122
clrscr();
cout<<"\n\t\t\t FLUID PROPERTIES INPUT";
cout<<"\n\t\t\t --------------------------";
cout<<"\n\n Cold/Outer Fluid Properties at "<<Tci<<" degree celcius";
cout<<"\n---------------------------------------------------------------";
cout<<"\n\n Density (kg/m^3) : ";
cin>>dens_o;
cout<<"\n Kinematic Viscosity (m^2/s) : ";
cin>>okinvisco;
cout<<"\n Prandtl Number : ";
cin>>Pro;
cout<<"\n Thermal Conductivity (W/mK) : ";
cin>>ko;
cout<<"\n Specific Heat (J/kgK) : ";
cin>>cpc;
cout<<"\n\n Hot/Inner Fluid Properties at "<<Thi<<" degree celcius";
cout<<"\n ------------------------------------------------------------";
cout<<"\n\n Density (kg/m^3) : ";
cin>>dens_i;
cout<<"\n Kinematic Viscosity (m^2/s) : ";
cin>>ikinvisco;
cout<<"\n Prandtl Number : ";
cin>>Pri;
cout<<"\n Thermal Conductivity (W/mK) : ";
cin>>ki;
cout<<"\n Specific Heat (J/kgK) : ";
cin>>cph;
break;
case '3': exit(1);
default : goto MENU;
}
123
getch();
textcolor(7);
clrscr();
cout<<"\n\n\n\n The Outer Surface Fouling Resistance, Ro = 0.0001 m^2K/W";
Ro=0.0001;
cout<<"\n\n The Inner Surface Fouling Resistance, Ri = 0.0001 m^2K/W";
Ri=0.0001;
cout<<"\n\n Thermal Conductivity of Wall, kw = 60 W/mK";
cout<<"\n\n\n Wall material is assumed to be 1% Chrome Steel";
kw=60.0;
getch();
if (mh*cph<mc*cpc)
{
mC_min=mh*cph;
mC_max=mc*cpc;
}
else
{
mC_min=mc*cpc;
mC_max=mh*cph;
}
clrscr();
cout<<"\n\n NO FIN CALCULATIONS";
cout<<"\n-----------------------";
NFA=(pi*outID*outID/4.000)-(pi*inOD*inOD/4.00); //Net Flow Area
WetP=pi*outID+pi*inOD; //Wetted Perimeter
dh=4.000*NFA/WetP; //Hydraulic Mean Diameter
de=4.000*NFA/(WetP-pi*outID); //Heat Transfer Mean Diameter
NFAi=(pi*inID*inID)/4; // Net Flow Area for inner pipe
oflmeanvel=mc/(NFA*dens_o); //Mean Flow Velocity in the annulus
Reo=oflmeanvel*dh/okinvisco; //Reynolds Number in the annulus
124
Nuo=0.023*pow(Reo,0.8)*pow(Pro,0.4); //Nusselt Number in the annulus --- Dittus-Boelter equation
for heating
ho=Nuo*ko/de; //Outer Surface Convection Heat Tranfer Coefficient
iflmeanvel=mh/(NFAi*dens_i); //Mean Flow Velocity in the pipe
Rei=iflmeanvel*inID/ikinvisco; //Reynolds Number in the pipe
Nui=0.023*pow(Rei,0.8)*pow(Pri,0.3);//Nusselt Number in the pipe --- Dittus-Boelter equation for
cooling
hi=Nui*ki/inID; //Inner Surface Convection Heat Tranfer Coefficient
cout<<"\n\n Outer Surface Convection Heat Transfer Coefficient : ";
cout<<"\n\n\t ho = "<<ho<<" W/m^2K";
cout<<"\n\n Inner Surface Convection Heat Transfer Coefficient : ";
cout<<"\n\n\t hi = "<<hi<<" W/m^2K";
getch();
Ufs=1.000/(Ro+(1.000/ho)); //Outer surface convection with fouling
Atot[0]=Ao=pi*inOD*length; //Unfinned pipe outer surface area
Rw=(inOD*log(inOD/inID))/(2.000*kw); // Wall thermal resistance
Uo[0]=1.000/((1.000/Ufs)+Rw+(Ri+(1/hi))*(Ao/(pi*inID*length)));//Overall Heat Transfer Coefficient
NTU[0]=(Uo[0]*Ao)/mC_min; //Number of Transfer Units
Vol[0]=length*(pi*(inOD*inOD-inID*inID)/4); //Volume of the pipe material required
R=mC_min/mC_max; // ratio of specific heats
eff[0]=(1.000-exp(-NTU[0]*(1.000-R)))/(1.000-R*exp(-NTU[0]*(1.000-R)));//heat exchanger
effectiveness
Heat[0]=eff[0]*(mC_min*(Thi-Tci)); // Net Heat Transfer
Tho[0]=Thi-(Heat[0]/(mh*cph)); // Hot Fluid Outlet Temp.
Tco[0]=Tci+(Heat[0]/(mc*cpc)); // Cold Fluid Outlet Temp.
cout<<"\n\n\n\n NTU = "<<NTU[0]<<"\n\n Net Heat Transfer Area, Atot = "<<Atot[0]<<"
m^2"<<"\n\n Heat Exchanger Effectiveness = "<<eff[0]<<"\n\n Net Heat Transfer = "<<Heat[0]<<" W";
cout<<"\n\n Hot Inlet Temp = "<<Thi<<"\t Hot Outlet Temp = "<<Tho[0]<<"\n\n Cold Inlet Temp =
"<<Tci<<"\t Cold Outlet Temp = "<<Tco[0]<<"\n\n\n Overall Heat Transfer Coefficient, Uo =
"<<Uo[0];
getch();
fin_h=0.001;
clrscr();
for (int i=2;(2.000*fin_h+inOD)<=outID;i++)
123
{
clrscr();
cout<<"\n\n\n Step No. "<<i-1;
cout<<"\n\n Fin Height = "<<fin_h*1000<<" mm";
// FLUID FLOW CALCULATIONS
NFA=(pi*outID*outID/4.000)-((pi*inOD*inOD/4.00)+noffins*fin_h*fin_t); //Net Flow Area
WetP=pi*outID+pi*inOD+2.00*noffins*fin_h-noffins*fin_t; //Wetted Perimeter
dh=4.000*NFA/WetP; //Hydraulic Mean Diameter
de=4.000*NFA/(WetP-pi*outID); //Heat Transfer Mean Diameter
NFAi=(pi*inID*inID)/4; // Net Flow Area for inner pipe
oflmeanvel=mc/(NFA*dens_o); //Mean Flow Velocity in the annulus
Reo=oflmeanvel*dh/okinvisco; //Reynolds Number in the annulus
Nuo=0.023*pow(Reo,0.8)*pow(Pro,0.4); //Nusselt Number in the annulus --- Dittus-Boelter equation
for heating
ho=Nuo*ko/de; //Outer Surface Convection Heat Tranfer Coefficient
iflmeanvel=mh/(NFAi*dens_i); //Mean Flow Velocity in the pipe
Rei=iflmeanvel*inID/ikinvisco; //Reynolds Number in the pipe
Nui=0.023*pow(Rei,0.8)*pow(Pri,0.3);//Nusselt Number in the pipe --- Dittus-Boelter equation for
cooling
hi=Nui*ki/inID; //Inner Surface Convection Heat Tranfer Coefficient
cout<<"\n\n Outer Surface Convection Heat Transfer Coefficient : ";
cout<<"\n\n\t ho = "<<ho<<" W/m^2K";
cout<<"\n\n Inner Surface Convection Heat Transfer Coefficient : ";
cout<<"\n\n\t hi = "<<hi<<" W/m^2K";
// HEAT TRANSFER CALCULATIONS
Ufs=1.000/(Ro+(1.000/ho)); //Outer surface convection with fouling
m=sqrt((Ufs*2.000)/(fin_t*kw));
etaf=(tanh(m*fin_h))/(m*fin_h); //Fin eficiency
Af=2.000*noffins*length*fin_h; //Fin Surface Area
Ab=length*(pi*inOD-noffins*fin_t); //Base Area
Atot[i-1]=Af+Ab; //Net Outer Surface Area
etaf1=(Af*etaf+Ab)/Atot[i-1]; //Net fin efficiency of the outer surface
126
Ao=pi*inOD*length; //Unfinned pipe outer surface area
Rw=(inOD*log(inOD/inID))/(2.000*kw); // Wall thermal resistance
Uo[i-1]=1.000/((1.000/(Ufs*etaf1))+Rw*(Atot[i-1]/Ao)+(Ri+(1/hi))*(Atot[i-
1]/(pi*inID*length)));//Overall Heat Transfer Coefficient
NTU[i-1]=(Uo[i-1]*Atot[i-1])/mC_min; //Number of Transfer Units
Vol[i-1]=length*(pi*(inOD*inOD-inID*inID)/4+fin_h*fin_t*noffins); //Volume of the pipe material
required
R=mC_min/mC_max; // ratio of specific heats
eff[i-1]=(1.000-exp(-NTU[i-1]*(1.000-R)))/(1.000-R*exp(-NTU[i-1]*(1.000-R)));//heat exchanger
effectiveness
Heat[i-1]=eff[i-1]*(mC_min*(Thi-Tci)); // Net Heat Transfer
Tho[i-1]=Thi-(Heat[i-1]/(mh*cph)); // Hot Fluid Outlet Temp.
Tco[i-1]=Tci+(Heat[i-1]/(mc*cpc)); // Cold Fluid Outlet Temp.
// DISPLAY RESULTS
cout<<"\n\n\n\n NTU = "<<NTU[i-1]<<"\n\n Net Heat Transfer Area, Atot = "<<Atot[i-1]<<"
m^2"<<"\n\n Heat Exchanger Effectiveness = "<<eff[i-1]<<"\n\n Net Heat Transfer = "<<Heat[i-1]<<"
W";
cout<<"\n\n Hot Inlet Temp = "<<Thi<<"\t Hot Outlet Temp = "<<Tho[i-1]<<"\n\n Cold Inlet Temp =
"<<Tci<<"\t Cold Outlet Temp = "<<Tco[i-1]<<"\n\n\n Overall Heat Transfer Coefficient, Uo =
"<<Uo[i-1];
getch();
fin_h=fin_h+.001; // Fin Height Incrementation
} // END OF LOOP
clrscr();
// TABLE OF VALUES OBTAINED AT VARIOUS FIN HEIGHTS
gotoxy(1,1);cout<<"No";gotoxy(5,1);cout<<"FinHt";gotoxy(12,1);cout<<"Atot";gotoxy(21,1);cout<<"U
o";
gotoxy(32,1);cout<<"Eff..";gotoxy(42,1);cout<<"Heat";gotoxy(60,1);cout<<"Tho";gotoxy(70,1);cout<<"
Tco";
gotoxy(1,2);cout<<"********************************************************************
***********";
int yp;
for(int k=1;k<i;k++)
{
127
if (!(k%40))
{
getch();
clrscr();
gotoxy(1,1);cout<<"No";gotoxy(5,1);cout<<"FinHt";gotoxy(12,1);cout<<"Atot";gotoxy(21,1);cout<<"U
o";
gotoxy(32,1);cout<<"Eff..";gotoxy(42,1);cout<<"Heat";gotoxy(60,1);cout<<"Tho";gotoxy(70,1);cout<<"
Tco";
gotoxy(1,2);cout<<"********************************************************************
***********";
}
if(k>=40)
yp=k-37;
else
yp=k+2;
gotoxy(1,yp);cout<<k;gotoxy(5,yp);cout<<.001*(k-1);gotoxy(12,yp);cout<<Atot[k-
1];gotoxy(21,yp);cout<<Uo[k-1];
gotoxy(32,yp);cout<<eff[k-1];gotoxy(42,yp);cout<<Heat[k-1];gotoxy(60,yp);cout<<Tho[k-
1];gotoxy(70,yp);cout<<Tco[k-1];
}
getch();
// GRAPH PLOTTING
// Plot a graph of Fin Height in mm versus the Heat Exchanger Effectiveness
/* request auto detection */
int gdriver = DETECT, gmode, errorcode;
/* initialize graphics mode */
initgraph(&gdriver, &gmode, "..\\bgi");
/* read result of initialization */
errorcode = graphresult();
if (errorcode != grOk) /* an error occurred */
{
printf("Graphics error: %s\n", grapherrormsg(errorcode));
128
printf("Press any key to halt:");
getch();
exit(1); /* return with error code */
}
int ii,scale=500;
if(eff[i-2]*500>getmaxy()-30)
scale=300;
cleardevice();
setcolor(getmaxcolor()-1);
setfillstyle(SOLID_FILL, getmaxcolor()-4);
line(15, 15, 15, getmaxy()-15);
line(15,getmaxy()-15,getmaxx()-15,getmaxy()-15);
line(15,15,getmaxx()-15,15);
line(getmaxx()-15,15,getmaxx()-15,getmaxy()-15);
// X-AXIS PLOTTING
for (k=0;k<15;k++)
{
fillellipse(15+k*40, getmaxy()-15, 2,2);
}
// Y-AXIS PLOTTING
for (k=1;k<=10;k++)
{
fillellipse(15,getmaxy()-15-k*scale/10, 2,2);
}
settextstyle(0,0,1);
outtextxy(100,getmaxy()-10,"Fin Height in mm (5mm per Division) -->");
settextstyle(0,1,1);
outtextxy(10,50,"Effectiveness % (10% per Division) -->");
int x1=15,y1=getmaxy()-15,x2,y2;
getch();
// Plotting the Points and the conneting lines
129
for (ii=0;ii<i-1;ii++)
{
setcolor(5);
setfillstyle(SOLID_FILL, getmaxcolor()-5);
fillellipse(x1, y1, 3,3);
x2=15+(ii)*8;
y2=(getmaxy()-15)-int(eff[ii]*scale);
line(x1,y1,x2,y2);
getch();
x1=x2;
y1=y2;
}
setcolor(9);
line(x2,y2,15,y2);
getch();
double perc;
// To find the optimum fin height
k=i-2;
for(ii=0;ii<i-1;ii++)
{
perc=(eff[ii+1]*100.0-eff[ii]*100.0)/(eff[ii]*100.0)*100.0;
if(perc<=0.3) // Percentage increase in Eff. less than 0.3%
{
k=ii;
break;
}
}
int opt=k;
setcolor(7);
// PLOT THE OPTIMUM POINT
130
line(15+k*8,y2,15+k*8,getmaxy()-15);
getch();
setcolor(10);
line(15+k*8,(getmaxy()-15)-int(eff[k]*scale),15,(getmaxy()-15)-int(eff[k]*scale));
getch();
//Percentage increase in volume and Percentage increase in Heat are plotted against fin height
cleardevice();
setcolor(getmaxcolor()-1);
setfillstyle(SOLID_FILL, getmaxcolor()-4);
line(15, 15, 15, getmaxy()-15); //Left Vertical
line(15,getmaxy()-15,getmaxx()-15,getmaxy()-15); //Bottom Horizontal
line(15,15,getmaxx()-15,15); //Top Horizontal
line(getmaxx()-15,15,getmaxx()-15,getmaxy()-15); //Right Vertical
// X-AXIS DIVISION PLOTTING
for (k=0;k<15;k++)
{
fillellipse(15+k*50, getmaxy()-15, 2,2);
}
// Y-AXIS DIVISION PLOTTING
for (k=1;k<=10;k++)
{
fillellipse(15,getmaxy()-15-k*80, 2,2);
}
settextstyle(0,0,1);
outtextxy(100,getmaxy()-10,"Fin Height in mm (5mm per Division) -->");
settextstyle(0,1,1);
outtextxy(10,10,"Percentage Increase in Vol & Heat % (10% per Division) -->");
long double percVol,percHeat;
int ya1,yb1,ya2,yb2;
x1=15,ya1=getmaxy()-15,yb1=getmaxy()-15;
for (k=1;k<(i-1);k++)
131
{
getch();
percVol=(Vol[k]-Vol[k-1])/Vol[k-1]*100;
percHeat=(Heat[k]-Heat[k-1])/Heat[k-1]*100;
setcolor(5);
setfillstyle(SOLID_FILL, getmaxcolor()-5);
fillellipse(x1, ya1, 3,3);
setfillstyle(SOLID_FILL, getmaxcolor()-6);
fillellipse(x1, yb1, 3,3);
x2=15+k*10;
ya2=(getmaxy()-15)-int(percVol*8.00);
yb2=(getmaxy()-15)-int(percHeat*8.00);
line(x1,ya1,x2,ya2);
setcolor(6);
line(x1,yb1,x2,yb2);
x1=x2;
ya1=ya2;
yb1=yb2;
}
getch();
cleardevice();
getch();
/* clean up */
closegraph();
clrscr();
// RESULTS AND CONCLUSIONS
cout<<"\n RESULTS AND CONCLUSIONS";
cout<<"\n---------------------------";
cout<<"\n Graph 1. The effectiveness is found to be increasing as the fin height increases. But beyond
a point it becomes almost constant.";
cout<<"\n Graph 2. ";
cout<<"\n\n The optimum value of the fin height obtained from the graphs is : "<<opt<<" mm";
132
cout<<"\n\n Corresponding Net Heat Transfer Rate is "<<Heat[opt]<<" W";
cout<<"\n\n Heat Exchanger Effectiveness is "<<eff[opt]*100<<" %";
cout<<"\n\n Percentage increase in Heat Transfer Rate from unfinned pipe is "<<(Heat[opt]-
Heat[0])/Heat[opt]*100<<" %";
cout<<"\n\n Percentage increase in Volume when fins are added = "<<(Vol[opt]-
Vol[0])/Vol[opt]*100<<" %";
getch();
clrscr();
goto MENU;
}
133
REFERENCES
1. Wang, C.C., Lee, W.S., Sheu, W.J., 2001, A comparative study of compact
enhanced fin-and-tube heat exchangers, International Journal of Heat and Mass
transfer, vol. 44: p.3565-3573.
2. Du, Y.J., Wang, C.C., 2000, An experimental study of the air-side performance of
the superslit fin-and-tube heat exchangers, International Journal of Heat and Mass
Transfer, vol. 43: p.4475-4482.
3. Kim, N.H., Youn, B., Webb, R.L., 1999, Air side heat transfer and friction
correlations for plain fin-and-tube heat exchangers with staggered tube arrangements,
ASME Transaction, vol.121, august 1999.
4. Wang, C.C., Webb, R.L., Chi, K.Y., 2000, Data reduction for air-side performance
of fin-and-tube heat exchangers, Experimental Thermal and Fluid Science, vol. 21:
p.218-226.
5. Creed Taylor, December 2004 ,Measurement Of Finned-Tube Heat Exchanger
Performance. P 9-23.
6. Jeffrey Siegel and Van P. Carey, September 2001, Fouling of HVAC Fin and Tube
Heat Exchangers,p 1-4.
7. Sarit.K.Das, Process Heat Transfer, Alpha Science Intl Ltd, p 89-230
8. Adrian Bejan , Allan D. Kraus Heat Transfer Handbook, Wiley, New York, 2003.
9. Stephen Schneider, December 2000, Water Heat exchanger Optimisation for Space
Flight Applications. University of Colorado.
10. Wolverine Tube Heat Transfer Data Book, www.wlv.com/products/databook/.
11. Latif.M.Jiji, 2006, Heat Convection. Springer; 1 edition, P293-343.
12.Hans.Dieter Baehr, Karls Stephan, Heat and Mass Transfer, 2
nd
Edition, Berlin:
Springer 2006.
13. T.Kuppan, Heat Exchanger Design Handbook, Marcel Dekker 2003. P 1-158.
14. http://en.wikipedia.org/wiki/Heat_exchanger.
15. http://en.wikipedia.org/wiki/Heat_transfer_coefficient.
16. google.com/heat exchangers :pdf