CHAPTER THREE

CONVECTION

A large majority of practical applications of heat transfer in the chemical process

industries involve either heat transfer to a fluid or heat transfer from a fluid.
The heat flow mechanism in solids is by conduction; whereas the heat flow mechanism
in fluids is due to convection. Convection is the transfer of heat from one point to another
point within a fluid by mixing of hot and cold portions of the fluid.
Heat transfer by convection occurs as a result of the movement of the fluid on a
macroscopic scale in the form of circulating currents. The circulating currents may be set up
either by heat transfer process itself or some external agency. It is restricted to the heat flow
in fluids.
In the case of convective heat transfer, the physical mixing of the hot and cold portions of
a fluid is responsible for the flow of heat from one place to another within the fluid.
There are two types of convection.
Convection is classified as :
(i) free or natural convection and (ii) forced convection.
When the circulating currents arise from the heat transfer process itself, i.e., from the
density differences arising in turn due to temperature differences / gradients within the fluid
mass, the mode of heat transfer is called free or natural convection.
Examples of natural convection :
1. Heating of a vessel containing liquid by means of a gas flame situated underneath.
The liquid at the bottom of the vessel gets heated, expands and rises because its density
has become less than that of the remaining liquid. Cold liquid of higher density takes its
place and a circulating current is set up.
2. The flow of air across a heated radiator/heat of a room by means of a steam radiator.
When the circulating currents are produced by an external agency such as an agitator in a
reaction vessel, pump, fan or blower, the mode of heat transfer is called forced convection.
Here fluid motion is independent of density gradients.
Example of forced convection :
Heat flow to a fluid pumped through a heated pipe.
In general, higher rates of heat transfer are obtained in forced convection as compared to
natural convection owing to a greater magnitude of circulation in the forced circulation.
In the case of convective heat transfer taking place from a surface to a fluid, the
circulating currents die out in the immediate viscinity of the surface and a film of the fluid,
(3.1)
Unit Operations – II 3.2 Convection

free of turbulence, covers the surface. Heat transfer through this film takes by thermal
conduction. Since the thermal conductivity of most fluids is low, the main resistance to heat
transfer lies in the film. Therefore, an increase in the velocity of the fluid over the surface
results in improved heat transfer mainly because of reduction in the thickness of the film.
If the resistance to heat transfer is considered as lying within the film covering the
surface, the rate of heat transfer Q is given by
Q = kA ∆T/x
The effective thickness x is not generally known and therefore this equation is usually
rewritten in the form :
Q = hA ∆T
This is the basic equation for the rate of heat transfer by convection under steady state
conditions, where 'h' is called the film heat transfer coefficient or surface coefficient or
simply film coefficient. The value of 'h' depends upon the properties of the fluid within the
film region, hence it is called the film heat transfer coefficient. It depends upon the various
properties of the fluid, linear dimension of surface and fluid velocity (i.e. the nature of flow).
Numerically, heat transfer coefficient (h) is the quantity of heat transferred in a unit
time through a unit area at a temperature difference of one degree between the surface and
surrounding. h has the units of W/(m2·K) in the SI system. The term 1/h is called as the
thermal resistance.
The process of transfer of heat from a hot fluid to a cold fluid through metal wall is very
common in the chemical and process industry. The heat transferred may be latent heat of a
phase change, e.g., condensation, vaporisation etc.; or may be sensible heat. In all process
equipments, e.g., heater, cooler, condenser, reboiler etc. heat is transferred by conduction and
convection.
Individual and Overall heat transfer coefficients :
Consider that a hot fluid is flowing through a circular pipe and a cold fluid is flowing on
the outside of the pipe. The heat will flow from the hot fluid to the cold fluid through a series
of resistances. Generally, the velocity of the fluid may be considered to be zero at the solid
surface and it rapidly increases as we move away from the wall surface. It is found that even
in the turbulent flow, where convective heat flow occurs from a surface to a fluid, the thin
film of the fluid free of turbulence (viscous sublayer) exists at the wall surface. This thin film
of fluid covering the surface is of great importance in determining the rate of heat transfer as
all the heat reaching the bulk of the cold fluid must pass through the film of fluid by
conduction. The thermal conductivities of the fluids are very low so that the resistance
offered by the film to the heat flow is very large even though the film is thin. Beyond this
film, the turbulence present brings about rapid equalisation of temperature.
The temperature gradients for the situation under consideration are shown in Fig. 3.1.
The dotted lines Y1 Y2 and Z1 Z2 represent the boundaries of thin films (hot and cold fluid
films). The flow of fluid to the left of Y1 Y2 and right of Z1 Z2 is turbulent. The temperature
gradient from the bulk of the hot fluid to the metal wall is represented by Ta T' T2, where Ta is
the maximum temperature of the hot fluid, T' is the temperature at the boundary between
Unit Operations – II 3.3 Convection

turbulent and viscous regions and T2 is the temperature at the actual interface between fluid
and solid. Similarly, the temperature gradient in the cold fluid is represented by lines
T3 T" Tb. In heat transfer calculation, for convenience, average temperature of the fluid is
usually used rather than the maximum temperature or the temperature at the outer surface of
the film. The average temperature (T1) of the hot fluid is represented by the line marked NN
and similarly the average temperature (T4) of the cold fluid is represented by the line marked
MM.
Y1 Z1
Ta
N Cold fluid film
N
T1 T'

T2
Hot fluid T3

Hot fluid film Cold fluid

T" T4
M
M
Tb
X2
X1

Y2 Z2
Metal wall

Fig. 3.1 : Temperature gradient in forced convection

The temperature change from T1 to T2 is taking place in the hot fluid film of thickness x1.
The rate of heat transfer through this film by conduction is given by
k1 A1 (T1 – T2)
Q = x1 … (3.1)

The effective film thickness x1 depends upon the nature of flow and nature of the surface,
and is generally not known. Therefore Equation (3.1) is usually rewritten as
Q = hi Ai (T1 – T2) … (3.2)
where hi is known as the inside heat transfer coefficient or the surface coefficient or the
simply film coefficient.
As seen from Equation (3.2), the film coefficient is a measure of the rate of heat transfer
for unit temperature difference and unit surface of heat transfer and it indicates the rate or
speed of transfer of heat by a fluid having a variety of the physical properties under varying
degrees of agitation. In the SI system, it has the units of W/(m2·K).
Unit Operations – II 3.4 Convection

The overall resistance to heat flow from the hot fluid to the cold fluid is made up of three
resistances in series. They are :
1. Resistance offered by the film of hot fluid.
2. Resistance offered by the metal wall and
3. Resistance offered by the film of cold fluid.
The rate of heat transfer through the metal wall is given by :
kAw (T2 – T3)
Q = xw … (3.3)
where Aw – log mean area of pipe
xw – thickness of pipe wall
and k – thermal conductivity of material of pipe.
The rate of heat transfer through the cold fluid film is given by :
Q = ho Ao(T3 – T4) … (3.4)
where ho is the outside film coefficient or individual heat transfer coefficient.
Equation (3.2) can be rearranged as :
Q
T1 – T2 = h A … (3.5)
i i
Similarly, Equations (3.3) and (3.4) can be rearranged as
Q
T2 – T3 = (kA /x ) … (3.6)
w w
Q
and T3 – T4 = h A … (3.7)
o o
Adding Equations (3.5), (3.6) and (3.7), we get
(T1 – T2) + (T2 – T3) + (T3 – T4) = Q h A + (kA /x ) + h A 
1 1 1
… (3.8)
 i i w w o o

(T1 – T4) = Q h A + (kA /x ) + h A 

1 1 1
∴ … (3.9)
 i i w w o o
where T1 and T4 are the average temperatures of the hot and cold fluid respectively. Therefore
equations similar to Equation (3.1) in terms of overall heat transfer coefficients can be
written as
Q = Ui Ai (T1 – T4) … (3.10)
or Q = Uo Ao (T1 – T4) … (3.11)
where Ui and Uo are the overall heat transfer coefficients based on the inside area and
outside area, respectively.
Equations (3.10) and (3.11) state that the rate of heat transfer is a product of three
factors namely the overall heat transfer coefficient, the area of heating surface and the
temperature drop.
Equation (3.11) can be rearranged as :
Q
(T1 – T4) = U A … (3.12)
o o
Unit Operations – II 3.5 Convection

Comparing Equations (3.9) and (3.12), we get

1 1 1 1
Uo Ao = hi Ai + (kAw/xw) + ho Ao … (3.13)

1 1 Ao xw  Ao  1
Uo = hi  A  + k A  + h … (3.14)
 i  w o
where Ao – area of heat transfer based on the outside diameter, i.e., the outside
area of the tube
Ai – area of heat transfer based on the inside diameter, i.e., the inside area
of the tube.
We have Ai = π Di L (where L is length of pipe)
Ao = π Do L
Ao Do
∴ Ai = Di … (3.15)
Ao Do
Similarly, Aw = Dw … (3.16)
where Dw = logarithmic mean diameter
Dw = 2 · rm where rm – logarithmic mean radius.
Putting the values of area ratios in Equation (3.14), we get
1 1 Do xw · Do 1
Uo =  
hi  Di  + k Dw + ho … (3.17)
Similarly,
1 1 xw Di 1  Di 
Ui = hi + k · Dw + ho Do … (3.18)
For thin walled tubes, the inside and outside radii are not much different from each other,
and hence the overall heat transfer coefficient Uo or Ui may be replaced simply by 'U' and is
written in terms of hi, ho, etc. as :
1 1 xw 1
U = hi + k + ho … (3.19)
1 1 1 1
Or U = hi + (k/xw) + ho … (3.20)
1
Or U = 1/h + x /k + 1/h
i w o
When the metal wall resistance is very small in comparison with the resistances of fluid
films, then Equation (3.19) reduces to :
1 1 1
U = ho + hi … (3.21)

1 ho + hi
U = hi ho … (3.22)
Unit Operations – II 3.6 Convection

Fouling Factor :
When the heat transfer equipment is put into service, after sometime, scale, dirt and other
solids deposit on both sides of the pipe wall, providing two more resistances to the heat flow.
The added resistances must be taken into account in the calculation of the overall heat
transfer coefficient. The additional resistances reduce the original value of U and thus the
required amount of heat is no longer transferred by the original heat transfer surface. Hence,
heat transfer equipments are designed by taking into account the deposition of dirt and scale
by introducing a resistance Rd known as the fouling factor (it is a thermal resistance due to
scale).
Equation (3.19) then becomes
1 1 xw 1
U = hi + k + ho + Rd … (3.23)

The overall heat transfer coefficient calculated by taking into account Rd is known as the
'design or dirty overall heat transfer coefficient' and the one calculated without taking into
account the term Rd is known as the 'clean overall heat transfer coefficient' – [as given by
Equation (3.19)].
The dirty overall coefficient [calculated by Equation (3.23)] is always less than the clean
overall coefficient. Fouling factor Rd is composed of Rdi, the dirt factor for an inner fluid at
the inside surface of a pipe and Rdo, the dirt factor for an outer fluid at the outside surface of
the pipe; Rd = Rdi + Rdo. The dirt or fouling factor Rd has the units of (m2·K)/W.
Resistance form of overall coefficient :
The reciprocal of an overall heat transfer coefficient is an overall resistance and it may be
given by the following equation :
1 1 Do xw  Do  1
= D  + k D  + h … (3.24)
Uo hi  i  w o

The individual terms on R.H.S. of the above equation represent the individual resistances
of the two fluids and of the metal wall.
The overall temperature drop is proportional to 1/U. Similarly, the individual
temperature drops in the two fluids and metal wall are proportional to the respective
individual resistances. Therefore,
∆T ∆Ti ∆Tw ∆To
1/Uo = Do/Di hi = (xw/k) (Do/Dw) = 1/ho … (3.25)

where, ∆T is the overall temperature drop

∆Ti is the temperature drop through the inside fluid
∆Tw is the temperature drop through the metal wall.
∆To is the temperature drop through the outside fluid.
Unit Operations – II 3.7 Convection

Magnitude of film heat transfer coefficients :

Table 3.1 gives the range of values of the film heat transfer coefficients for various
processes of heat transfer.
Table 3.1 : Magnitudes of film heat transfer coefficients (hi or ho)
Type of Processes Range of values of h in
W/(m2·K)
No phase change :
Water (heating or cooling) 300 – 20,000
Gases (heating or cooling) 20 – 300
Air (heating or cooling) 1 – 50
Oils (heating or cooling) 50 – 1500
Organic solvents (heating or cooling) 350 – 3000
Condensing :
Condensing steam (film-type condensation) 6000 – 20,000
Condensing organic vapours 1000 – 2000
Ammonia 3000 – 6000
Condensing steam (drop-wise condensation) 30,000 – 100,000
Evaporation :
Water 2000 – 12000
Organic solvents 600 – 2000
Ammonia 1100 – 2300
Table 3.2 gives the range of values of the overall heat transfer coefficients for various
fluid systems in a shell and tube equipment.
Table 3.2 : Approximate range of values of overall heat transfer coefficients, U
Hot side (Hot fluid) Cold side Overall U in
(Cold fluid) W/(m2·K)
Heat exchangers (without phase change) :
Water Water 900 – 1700
Gases Water 20 – 300
Organic solvents Water 300 – 900
Water Brine 600 – 1200
Gases Brine 20 – 300
Heavy organics Heavy organics 50 – 300
Heaters :
Steam Water 1500 – 4000
Steam Gases 30 – 300
Dowtherm Gases 20 – 200
Steam Light oils 300 – 900
Steam Organic solvents 600 – 1200
… Contd.
Unit Operations – II 3.8 Convection

Evaporators :
Steam Water 2000 – 4000
Steam Organic solvents 600 – 1200
Steam Light oils 400 – 1000
Water Refrigerants 400 – 900
Condensers :
Steam Water 2000 – 4000
Saturated organic solvents Water 600 – 1200
Low boiling hydrocarbons Water 400 – 1200
Organic solvents with high noncondensable Water 100 – 500
Classification of individual heat transfer coefficients :
The problem of predicting the rate of heat transfer from one fluid to another through a
metal wall reduces essentially to the problem of predicting the values of the film coefficients
of the fluids involved in the overall process. In practice, we come across a variety of cases
and each one must be considered separately. The cases that come across in practice may be
classified as
(a) Heat transfer to or from fluids inside tubes, without phase change.
(b) Heat transfer to or from fluids outside tubes, without phase change.
(c) Heat transfer from condensing vapours.
(d) Heat transfer to boiling liquids.
Sensible heat :
It is the heat that must be transferred either to raise or lower the temperature of a
substance or mixture of substances. It is the heat added or removed from a system to increase
or decrease the temperature of the system without changing its phase. The sensible heat
transfer is given by
⋅
Q = m Cp ∆T or m Cp ∆T
Latent heat :
When a matter undergoes a phase change, the enthalpy change associated with unit
amount of matter at constant temperature and pressure is known as the latent heat of phase
⋅ ⋅
change. The latent heat transfer is given by Q = mλ or mλ where m is the mass and m is the
mass flow rate.
Heat transfer to fluids without phase change : In most of the heat exchange
applications, heat is transferred between fluid streams without a phase change in the fluids.
Examples of this type of heat transfer are :
Heat exchange between hot and cooler petroleum streams.
Heat transfer from a stream of hot gas to cooling water.
Cooling of a hot liquid stream by cooling water/air.
In such cases the two streams are separated by a metal wall that constitutes the heat
transfer surface. The surface may consist of tubes, or other channels of constant cross section
or of flat plates.
A fluid being heated or cooled may be flowing in laminar flow, in turbulent flow or in
the transition range between laminar and turbulent.
Unit Operations – II 3.9 Convection

An equation for predicting the film coefficient or the surface coefficient in any particular
case must include all the properties of the fluid and conditions of its flow that affect the
problem. In a particular case the factors that might be considered are the diameter of pipe,
the velocity of flowing fluid, density, viscosity, thermal conductivity, specific heat of the
fluid, etc. The dimensional analysis is one of the most useful methods to assemble these
variables/factors into an equation. This method results in arranging the variables into various
dimensionless groups.
It is found that an equation for the film coefficient for heat transfer to or from a flowing
fluid without phase change will probably be of the form :
NNu = f NRe‚ NPr‚ NGr ‚ D
L
 
Application of Dimensional Analysis to heat transfer by convection :
Dimensional analysis is a method of correlating a number of variables into a single
equation that clearly states an effect. When the value of a given particular physical quantity is
influenced by a number of variables, then it is impossible to determine their individual
effects by the experimental methods. In such cases, a problem can be more easily handled by
using the method of dimensional analysis wherein the variables are arranged in
dimensionless groups which are significantly less than the number of variables. Dimensional
analysis finds application in many areas of chemical engineering such as fluid flow, heat
transfer, etc.
The rate of heat transfer per unit area depends on the viscosity µ, density ρ, specific heat
capacity Cp, thermal conductivity of the fluid k, a linear dimension of the surface l, the
velocity of the flowing fluid u, the temperature difference, ∆T, and the product of the thermal
expansion coefficient, β and the acceleration due to gravity, g. The relationship between the
heat flux and the above cited parameters may be given as
Q a lb ρc Cd ∆Te (β·g)f ki, µm]
A = α [u p … (3.26)
The dimensions of the variables in terms of length L, mass M, time θ, temperature T, and
heat H are given as follows :
Q/A rate of heat transfer per unit area H L–2 θ–1
u velocity of the flowing fluid L θ–1
l linear dimension of the surface L
µ viscosity of the flowing fluid M L–1 θ–1
ρ density of the flowing fluid M L–3
k thermal conductivity of the flowing fluid H L–1 θ–1 T–1
Cp specific heat of the flowing fluid H M–1 T–1
∆T temperature difference T
(β · g) coefficient of thermal expansion times T–1 L θ–2
the acceleration due to gravity.
Unit Operations – II 3.10 Convection

Substituting the dimensions of each term in Equation (3.26) gives

(HL–2 θ–1) = α (Lθ–1)a (L)b (ML–3)c (HM–1 T–1)d (T)e (T–1 Lθ–2)f ×
× (HL–1 θ–1 T–1)i × (ML–1 θ–1)m … (3.27)
Equating the indices of the dimensions L, θ, H, etc. on LHS with those on RHS of
Equation (3.27), we get
in length – 2 = a + b – 3c + f – i – m
in mass 0 = c–d+m
in time – 1 = – a – 2f – i – m
in temperature 0 = –d+e–f–i
in heat 1 = d+i
There are five equations (in length, mass, time, temperature and heat) and nine variables
so that the above equation (i.e., Equation 3.27) will be rearranged in four dimensionless
groups in terms of any three indices. Let us choose a, f and d be the three indices.
Let us obtain the other constants in terms of a, f and d.
We have 1 = d+i
∴ i = 1–d
We have – 1 = – a – 2f – i – m
∴ – 1 = – a – 2f – (1 – d) – m
∴ m = d – a – 2f
We have 0 = –d+e–f–i
0 = – d + e – f – (1 – d)
∴ e = 1+f
0 = c–d+m
0 = c – d + d – a – 2f
∴ c = a + 2f
– 2 = a + b – 3c + f – i – m
– 2 = a + b – 3 (a + 2f) + f – (1 – d) – (d – a – 2f)
∴ b = – 1 + a + 3f
Substituting for b, c, e, i and m, Equation (3.27) becomes
Q a –1 + a1 + 3f (ρ) a+2f (C )d (∆T)1+f × (βg)f (k)1–d (µ)d – a – 2f]
A = α [(u) (l) p … (3.28)

Collecting the terms, we get

Q k ∆T l uρa Cp µd βg ∆T l3 ρ2f
= α  l       … (3.29)
A   µ   k   µ2 
Unit Operations – II 3.11 Convection

Ql l · uρa Cp · µd βg · ∆T l3 · ρ2f

= α       … (3.30)
A ∆Tk  µ   k   µ2 
Since Q = hA ∆T, Q/A ∆T = h. Therefore,
hl l uρa Cp · µd βg · ∆T · l3 · ρ2f
= α       … (3.31)
k  µ   k   µ2 
If the linear dimension of the surface is D (for circular pipes), then Equation (3.31)
becomes
hD Duρa Cp µd β·g ∆T D3 ρ2f
= α       … (3.32)
k  µ   k   µ2 
The groups on both the sides of Equation (3.32) are dimensionless
hl hD
k = k (for pipe) = Nusselt number, NNu
luρ Duρ
= (for pipe)
µ µ
= Reynolds number, NRe
Cp µ
k = Prandtl number, NPr
βg ·∆Tl3 ρ2 βg · ∆T D3 ρ2
and = (for pipe)
µ2 µ2
= Grashof number, NGr
Physical significance of the groups :
Inertia forces
Reynolds number = Viscous forces
Buoyancy forces × inertia forces
Grashof number = (viscous forces)2
molecular diffusivity of momentum
Prandtl number = molecular diffusivity of heat
Wall heat transfer rate
Nusselt number = Heat transfer by conduction

The above relationship given by Equation (3.32) can be written as

NNu = α [NRe]a [NPr]d [NGr]f … (3.33)
For natural convection where there is a buoyancy effect, NGr influences the heat transfer
characteristics more than NRe. Therefore, NNu is a function of NGr and NPr.
For natural convection :
NNu = f (NPr, NGr) … (3.34)
Unit Operations – II 3.12 Convection

For forced convection, Reynold's number influences the heat transfer characteristics and
the Grashof number may be omitted. Therefore, for forced convection, we have
NNu = f (NRe, NPr) … (3.35 a)
I. Application of dimensional analysis for a forced convection equation :
To derive a relationship for heat transfer coefficient, h for forced convection heat transfer
on the assumption that the coefficient h is a function of the following variables :
l – linear dimension of the surface
Cp – specific heat of the fluid
ρ – density of the fluid
µ – viscosity of the fluid
k – thermal conductivity of the fluid
and u – velocity of the fluid.
Solution : It is given that h is a function of L, u, ρ, µ, Cp and k. Therefore, we can
write
h = f (ρ, u, L, µ, Cp, k)
e
h = A (ρa ub Lc µd Cp kf) … (3.35 b)
where A, a, b, c, d, e, and f are constants.
Express the energy terms mechanically by dimensions of the variables :
The dimensions of each variable in terms of M, L, T, θ and F are :
Parameter Units Fundamental dimensions
h N.m/(m ·s·K) = W/(m ·K)
2 2
FL/L2θT = FL–1 θ–1 T–1
k N·m/(m·s·K) = W/(m·K) FL/LθT = Fθ–1 T–1
Cp N·m/(kg·K) = J/(kg·K) FL/MT = FLM–1 T–1
µ kg/(m·s) M/Lθ = ML–1 θ–1
u m/s L/θ = Lθ–1
l m L = L
ρ kg/m 3
M/L3 = ML–3
Substituting the dimensions of all parameters in the above equation, we get
FL–1 θ–1 T–1 = A (ML–3)a (Lθ–1)b (L)c (ML–1 θ–1)d (FLM–1 T–1)e (Fθ–1 T–1)f … (1)
Equating the indices/powers of M, L, θ, T and F, on LHS with those on RHS of the above
equation gives
Force, F : 1 = e + f … (i)
Mass, M : 0 = a + d – e … (ii)
Length, L : – 1 = – 3a + b + c – d + e … (iii)
Temperature, T : – 1 = – e – f … (iv)
Time, θ : – 1 = – b – d – f … (v)
Let us obtain the constants c, d, etc. in terms of a and e.
Unit Operations – II 3.13 Convection

From Equation (i),

f = 1–e
From Equation (ii),
d = e–a
From Equation (v),
–1 = –b–d–f
b = 1 – d – f = 1 – (e – a) – f
b = 1–e+a–f
b = a + (1 – e) – f = a + f – f
∴ b = a
From Equation (iii),
– 1 = – 3a + b + c – d + e
c = 3a – b + d – e – 1
= 3a – a + (e – a) – e – 1
c = 2a – a – 1 = a – 1
Substituting back, Equation (1) becomes
h = A (ρ)a (u)a (L)a–1 (µ)e–a (Cp)e (k)1–e
Collecting the terms, we get

Luρa Cp µe

h = A  (k/L)
 µ   k 
hL Luρa Cp µe
∴ = A   … (3.35 c)
k  µ   k 
In the case of pipe, the linear dimension is D (inside diameter) and the above equation
becomes
hD Duρa Cp µe
= A  
k  µ   k 
We can obtain the same expression if we express energy terms thermally by the
dimensions of the variables.
Equation (3.35 c) is the required relationship; where A, a and e must be evaluated from a
minimum of three sets of experimental data.
This equation contains three dimensionless groups/numbers :
(i) Nusselt number, NNu = hL/k
Unit Operations – II 3.14 Convection

(ii) Reynolds number, NRe = Luρ/µ

(iii) Prandtl number, NPr = Cp µ/k.
Let us obtain the same relationship by expressing energy terms thermally.
Expressing the energy terms thermally by dimensions of the variables :
Parameter Units Fundamental dimensions
h W/(m ·K) = J/(s·m ·K)
2 2
H/θL2T = Hθ–1 L–2 T–1
k W/(m·K) = J/(s·m·K) H/θLT = H θ–1 L–1 T–1
Cp J/(kg·K) H/MT = HM–1 T–1
µ kg/(m·s) M/Lθ = ML–1 θ–1
u m/s L/θ = Lθ–1
l m L = L
ρ kg/m 3
M/L3 = ML–3
Substituting the dimensions of all the parameters, we get
Hθ–1 L–2 T–1 = A (ML–3)a (Lθ–1)b (L)c (ML–1 θ–1)d (HM–1 T–1)e (Hθ–1 L–1 T–1)f
Equating the indices/powers of M, L, θ, etc. on LHS and RHS of the above equation
gives
H : 1 =e+f … (i)
M : 0 =a+d–e … (ii)
L : – 2 = – 3a + b + c – d – f … (iii)
θ : –1 = –b–d–f … (iv)
T : –1 =– e–f … (v)
Let us obtain indices f, d, etc. in terms of a and e.
From equation (i), f = 1–e
From equation (ii), d = e–a
From equation (iv), b = 1 – d – f = 1 – e + a – 1 + e = a
From equation (iii), c = a–1
Substituting back values of the indices b, c, d and f in Equation (3.35 b),
h = A (ρ)a (u)a (L)a–1 (µ)e–a (Cp)e (k)1–e
Collecting the terms, we get
Luρa
h = A  (Cpµ/k)e (k/L)
 µ 
hL Luρa
∴ = A   (Cpµ/k)e … (3.35 d)
k  µ 
II. Application of dimensional analysis for a natural convection equation :
To derive a relationship for the heat transfer coefficient h for natural convection between
a surface and a fluid assuming that the coefficient h is a function of the following variables :
l – a characteristic linear dimension of the surface,
ρ – density of the fluid,
µ – viscosity of the fluid,
k – thermal conductivity of the fluid,
Unit Operations – II 3.15 Convection

Cp – specific heat of the fluid,

β·g – the product of the acceleration due to gravity and coefficient of
cubical expansion of the fluid,
∆T – the temperature difference between the fluid and the surface.
Solution : It is given that the coefficient h is a function of l, k, ρ, Cp, µ, βg and ∆T.
Therefore, we can write :
h = f [l, ρ, µ, k, Cp, β·g, ∆T]
∴ h = α [la ρb µc kd Cpe (βg)f (∆T)g]
The dimensions of each variable in terms of M, L, T, F and θ are :
Parameter Units Fundamental dimensions
h N.m/(m ·s·K) = W/(m ·K)
2 2 FL/L2θT = FL–1 θ–1 T–1
k N·m/(m·s·K) = W/(m·K) FL/LθT = Fθ–1 T–1
Cp N·m/(kg·K) = J/(kg·K) FL/MT = FLM–1 T–1
µ kg/(m·s) M/Lθ = ML–1 θ–1
l m L = L
ρ kg/m 3 M/L3 = ML–3
βg m/(s2·K) L/θ2T = L θ–2 T–1
∆T K T = T
Substituting the dimensions of each variable in the above equation, we get :
FL–1 θ–1 T–1 = α (L)a (ML–3)b (ML–1 θ–1)c (Fθ–1 T–1)d (FLM–1 T–1)e (Lθ–2 T–1)f × (T)g
Equating the indices/powers on the LHS and RHS of the above equation in force, mass,
etc., we get
Force, F : 1 = d+e … (i)
Mass, M : 0 = b+c–e … (ii)
Length, L : – 1 = a – 3b – c + e + f … (iii)
Temperature, T : –1 =–d–e–f+g … (iv)
Time, θ : – 1 = – c – d – 2f … (v)
Let us obtain the constants a, b, c and d in terms of e and f. From Equation (i),
d = 1–e
From Equation (v),
c = 1 – d – 2f, but d = 1 – e
∴ c = 1 – (1 – e) – 2f = e – 2f
From Equation (ii), we have
b = – c + e but c = e – 2f
∴ b = – (e – 2f) + e = 2f
Unit Operations – II 3.16 Convection

From Equation (iii), we have

a = – 1 + 3b + c – e – f
a = – 1 + 3 (2f) + (e – 2f) – e – f = – 1 + 3f
From Equation (iv), we have
g = – 1 + d + e + f = – 1 + 1 + f = f [as d + e = 1]
Substituting back the values of constants a, b, c and d, the equation for h becomes
h = α (l)–1 + 3f (ρ)2f (µ)e–2f (k)1–e (Cp)e (βg)f (∆T)f
Collecting the terms, we get
Cpµe l3 ρ2 βg ∆Tf
h = α k    (k/L)
   µ2 
hL Cp µ l3 ρ2 βg ∆Tf
e
 
 k   
∴ = α … (3.35 e)
k µ2 
Equation (3.35 e) is the desired relationship for natural convection.
Cp µ
Here k is the Prandtl number (NPr)
hL/k is the Nusselt number (NNu)
l3 ρ2 β·g ∆T
and is the Grashof number (Nar)
µ2
Empirical equations for calculation of heat transfer coefficients in laminar, turbulent
and transition region in forced convection :
[A] Film coefficients-in pipes – Laminar flow :
Heat transfer to laminar flow may be occurred whenever fluids to be heated or cooled are
viscous or flow rates are low. In polymer, food or pharmaceutical industry, we may come
across a heat exchanger where the flow is in the laminar region (NRe < 2100).
For heating or cooling of viscous fluids, an empirical correlation is
NNu = 1.86 [(NRe) (NPr) (D/L)]1/3 [µ/µw]0.14 … (3.36)
Equation (3.36) is the Sider-Tate equation for the calculation of the heat transfer
coefficient for laminar flow of fluids in horizontal tubes or pipes. Equation (3.36) is valid for
NRe < 2100 and (NRe NPr , D/L) > 100
where D is the diameter of pipe and L is the length of pipe.
The term (µ/µw)0.14 is called as the Sider-Tate correction. The term (µ/µw) is the ratio of
the viscosity of the fluid at the bulk temperature and the viscosity of the fluid at the wall
temperature.
hD Duρ Cp µ D1/3  µ 0.14
= 1.86       … (3.37)
k  µ   k   L  µw
where h is the inside heat transfer coefficient, D is the inside diameter of pipe.
Unit Operations – II 3.17 Convection

In the above equation all the physical properties of the fluid are taken at the mean bulk
temperature of the flowing fluid, i.e., at the arithmetic mean of inlet and outlet temperatures
of the fluid flowing through pipe (Ti + To)/2 and µw is evaluated at the average wall
temperature.
Graetz number is defined by the following equation
.
m Cp
NGz = kL … (3.38)
.
where m is the mass flow rate
. π
m = ρuA = 4 ρuD2 … (3.39)
π ρu Cp D2
NGz = 4 kL
π Duρ Cp µ D
= 4 k ·L
µ
π D
= 4 NRe · NPr · L … (3.40)
D
By substituting for NRe · NPr · L from Equation (3.40), Equation (3.36) becomes :
1/3
 µ 0.14
NNu = 1.86  (NGz) ×  
4
π  µw
NNu = 2 [NGz] [µ/µw]
1/3 0.14 … (3.41)
m. C 1/3
hD  p
k = 2.0  k L  [µ/µw] … (3.42)
0.14

Equation (3.42) is the Sider-Tate equation for laminar flow/viscous flow in terms of the
Graetz number.
The Peclet number is defined as the product of the Reynolds number and Prandtl number.
Duρ Cp µ Du
NPe = NRe · NPr = · k = … (3.43)
µ α
α = k/ρCp, α is the thermal diffusivity and has units of m2/s.
The choice among these groups is arbitrary.
For thin fluids in laminar flow :
For values of (NRe · NPr D/L) > 12, the following empirical equation is applicable :
NNu = 1.62 NRe · NPr · L 
D 1/3
… (3.44)
 
Equation (3.44) can be written as
m. C 1/3
NNu = 1.75  k L 
p
… (3.45)
 
For viscous liquids in laminar flow, Equation (3.42) is applicable in which variation of
viscosity with temperature is introduced.
Unit Operations – II 3.18 Convection

[B] Film coefficients - in pipes – Turbulent flow :

The most important situation in heat transfer is the flow of heat in a stream of fluid in
turbulent flow in pipes or tubes. Turbulence is encountered at NRe > 2100. As the rate of heat
transfer is greater in turbulent flow than in laminar flow, most of the heat transfer equipments
are operated in the turbulent range.
The Dittus-Boelter equation for predicting the heat transfer coefficient for turbulent flow
in tubes or pipes is
Duρ Cp µ
0.8 a
hD
= 0.023   … (3.46)
k  µ   k 
where 'a' has a value of 0.4 for heating and 0.3 for cooling. It is commonly used for water like
materials.
Here also the fluid properties are evaluated at the arithmetic mean bulk temperature,
i.e., at (Ti + To)/2 of the flowing fluid. Ti and To are the inlet and outlet temperatures.
For heating :
NNu = 0.023 [NRe]0.8 [NPr]0.4 … (3.47)
For cooling :
NNu = 0.023 [NRe]0.8 [NPr]0.3 … (3.48)
Equations (3.47) and (3.48) are valid for Reynolds number greater than 10000 and
Prandtl number between 0.7 and 160.
For turbulent flow in tubes/pipes, the Sieder-Tate equation that takes into account the
variation of the viscosity of the fluid near the wall with thermal gradients is
hD Duρ
0.8
Cp µ1/3
  (µ/µw)0.14
k = 0.023  µ   k 
… (3.49)

Equation (3.49) is the Sieder-Tate equation.

Equation (3.49) is valid for NRe > 10,000 and 0.7 < NPr < 700 and L/D > 60. The
term (µ/µw)0.14 is called as the Sieder-Tate correction factor.
Here all the fluid properties are taken at the mean bulk temperature, i.e., at (Ti + To)/2,
where Ti is inlet fluid temperature and To is the fluid outlet temperature. µw is the viscosity of
fluid at the wall temperature.
An alternate equation that in many ways is more convenient is the one proposed by
Colburn.
In the Colburn equation, the Stanton number (Nst) is used instead of the Nusselt number
(NNu) :
h
Nst = … (3.50)
Cp ρu
Unit Operations – II 3.19 Convection

It should be noted that :

h hD µ k
= k · · … (3.51)
Cp ρu Duρ Cp µ
–1 –1
Nst = NNu · NRe · NPr
NNu = Nst · NRe · NPr … (3.52)
E.M. Sider and C.E. Tate proposed the following empirical equation for predicting the
heat transfer coefficient for turbulent flow.
NNu = 0.023 (NRe)0.8 (NPr)1/3 (µ/µw)0.14 … (3.53)
Substituting the value of NNu from Equation (3.52) into Equation (3.53), we get
Nst · NRe · NPr = 0.023 (NRe)0.8 (NPr)1/3 (µ/µw)0.14 … (3.54)
2/3
Nst · NPr · (µw/µ)0.14 = 0.023 (NRe)–0.2 … (3.55)
h Cp µ2/3 µw0.14 Duρ– 0.2
    = 0.023   … (3.56)
Cp ρu  k  µ  µ 
Equation (3.56) is the Colburn equation.
Turbulent flow of gases :
For heating or cooling of gases in turbulent flow, for which the Prandtl group usually has
a value of about 0.74, an empirical correlation is
NNu = 0.02 (NRe)0.8 … (3.57)
For turbulent flow in pipes, the following equations hold good in SI units :
1. For air at 1 atm. pressure
3.5 u0.8
h = D0.2 … (3.58)

2. For water in tubes

u0.8
h = 1063 (1 + 0.00293 T) D0.2 W/(m2·K) … (3.59)

where T is average bulk temperature in K.

3. For organic liquids :
u0.8
h = 423 D0.2 … (3.60)

Mass velocity is given by the relation :

.
G = m /A
= ρuA/A
= ρu, kg/(m2 · s) … (3.61)
Unit Operations – II 3.20 Convection

Therefore, in all equations involving the Reynolds group, we can write NRe also as
DG
NRe = … (3.62)
µ
In all the equations cited above use the following SI units :
D in m, u in m/s, L in m, G in kg/(m2·s), ρ in kg/m3, Cp in J/(kg·K), µ in Pa·s =
kg/(m·s) = (N·s)/m2 and k in W/(m·K) ≡ J/(m·s·K) so that the unit of h will be W/(m2·K).
Flow in non-circular cross-sections :
To use the Sider-Tate or Colburn equation for cross sections other than circular, it is only
necessary to replace the diameter D in both Reynolds and Nusselt numbers by the equivalent
diameter, De, defined mathematically as
De = 4 × rH … (3.63)
4 × cross-sectional area
De = wetted perimeter
For a circular pipe, De is
4 (π/4 D2)
De = = D
πD
For an annulus of outer diameter Do and inner diameter Di, De is given by

π 2 2 
4 4 [Do – Di ] 2 2
  Do – Di
De = = Di … (3.64)
π Di
where, Do = inner diameter of outer pipe (outside diameter of annulus)
Di = outer diameter of inner pipe (inside diameter of annulus)
Annulus
Inner pipe

Outer pipe

Fig. 3.2 : Double pipe heat exchanger

For a duct of rectangular cross-section x by y, De is given by
4 xy 2xy
De = 2 (x + y) = (x + y) … (3.65)
Unit Operations – II 3.21 Convection

The Colburn analogy; Colburn j factor :

For NRe between 5000 – 200000, the friction factor (f) for a smooth pipe is given by the
following empirical equation
Duρ– 0.2
f = 0.046   … (3.66)
 µ 
Comparison of Equation (3.66) with Equation (3.55) for heat transfer in turbulent flow
inside a pipe/tube gives
h Cp µ2/3 µw0.14 f
    ≡ j = … (3.67)
Cp ρu  k   µ  H 2
Equation (3.67) is the Colburn analogy between heat transfer and fluid friction. The
factor jH is called the Colburn j factor. Equation (3.55) can be written in the j-factor form as
given below :
jH = 0.023 (NRe)–0.2 … (3.68)
For laminar flow region, we have
m. C 1/3 0.14
µ
= 2.0  k L 
p
NNu
  µ  … (3.69)
 w
Using Equation (3.40), Equation (3.69) becomes
π D 
1/3
NNu = 2.0 4 L NRe · NPr [µ/µw]0.14 … (3.70)
 
Multiplying Equation (3.70) by (1/NRe) (1/NPr), we get
h Cp µ2/3 µw0.14 D
1/3
Duρ
– 2/3
    = j = 1.86   … (3.71)
Cp ρu  k   µ  H
L  µ 
[C] Film coefficients-in pipes - in Transition Region :
Equation (3.37) is valid for Reynolds numbers less than 2100 and turbulent flow
equation for predicting the heat transfer coefficient, i.e., Equation (3.55) is valid for
Reynolds numbers more than 10,000. The range of Reynolds numbers between 2100 and
10000 is called the transition region. For this, no simple equation is applicable and therefore
to obtain h, a graphical method is used. In the graphical method the Colburn j factor is
plotted against NRe at several values of L/D - with L/D as a parameter.
Equation (3.71) shows that for each value of L/D, a logarithmic plot of L.H.S.,
i.e., jH v/s NRe gives a straight line with a slope equal to (– 2/3). The lines terminate at
NRe = 2100.
Equation (3.66) when plotted on the same coordinates gives a straight line with a slope
equal to (– 0.20) for NRe > 10,000.
The transition region is represented by curved lines for NRe between 2100 and 10000.
Unit Operations – II 3.22 Convection

Fig. 3.3 covers NRe between 10000 and 30000, so Fig. 3.3 is referred in this range.
(For NRe < 1000 and NRe > 30000), Equations (3.37) and (3.55), respectively can be
used.
For the transition region, i.e., for 2100 < NRe < 10000, the following empirical equation
can be used to calculate the film heat transfer coefficient.

NNu = 0.116 [(NRe)2/3 – 125] [NPr]1/3 1 +  L   [µ/µw]0.14 … (3.72)

D 2/3
   

0.01
0.14
w
(mm )
2/3

L/
D
= (C G) ( k )
C m

L/ =
p

D 50
=
10
0
i
h
p
JH

0.001
3 4
10 NRe = DG/m 10

Fig. 3.3 : Tube side heat transfer curves transition range

First of all NRe is evaluated from the data provided and then from Fig. 3.3, jH is obtained
for the corresponding NRe and L/D and in turn, it gives the value of the tube side heat transfer
coefficient.
Wilson Plot :
This plot is used to determine the film heat transfer coefficients.
Consider a shell and tube heat exchanger wherein steam is condensing on the shell side
and a cold fluid is flowing through the tubes in the turbulent flow region. The overall heat
transfer coefficient is determined by the direct measurements of heat transfer rate, overall
temperature difference and area at various cold fluid velocities. In this case, the condensing
steam side coefficient (ho) remains almost constant and the resistance offered by the metal
wall is also constant. Assuming clean tubes, Equation (3.19) reduces to
1 1
U = hi + C … (3.73)

where 'C' is a constant.

Unit Operations – II 3.23 Convection

From Equation (3.54) for turbulent flow, we can write

NNu ∝ (NRe)0.8 … (3.74)
∴ hi ∝ (u) 0.8 … (3.75)
hi = a (u)0.8 … (3.76)
With this, Equation (3.73) becomes
1 1
U = a (u)0.8 + C … (3.77)
where 'u' is the linear velocity of the cold fluid. A plot of 1/U v/s 1/(u)0.8 results in a straight
xw 1
line with a slope equal to 1/a and an intercept equal to k + h . The value of ho is obtained
o
from the intercept and 'a' represents the value of film coefficient hi for a unit velocity of the
cold fluid. Such a plot shown in Fig. 3.4 is known as the Wilson plot.

1
U

1
(u)0.8

Fig. 3.4 : Wilson plot

Flow arrangements in heat exchangers :
A heat exchange is a device used for exchanging heat between two fluids that are at
different temperatures.
There are three basic flow arrangements.
1. Parallel flow/Co-current flow 2. Counter current flow and 3. Cross flow
Consider a double pipe heat exchanger wherein a hot fluid is flowing through the inside
pipe and a cold fluid is flowing through the annular space for the explanation of parallel and
counter current flow.
When both the fluids flow in the same direction from one end of a heat exchanger to
other end through the heat exchanger, then the flow is called a co-current or parallel flow.
Such a flow is shown in Fig. 3.5 (a). The temperature - length curve for parallel flow
arrangement is shown in Fig. 3.5 (b).
When the fluids are flowing through a heat exchanger in opposite directions with respect
to each other (i.e., one fluid enters at one end of the heat exchanger and the other fluid enters
at the opposite end of the heat exchanger), then the flow is called a counter current flow.
Such a flow arrangement is shown in Fig. 3.6 (a). The temperature - length curve for counter
flow arrangement is shown in Fig. 3.6 (b).
When the fluids are directed at right angles to each other through a heat exchanger, then
the flow arrangement is called a cross-flow.
Unit Operations – II 3.24 Convection
Cold fluid
out (Tc2)

Hot fluid Hot fluid

in (Th1) out (Th2)

Cold fluid
in (Tc2)
Fig. 3.5 (a) : Co-current/Parallel flow in heat exchanger
Th Hot fluid flow
1
Temperature

Th
2
Tc
2
DT1 = Th – Tc
1 1
Tc Cold fluid flow DT2 = Th – Tc
1 2 2

0 Length L
(From cold fluid inlet)
Fig. 3.5 (b) : Temperature - length curve for parallel flow
Cold fluid
in (Tc1)

Hot fluid Hot fluid

in (Th1) out (Th2)

Cold fluid
out (Tc2)
Fig. 3.6 (a) : Counter-current flow in heat exchanger
Hot fluid flow
Temperature

Th
1
Th Tc
2 2

DT1 = Th – Tc
2 1
Tc Cold fluid flow DT2 = Th – Tc
1 1 2

0 Length L
(From cold fluid inlet)

Fig. 3.6 (b) : Temperature - length curve for counter-current flow

Unit Operations – II 3.25 Convection
Hot fluid in

Cold fluid in Cold fluid out

Hot fluid out

Fig. 3.7 : Cross flow heat exchanger
Heat exchangers according to the flow arrangement :
Parallel flow heat exchanger : It is the one in which two fluid streams enter at one end,
flow through it in the same direction and leaves at the other end. For example, double pipe
heat exchangers, shell and tube heat exchangers can be operated in a parallel flow fashion.
Counter current flow heat exchanger : It is the one in which two fluid streams flow in
opposite directions.
For example, double pipe heat exchangers, shell and tube heat exchangers can be
operated in this way.
Cross flow heat exchanger : It is the one in which one fluid moves through the
exchanger at right angles to the flow path of the other fluid. Cross flow heat exchangers are
used for air or gas heating. For example, plate-fin heat exchangers.
Counter current flow v/s Co-current flow :
The temperature gradient in case of parallel flow is maximum at the entrance and
continuously decreases towards the exit, whereas the temperature gradient is fairly constant
over the length of heat exchanger in case of counter current flow. Hence, with a counter
current flow arrangement, the heating surface has nearly constant capacity through the
exchanger and with a parallel flow arrangement, the capacity at exit is much less as
compared to that at the entrance. With parallel flow arrangement the lowest temperature
theoretically attainable by a hot fluid is that of the outlet temperature of a cold fluid. Under
this condition, the log mean temperature difference would be zero and the heat transfer
surface requirement would be infinite. [Q = U.A. ∆Tlm, U and A are finite]. With parallel
flow, it is not possible to bring the hot fluid temperature below the outlet temperature of the
cold fluid and thus has a considerable effect on the ability of heat exchanger to recover heat.
In a parallel flow heat exchanger, heat transfer is restricted by the cold fluid outlet
temperature rather than the cold fluid inlet temperature and hence the counter flow
arrangement is very common in heat transfer apparatus. The parallel flow arrangement is
used whenever it is necessary to limit the maximum temperature of the cooler fluid.
In counter current flow, it is possible for the cooling liquid to leave at a higher
temperature than the heating fluid, and one of the greatest advantages of counter-flow is that
it is possible to extract a higher proportion of the heat content of the heating fluid. For the
same terminal temperatures, the logarithmic mean temperature difference for counter flow is
appreciable greater than that for co-current flow. Thus, for the same terminal temperatures
and same heat load, the heat transfer area required for a counter flow heat exchanger is less
Unit Operations – II 3.26 Convection

than it for a parallel flow heat exchanger. The rate of heat transfer in counter current flow
heat exchangers is more than it in co-current flow exchangers.
Co-current flow
Counter-current flow Concurrent flow
360
360
340 340
44 60 24
316
40
316 300
300
DTlm = 41.9 K DTlm = 39.3 K
Fig. 3.8
In counter current flow arrangement, temperature difference will show less variation
throughout the heat exchanger. In co-current flow, the temperatures of the two streams
progressively approach one another and the temperature difference will show more variation
throughout the heat exchanger.
Range :
It is the actual rise or fall of temperature of a fluid. If Th1 is the inlet temperature and
Th2 is the outlet temperature of a hot fluid, then Th1 – Th2 is the range for the hot fluid.
Similarly if Tc1 and Tc2 are the inlet and outlet temperatures of a cold fluid, then Tc2 – Tc1 is
the range for the cold fluid.
Approach :
It is the terminal point temperature difference between hot and cold fluids. Thus Th1 is
the inlet temperature of a hot fluid and Tc2 is the outlet temperature of a cold fluid at one end
of counter current heat transfer apparatus, then Th1 – Tc2 is called the approach.
In a condenser where the vapour entering it is not superheated and condensate is not
subcooled below its boiling temperature, the temperature throughout the shell side of the
condenser is constant as pressure in the shell space is constant. Fig. 3.9 shows a temperature
v/s length plot for a condenser. The temperature of the coolant used continuously increases
as it passes through the tubes of the condenser.
Temperature of
condensing vapour
Th Th
Temperature

Tc
2

DT1 = Th – Tc
Tc Tube side fluid 1
1 (Coolant) DT2 = Th – Tc
2

0 Length L

Fig. 3.9 : Temperature-length curve for condenser

Unit Operations – II 3.27 Convection

Energy balances :
The energy balances are of great importance in attacking the heat transfer problems.
Consider the heat exchanger shown in Fig. 3.10.
Cold fluid in Cold fluid out
Hot fluid out Heat Exchanger Hot fluid in

Fig. 3.10 : Energy balance

. .
Let mc and mh be the mass flow rates of the cold and hot fluids respectively.
Let Hc1 and Hc2 be the enthalpies per unit mass of the cold fluid entering and leaving
the heat exchanger, respectively.
Similarly, let Hh1 and Hh2 be the enthalpies per unit mass of the hot fluid entering and
leaving the heat exchanger respectively.
Qc and Qh be the rate of heat addition to the cold fluid and hot fluid respectively. The
sign of Qh is negative since the hot fluid losses/gives out, rather than gains heat, while the
sign of Qc is positive. At steady state, the heat given out by the hot fluid is equal to the heat
gained by the cold fluid if no heat loss from the exchanger.
∴ Qc = – Qh … (3.78)
. .
mc (Hc2 – Hc1 ) = – mh [Hh2 – Hh1 ] = Q … (3.79)
. .
mc [Hc2 – Hc1 ] = mh [Hh2 – Hh1 ] = Q … (3.80)
If only sensible heat transfer takes place between hot and cold fluids and if constant heat
capacities are assumed, then Equation (3.80) can be written as
. .
mc Cp [Tc – Tc ] = mh Cp (Th – Th ) = Q … (3.81)
c 2 1 h 1 2
where Cpc and Cp are the heat capacities of the cold and hot fluids respectively.
h
Tc1, Tc2 and Th1 and Th2 have their usual meaning. Equation (3.81) is the overall energy
or enthalpy balance over a heat exchanger equipment for sensible heat changes in both the
fluids.
For the latent heat transfer (i.e., condensation of vapours) from the condensing vapours to
a coolant, the energy balance will be :
. .
mh · λ = mc Cpc (Tc2 – Tc1) = Q … (3.82)
.
where mh – rate of condensation of vapour
λ – latent heat of condensation of vapour
Equation (3.82) is applicable to a condenser where the vapour entering is saturated and
no subcooling of the condensed liquid takes place.
In the SI system, the mass flow rate is expressed usually in kg/s, the specific heat in
J/(kg·K) or kJ/(kg·K) and the rate of heat transfer in W ≡ J/s.
Unit Operations – II 3.28 Convection

Sensible Heat :
It is the heat that must be transferred to raise or lower the temperature of a substance or a
mixture of substances.
Latent Heat :
When matter undergoes a phase change (vapour to liquid, liquid to solid, etc.), the
enthalpy change associated with a unit amount of matter at constant temperature and pressure
is known as the latent heat of phase change. It is denoted by 'λ' and has the units of kJ/kg.
Heat Flux :
It is defined as the rate of heat transfer per unit area (or defined as the amount of heat
transfer per unit area per unit time). It can be based on the outside or inside heat transfer
area. It has the units of W/m2.
Log mean temperature difference :
The heat transfer flux is directly proportional to a driving force. The driving force for
heat flow is taken as Th – Tc, where Th and Tc are the temperatures of hot and cold fluids
respectively. As the term ∆T = Th – Tc varies along the length of a heat exchanger, the flux
also varies over the entire length. Consider a differential element of area dA through which a
differential amount of heat dQ flows under the driving force of ∆T. Then, dQ/dA is related to
∆T by the relation given below.

Th
1
DT2
Tc
2
Th
Temperature

DT1

Tc
1
DT v/s
Q
DT1
DT2
0 Q QT

Fig. 3.11 : Temperature v/s heat flow rate (counter-current flow)

dQ
dA = U · (∆T) = U · (Th – Tc) … (3.83)
where U is the local overall heat transfer coefficient, dQ/dA is the local heat flux and ∆T is
the local temperature difference.
Equation (3.83) needs to be integrated for its application to the entire area of the heat
exchanger.
Assumptions to be made for the integration of Equation (3.83) are :
1. Overall coefficient U is constant throughout the exchanger.
2. Specific heats of the hot and cold fluids are constant.
Unit Operations – II 3.29 Convection

3. Heat flow to and from the ambient is negligible.

4. Flow is steady and may be parallel or counter current type and
5. Temperatures of both the fluids are uniform over a given cross-section and may be
represented by their bulk temperatures.
Based on the assumptions (2) and (4), we get straight lines if Tc and Th are plotted
against Q and also a plot of ∆T v/s Q is a straight line. Such plots are shown in Fig. 3.11.
The slope of the plot of ∆T v/s Q is constant.
d (∆T) ∆T2 – ∆T1
Slope = dQ = QT … (3.84)

where ∆T1 and ∆T2 are the terminal temperature differences (i.e., approaches)
and QT is the rate of heat transfer in the entire heat exchanger.
Substituting for dQ from Equation (3.83), Equation (3.84) becomes
d (∆T) ∆T2 – ∆T1
= QT … (3.85)
U · ∆T dA
Rearranging Equation (3.85) gives
d (∆T) (∆T2 – ∆T1) U
= QT · dA … (3.86)
∆T
Integrating the above equation over the limits : A = 0, ∆T = ∆T1 and A = A, ∆T = ∆T2,
gives
∆T2 A
d (∆T) (∆T2 – ∆T1) U
⌠
⌡ ∆T = ⌠
⌡ dA … (3.87)
QT
∆T1 0

∆T2 (∆T2 – ∆T1)

ln   = U·A … (3.88)
 1
∆T QT
(∆T2 – ∆T1)
QT = U.A. … (3.89)
∆T2
ln  
∆T1
The heat transfer rate in the entire heat exchanger can be denoted by the symbol Q.
(∆T2 – ∆T1)
∴ Q = U.A … (3.90)
∆T2
ln  
∆T1
Q = U.A. ∆Tlm … (3.91)
(∆T2 – ∆T1)
where ∆Tlm = … (3.92)
∆T2
ln  
∆T1
∆Tlm is referred to as the logarithmic mean or the log mean temperature difference
(LMTD).
Unit Operations – II 3.30 Convection

LMTD for counter current flow :

Cold fluid
Tc1 –––––––––→ Tc2 , ∆T1 = Th2 – Tc1
Hot fluid
Th2 ←–––––––– Th1 , ∆T2 = Th1 – Tc2
LMTD for parallel/co-current flow :
Cold fluid
Tc1 –––––––––→ Tc2 , ∆T1 = Th1 – Tc1
Hot fluid
Th1 –––––––––→ Th2 , ∆T2 = Th2 – Tc2
∆T1 – ∆T2
LMTD = ∆Tlm =
∆T1
ln  
∆T2
In a 1-2 heat exchanger, the tube side fluid flows twice through the exchanger and the
shell side fluid flows once through it. Consequently, there is a combination of the co-current
and counter-current flow patterns in such multipass units. The rate of heat transfer in these
units using the corrected LMTD is given by
Q = U.A. FT (∆Tlm) … (3.93)
where FT = Correction factor for the LMTD and is usually taken as 0.85 – 0.90
and ∆Tlm = Log mean temperature difference calculated for counter current flow
from the terminal temperatures or counter current LMTD.
For cross flow arrangement, the true ∆Tlm is calculated as :
∆Tlm for cross flow = FT · ∆Tlm for counter current flow, where FT is a correction factor.
The heat transfer rate is given by Equation (3.93).
Heat transfer in condensation of single vapours :
The processes involving a phase change, like condensation and boiling are also very
important in industrial practice.
The change from liquid to vapour state is known as vaporisation and that from vapour to
liquid is known as condensation. In either case, the latent heats involved are identical. In the
condensation of a pure vapour, it is necessary to remove the latent heat of vaporisation.
Condensation is a convection process that involves a change of phase from vapour to liquid
and it occurs whenever a saturated vapour comes into contact of a cold surface, for example,
in surface condensers, heat transfer from the vapour to the surface takes place and the vapour
gets condensed on the surface.
The process of condensation, which is the reverse of boiling, occurs by two distinct
mechanisms/modes and that too at very different rates of heat transfer. The two distinct
mechanisms are :
(i) dropwise condensation and
(ii) filmwise condensation.
The condensing film coefficient depends upon the nature of surface and whether the
surface, on which condensation occurs, is mounted vertically or horizontally.
Unit Operations – II 3.31 Convection

For filmwise condensation on a vertical surface, the mean heat transfer coefficient is
given by
ρ2 g λ k31/4
hm = 0.943   … (3.94)
 µ L ∆Tf 
where hm - mean heat transfer coefficient over the tube length
L - tube length
g - gravitational acceleration
µ - liquid viscosity
λ - latent heat of vaporisation
k - thermal conductivity of liquid
∆Tf - temperature difference between vapour and metal
For filmwise condensation on a horizontal tube of outside diameter do, the mean heat
transfer coefficient hm is given as :
ρ2 g · λ k31/4
hm = 0.725   … (3.95)
 µ do ∆Tf 
Drop-wise Condensation :
When a saturated vapour comes into contact with a cold surface (a surface at a lower
temperature), it condenses and if condensate does not wet the surface, the droplets are
formed on the surface. These droplets grow and ultimately fall down the surface in a random
fashion under the influence of gravity leaving behind the bare metal surface on which further
condensation takes place. The condensation occurring by this mechanism is known as drop-
wise condensation.
Film-wise Condensation :
When a saturated vapour comes into contact with a cold surface, it condenses and if
condensate wets the surface, it forms a continuous film of condensate, which covers the
entire surface, through which heat must be transferred. The additional vapour is then required
to condense into the liquid film rather than directly on the surface. The condensate ultimately
flows down the surface under the influence of gravity. The condensation occurring by this
mechanism is called as filmwise condensation.
In film-wise condensation, the film covering the surface acts as a resistance to heat
transfer, while in drop-wise condensation, a large portion of the surface is directly exposed to
the vapour. Because of this the heat transfer coefficients (hence the heat transfer rates) in
drop-wise condensation are four to eight times larger than those for filmwise condensation.
Drop-wise condensation needs a physical preparation of surfaces which is very difficult
and needs to be promoted by introducing certain additives into the vapour stream. This type
of condensation is very unstable, i.e., difficult to maintain and also the design methods are
not available. Because of these reasons, dropwise condensation is not common in industry.
In general, smooth, clean, uncontaminated surfaces seem to promote film-wise condensation
and oily or greasy surfaces and presence of dirt on the surfaces, i.e., when impurities are
present and surface is contaminated, seem to promote dropwise condensation.
Unit Operations – II 3.32 Convection

Film-wise condensation is very common and reliable since most surfaces become wetted
after being exposed to the condensing vapours over a period of time and it is difficult to
achieve dropwise condensation and therefore the condensing equipments in use are designed
on the basis of film-wise condensation.
Condensers are widely used in the chemical industry. For example, in the separation of
the constituents of a liquid mixture by distillation, a condenser converts vapours from the top
of the column into liquid distillate and reflux. In power plants the surface condensers are
used for condensing steam from the exhaust of turbines into liquid. The home air
conditioning unit makes a use of air cooled condenser for liquefication of refrigerant.
Effect of non-condensable gases :
If a non-condensable gas is present in condensing vapours, for example, air in the
condensing vapours in a condenser because of say leakage, it hinders the process of heat
transfer. The non-condensable gas collects in the viscinity of condensate surface, and the
condensing vapour has to diffuse through the gas film. The presence of this diffusion
resistance in the process of condensation decreases the rate of condensation far below than
that for a pure material. Presence of about 1% by volume air in the condensing vapours can
reduce the heat transfer coefficient by 60% of its value for no air. Thus, an air vent is always
provided on almost all the condensers to eliminate air from the system.
Difference between drop-wise and film-wise condensation :
Drop-wise condensation Film-wise condensation
1. In drop-wise condensation the 1. In film-wise condensation the condensed
condensate does not wet the surface and liquid wets the surface and forms a
collects in the form of droplets. These continuous film of condensate on the
droplets grow for a while and then fall surface through which heat transfer takes
from the surface, leaving behind the bare place. The condensate flows down the
metal surface for further condensation. surface under the action of gravity.
2. Heat transfer coefficients are very high 2. Heat transfer coefficients are relatively
as the heat does not have to flow through very low since the heat does have to
a film by conduction. flow through a film by conduction.
3. Oily or greasy surfaces seem to tend 3. Smooth, clean uncontaminated surfaces
towards drop-wise condensation. seem to tend towards film-wise
condensation.
4. Drop-wise condensation is very difficult 4. Film-wise condensation is easily
to achieve. obtainable.
5. Drop-wise condensation is unstable and 5. Film-wise condensation is stable, easy,
difficult to maintain, uncommon and so reliable, common and so desired
not desired industrially. industrially.
6. Methods for predicting the film 6. Methods for predicting the film
coefficient are not available. coefficients are available.
7. Heat transfer coefficients and heat 7. Heat transfer coefficients and heat
transfer rates are very high. transfer rates are low.
Unit Operations – II 3.33 Convection

Table 3.3 : Average values of film coefficients (hm) for condensation of

pure saturated vapour on horizontal tubes
Vapour hm in W/(m2·K) Range of ∆Tf
Steam 10000 – 28000 1 – 11
Steam 18000 – 37000 4 – 37
Benzene 1400 – 2200 23 – 37
Methanol 2800 – 3400 8 – 16
Ethanol 1800 – 2600 6 – 22
Toluene 1100 – 1400 31 – 40
Heat transfer to boiling liquids :
The phenomena of boiling, opposite/reverse of condensation, is commonly encountered
in the unit operations such as distillation and evaporation and steam generation. In almost all
cases where the condensation is carried out, boiling apparatus associates it. In the chemical
industry usually the boiling takes place either on a hot submerged surface, e.g., kettle reboiler
or inside a vertical tube, e.g., vertical tube evaporator. In boiling practice, initially the vapour
is formed in the form of bubbles and afterwards as a distinct vapour phase above the liquid
interface.
Boiling is a convection process that involves a change in phase from liquid to vapour.
It may occur when a liquid is exposed to a solid surface maintained at a temperature higher
than the saturation temperature of the liquid. In this case, the heat flux will depend upon the
difference in temperature between the surface and the saturation temperature of the liquid.
The rate of heat transfer from the solid surface to the liquid is given by
Q = h (Tw – Ts)
where Tw is the temperature of the surface and Ts is the saturation of the liquid and the
difference Tw – Ts is known as the excess temperature.
When heat is added to a liquid from a submerged solid surface, then the boiling process
is termed as pool boiling. (Pool boiling is the type of boiling in which the heating surface is
surrounded by or submerged in a relatively large body of the liquid which is agitated by the
motion of the bubbles and by natural convection currents.) If the bulk liquid temperature is
less than the saturation temperature, the process is called as subcooled or local boiling. In
this case, the bubbles formed at the surface eventually condense in the liquid. If the liquid
temperature is equal to the saturation temperature, the boiling process is called as saturated
or bulk boiling. When heat is added to a liquid from a submerged surface and the liquid is
maintained at saturation temperature, then the process is called as saturated pool boiling.
The process of boiling depends upon the nature of the surface, thermophysical properties of
the liquid and vapour bubble dynamics. As the boiling process involves a large number of
variables, general equations describing it are not available.
In case of boiling accomplished by a hot submerged surface, the temperature of the liquid
is the same as the boiling point of the liquid at pressure prevailing in the apparatus. Bubbles
of vapour are formed at the heating surface, rise through the pool of liquid, and disengage
from the surface of the liquid. Vapours accumulate in a vapour space above the liquid surface
Unit Operations – II 3.34 Convection

and finally the vapour from the vapour space is removed as fast as it is formed through a
vapour outlet. This type of boiling is referred to as pool boiling of saturated liquid as the
vapours leaving are in equilibrium with the liquid at its boiling temperature.

Liquid

Heating
(a) (b) Q (c)
surface

Fig. 3.12 : Effect of surface tension on bubble formation

Heat transfer by vaporisation in the absence of mechanical agitation is a combination of
free convection and the additional convection generated by the rising stream of bubbles. The
surface tension of liquid against the heating surface is the important factor in controlling the
rate of bubble detachment. If the surface tension of the liquid against the heating surface is
large, the bubble tends to spread along the surface and blankets the surface rather than
leaving from the surface to make a space for the other bubbles. This is shown in Fig. 3.12 (c).
If the surface tension is low, liquid tends to wet the surface and the bubble will pinch off
easily and rises as shown in Fig. 3.12 (a). An example of a liquid having intermediate surface
tension is shown in Fig. 3.12 (b).
Pool Boiling of Saturated Liquid :
Consider a heating element, e.g., a horizontal tube bundle submerged in a pool of
Q
boiling liquid and steam is condensing on the inside of tubes. Let A be the heat flux and
∆T = (Tw – T) be the difference in temperature between the tube wall and the boiling liquid.
Start with a very small temperature drop and go on increasing ∆T (by increasing Tw and
Q
keeping T constant) in a stepwise manner. For each step, measure the value of A and ∆T till
high values of ∆T are obtained.
Maximum flux for
nuclear boiling
C
E

D AB - Natural convection
B BC - Nucleate boiling
Q/A

CD - Transition boiling
DE - Film boiling
A
DT = (Tw – T)
Fig. 3.13 : Heat flux v/s ∆T for water at 373 K (100 °C)
Q
When we plot A v/s ∆T on logarithmic co-ordinates, we will get a curve as shown in
Fig. 3.13.
Unit Operations – II 3.35 Convection

The curve obtained can be divided into four segments : AB, BC, CD and DE. Under very
small temperature drops, the rate of production of bubbles proceeds slowly and the
mechanism of heat transfer is free convection. Bubbles form on the surface, release from it,
rise to the surface of the liquid and finally disengage into a vapour space. Bubbles formed are
very few to disturb the natural convection currents set up in the liquid. This occurs over the
Q
segment AB wherein A is proportional to ∆ T1.25. The segment BC of the curve is also a
straight line with a slope greater than that for the segment AB. As the temperature of the
surface is further increased (i.e., ∆T increased), the rate of bubble production is large
(bubbles are formed more rapidly) and in turn it increases the velocity of the circulation
currents in the mass of liquid and both the heat flux and heat transfer coefficient become
greater than those in free convection. The temperature drop corresponding to point 'C' is
called the critical temperature drop and the corresponding flux is called the peak flux (the
critical heat flux). The action occurring below the critical temperature drop over the segment
BC is called nucleate boiling, where vaporisation takes place directly from the surface since
bubbles formed rise to the surface of the liquid. During nucleate boiling the most of the
surface is in direct contact with the liquid, and only small portion of the surface is occupied
by the bubbles. As the temperature drop is increased, more cite becomes active. Hence, more
bubbles are generated, agitation of the liquid is improved and both these lead to increase the
heat flux and heat transfer coefficient.
Over the segment CD as the temperature drop increases further, the flux decreases and
reaches a minimum at point D. Point D is known as the Leidenfrost point. The boiling
action over the segment CD is known as transition boiling. During this region the rate of
bubble production is so large that they tend to coalesce on the heating surface forming a layer
of insulating vapour (a vapour film). The layer is highly unstable and bubble issues from it in
the form of jets. Further increase in ∆T, increases the thickness of the insulating vapour film
which in turn reduces the number of explosions in a given time and hence both heat flux and
heat transfer coefficient decrease.

B D
A E
h

DT
Fig. 3.14 : Heat transfer coefficient v/s ∆T
Beyond point D, the hot surface gets covered with a stable vapour film that prevents the
inflow of fresh liquid to the surface and the heat is transferred through this vapour film by
conduction to the liquid for any further boiling. As the ∆T increases, the heat flux rises
slowly at first and then rapidly as the radiation heat transfer becomes important. Over the
segment DE, vaporisation takes place through a vapour film and is known as film boiling.
Unit Operations – II 3.36 Convection

The relationship between ∆T and h (heat transfer coefficient) from A to E is shown in

Fig. 3.14. In film boiling, the rate of heat transfer is low for a large temperature drop and the
temperature drop is not utilised effectively. Hence, film boiling is not usually desired in
commercial heat transfer equipments and the heat transfer equipments are generally designed
for the temperature drop less than the critical temperature drop.
SOLVED EXAMPLES
Example 3.1 : A hot fluid enters a double pipe heat exchanger at a temperature of
423 K (150 oC) and is to be cooled to 367 K (94oC) by a cold fluid entering at 311 K (38oC)
and heated to 339 K (66oC). Shall they be directed in parallel or counter-current flow ?
Solution : (I) For counter current flow :
Hot fluid
Th1 , 423 K –––––––––––→367 K, Th2
Cold fluid
Tc2 , 339 K –––––––––––→311 K, Tc1
Th1 – temperature of hot fluid at inlet = 423 K
Th2 – temperature of hot fluid at outlet = 367 K
Tc1 – temperature of cold fluid at inlet = 311 K
Tc2 – temperature of cold fluid at outlet = 339 K
∆T2 = Th2 – Tc1 = 367 – 311 = 56 K
∆T1 = Th1 – Tc2 = 423 – 339 = 84 K
Log mean temperature difference (LMTD) for counter current flow is given by
∆T1 – ∆T2 84 – 56
∆Tlm = = 84 = 69 K
ln 56
ln (∆T1/∆T2)
 
LMTD (counter-current flow) = 69 K
(II) For parallel flow :
hot fluid
423 K –––––––→ 367 K
cold fluid
311 K –––––––→ 339 K
∆T2 = Th2 – Tc2 = 367 – 339 = 28 K
∆T1 = Th1 – Tc1 = 423 – 311 = 112 K
Log mean temperature difference for parallel flow is given by
∆T1 – ∆T2 112 – 28
∆Tlm = =
∆T1 ln  28 
112
ln    
∆T2
LMTD (parallel flow) = 60.6 K … Ans.
Since the LMTD for the counter current flow is more than that for the co-current/parallel
flow, the fluids must be directed in the counter current fashion.
Unit Operations – II 3.37 Convection

Example 3.2 : Thermic fluid flowing at a rate of 5000 kg/h is to be cooled from
423 K (150oC) to 363 K (90oC) by circulating water at a rate of 15000 kg/h. If the water is
available at 303 K (30oC), find the outlet temperature of water.
Data : Specific heat of thermic fluid = 2.72 kJ/(kg·K)
Specific heat of water = 4.187 kJ/(kg·K)
Solution : Basis : 5000 kg/h of thermic fluid.
At steady state, the heat given out by the thermic fluid is equal to the heat gained by the
circulating water. Therefore,
· ·
Q = mt Cpt (T1 – T2) = mw Cpw (t2 – t1)
·
where, mt = 5000 kg/h (mass flow rate of thermic fluid)
Cpt = 2.72 kJ/(kg·K)
T1 = Th = Inlet temperature of thermic fluid
1
= 423 K
T2 = Th = Outlet temperature of thermic fluid
2
= 363 K
·
mw = 15000 kg/h (mass flow rate of circulating water)
Cpw = 4.187 kJ/(kg·K)
t1 = Tc = Inlet temperature of water
1
= 303 K
t2 = Tc = Outlet temperature of water = ?
2
We have :
· ·
mt Cpt (T1 – T2) = mw Cpw (t2 – t1)
5000 × 2.72 (423 – 363) = 15000 × 4.187 (t2 – 303)
Solving for t2 gives
t2 = 316 K (43oC)
Outlet temperature of water = 316 K (43°°C) … Ans.
Example 3.3 : Cold fluid is flowing through the heat exchanger at a rate of 15 m3/h.
It enters the heat exchanger at 303 K [30oC] and leaves at 328 K [55oC]. A hot thermic fluid
enters the heat exchanger at a rate of 21 m3/h at a temperature of 388 K [115oC]. Find the
area of heat transfer required assuming the flow to be counter current and overall heat
transfer coefficient be 3490 W/(m2·K).
Data : Density of cold fluid = 1000 kg/m3
Density of thermic fluid = 950 kg/m3
Specific heat of cold fluid = 4.187 kJ/(kg·K)
Specific heat of thermic fluid = 2.93 kJ/(kg·K)
Unit Operations – II 3.38 Convection

Solution : Basis : 21 m3/h of thermic fluid

·
mt = mass flow rate of thermic fluid
= volumetric flow rate × density
= 21 × 950
= 19950 kg/h
Volumetric flow rate of cold fluid = 15 m3/h
·
Similarly, mc = mass flow rate of the cold fluid
= 15 × 1000
= 15000 kg/h
T1 = Inlet (initial) temperature of the thermic fluid = 388 K
T2 = Outlet temperature of the thermic fluid, K
t1 = Initial temperature of the cold fluid = 303 K
t2 = Outlet temperature of the cold fluid = 328 K
Let us calculate T2.
At steady state, the heat gained by the cold fluid is equal to heat given out by the thermic
fluid. Therefore,
· ·
Q = mt · Cpt (T1 – T2) = mc · Cpc (t2 – t1)
19950 × 2.93 × (388 – T2) = 15000 × 4.187 (328 – 303)
∴ T2 = 361.2 K (88.2oC)
For counter current flow :
Thermic fluid
388 K ––––––––––→ 361.2 K
Cold fluid
328 K ←–––––– 303 K
∴ ∆T1 = T1 – t2 = 388 – 328 = 60 K
∆T2 = T2 – t1 = 361.2 – 303 = 58.2 K
Let us calculate LMTD.
∴ ∆Tlm = LMTD
(∆T2 – ∆T1) 60 – 58.2
= = = 59.1 K
∆T2  60 
ln   ln
58.2
∆T1
Let us calculate the rate of heat transfer based on the thermic fluid.
Q = Rate of heat transfer
·
= mt · Cpt (T1 – T2)
= 19950 × 2.93 × (388 – 361.2)
= 1566554 kJ/h
= 435154 J/s ≡ 435154 W
Unit Operations – II 3.39 Convection

Let us find the area of heat transfer.

We know : Q = U.A. ∆T lm
where U = overall heat transfer coefficient
= 3490 W/(m2·K)
∆Tlm = LMTD = 59.1 K
A = area of heat transfer in m2 = ?
Q = Rate of heat transfer
= 435154 W
∴ 435154 = 3490 × A × 59.1
A = Area of heat transfer
= 2.11 m2 … Ans.
Example 3.4 : Calculate the heat transfer area of a 1-2 heat exchanger from the
following data : Inlet and outlet temperatures of the hot fluid are 423 K (150oC) and
353 K (80oC) respectively. Inlet and outlet temperatures of the cold fluid are 303 K (30oC)
and 318 K (45oC) respectively.
∴ Overall heat transfer coefficient = 4100 W/(m2·K)
Heat loss = 407 kW
L.M.T.D. correction factor = 0.84
Solution : The rate of heat loss is given by
Q = U.A. FT ∆Tlm
Given : Q = 407 kW = 407 × 103 W
U = Overall heat transfer coefficient
= 4100 W/(m2·K)
A = Area of heat transfer, m2
∆Tlm = log mean temperature difference for counter current flow
FT = L.M.T.D. correction factor for 1-2 heat exchanger
= 0.84
Let us find LMTD for counter current flow.
hot fluid
423 K → 353 K
cold fluid
318 K ← 303 K
∆T1 = Th1 – Tc2 = 423 – 318 = 105 K

∆T2 = Th2 – Tc1 = 353 – 303 = 50 K

Unit Operations – II 3.40 Convection

∆Tlm = LMTD
∆T1 – ∆T2 (105 – 50)
= =
 1
∆T
ln  50 
105
ln    
∆T2
= 74.13 K
We have : Q = U.A. FT · ∆Tlm
407 × 103 = 4100 × A × 0.84 × 74.13
A = 1.594 m2 ~
– 1.6 m2
Area of heat transfer = 1.6 m2 … Ans.
Example 3.5 : Calculate the inside heat transfer coefficient for a fluid flowing at a rate
of 300 cm3/s through a 20 mm inside diameter tube of heat exchanger.
Data : Viscosity of flowing fluid = 0.8 (N.s)/m2
Density of flowing fluid = 1.1 g/cm3
Specific heat of fluid = 1.26 kJ/(kg·K)
Thermal conductivity of fluid = 0.384 W/(m·K)
Viscosity at wall temperature = 1.0 (N.s)/m2
Length of heat exchanger = 5 m
Solution : Basis : 300 cm3/s of volumetric flow rate of fluid
Volumetric flow rate = 300 cm3/s
= 300 × 10–6 m3/s
Density of the flowing fluid = 1.1 g/cm3
Density of the flowing fluid = 1100 kg/m3
Mass flow rate of the flowing fluid = 300 × 10–6 × 1100 = 0.33 kg/s
Inside diameter of the pipe = Di = 20 mm = 0.02 m
2
Area of pipe (cross section) = π/4 Di
= π/4 (0.02)2 = 3.14 × 10–4 m2
Volumetric flow rate
Velocity of flowing fluid = Cross section area
300 × 10–6
u = = 0.955 m/s
3.14 × 10–4
Let us calculate NRe. NRe = Reynolds number
Diuρ
=
µ
Unit Operations – II 3.41 Convection

where Di = 0.02 m
u = 0.955 m/s
ρ = 1100 kg/m3
µ = 0.8 (N.s)/m2 = 0.8 kg/(m.s)
0.02 × 0.955 × 1100
NRe = 0.8 = 26.26
As the Reynolds number is less than 2100, the flow is laminar.
Let us calculate NPr.
Cp µ
NPr = k
where, Cp = 1.26 kJ/(kg·K) = 1.26 × 103 J/(kg·K)
µ = 0.8 kg/(m.s)
k = 0.384 W/(m·K)
1.26 × 103 × 0.8
NPr = 0.384 = 2625

NRe · NPr · L = 26.26 × 2625 ×  5  = 275.73

D 0.02
 
µw = 1.0 (N.s)/m = 1.0 kg/(m·s)
2

Let us calculate hi.

The empirical relation applicable for laminar flow is
D 1/3  µ 0.14
NNu = 1.86 NRe · NPr · L   
  µw

NNu = 1.86 (26.26) (2625) × 5  × 1.0

0.02 1/3 0.8 0.14
   
= 11.71
hi Di
We have : NNu = k
where hi is the inside heat transfer coefficient.
hi × 0.02
∴ 11.71 = 0.384
hi = 1225 W/(m2·K) … Ans.
Example 3.6 : Calculate the heat transfer coefficient for fluid flowing through a tube
having inside diameter 40 mm at a rate of 5500 kg/h. Assume that the fluid is being heated.
Data : Properties of fluid at mean bulk temperature :
Viscosity of flowing fluid = 0.004 (N.s)/m2
Density of flowing fluid = 1.07 g/cm3
Specific heat of flowing fluid = 2.72 kJ/(kg.K)
Thermal conductivity of flowing fluid = 0.256 W/(m·K)
Make use of Dittus – Boelter equation.
Unit Operations – II 3.42 Convection

Solution : Basis : 5500 kg/h of fluid flowing through the tube

· 5500
Mass flow rate = m = 5500 kg/h = 3600 = 1.53 kg/s
Density of the fluid = ρ = 1.07 g/cm3 = 1070 kg/m3
· 1.53
Volumetric flow rate = m /ρ = 1070 = 1.43 × 10–3 m3/s

π 2
Cross-sectional area of the tube = 4 D i

where Di = 40 mm = 0.04 m
π
A = 4 (0.040) 2 = 1.256 × 10–3 m2
Let us calculate u.
1.43 × 10–3
Velocity of the flowing fluid = u = = 1.14 m/s
1.256 × 10–3
Let us calculate NRe.
Diuρ
Reynolds number = NRe =
µ
where Di = inside diameter of pipe
= 40 mm = 0.04 m
ρ = 1070 kg/m3
u = 1.14 m/s
µ = Viscosity
= 0.004 (N.s)/m2 = 0.004 kg/(m·s)
0.04 × 1.14 × 1070
NRe = 0.004 = 12198
Let us calculate NPr.
Cp · µ
NPr = k
Cp = 2.72 kJ/(kg·K) = 2.72 × 103 J/(kg·K)
k = 0.256 W/(m·K)
µ = 0.004 kg/(m·s)
2.72 × 103 × 0.004
NPr = 0.256 = 42.5
As NRe is greater than 10000, the flow is turbulent.
Let us obtain hi.
The empirical relationship for turbulent flow, i.e., the Dittus–Boelter equation for heating
is
NNu = 0.023 (NRe)0.8 (NPr)0.4
Unit Operations – II 3.43 Convection

NNu = 0.023 (12198)0.8 (42.5)0.4 = 191.5

NNu = Nusselt number
hi Di
k = 191.5
where, hi
= inside heat transfer coefficient
=
Di inside diameter of pipe
= 0.04 m
k = thermal conductivity of fluid
= 0.256 W/(m·K)
hi (0.04)
0.256 = 191.5
hi = 1225.5 W/(m2·K) … Ans.
Example 3.7 : Calculate the overall heat transfer coefficient from the following data :
Inside heat transfer coefficient = 5800 W/(m2·K)
Outside heat transfer coefficient = 1750 W/(m2·K)
Outside diameter of tube = 30 mm
Inside diameter of tube = 20 mm
Thermal conductivity of metal wall = 46.52 W/(m·K)
Solution : The relationship among U, hi, ho and k/x is
1 1 1 1
U = + +
ho hi k/x … (1)

where U = overall heat transfer coefficient

ho = outside heat transfer coefficient
= 1750 W/(m2·K)
hi = inside heat transfer coefficient
= 5800 W/(m2·K)
k = thermal conductivity of the metal wall
= 46.52 W/(m·K)
x = thickness of the metal wall of tube
= (O.D. – I.D.) / 2
(30 – 20)
= 2 = 5 mm = 0.005 m
Substituting the values in Equation (1) gives
1 1 1 1
U = 1750 + +
5800 46.52
0.005
U = 1175 W/(m ·K) 2

Overall heat transfer coefficient = 1175 W/(m2·K) … Ans.

Unit Operations – II 3.44 Convection

Example 3.8 : A hot fluid enters a double pipe heat exchanger at a temperature of
423 K (150oC) and to be cooled to 363 K (90oC) by a cold fluid entering at 308 K (35oC)
and heated to 338 K (65oC). Shall they be directed in parallel flow or counter current flow to
have a high rate of heat transfer ?
Solution :
(I) Parallel flow arrangement :
Hot fluid
423 K → 363 K
Cold fluid
308 K → 338 K
∆T1 = Th1 – Tc1 = 423 – 308 = 115 K
∆T2 = Th2 – Tc2 = 363 – 338 = 25 K
∆T1 – ∆T2
LMTD for parallel flow =
ln (∆T1/∆T2)
115 – 25
= 115 = 58.97 K

ln 25
 
(II) Counter current flow arrangement :
Hot fluid
423 K → 363 K
Cold fluid
338 K ← 308 K
∆T1 = Th1 – Tc2 = 423 – 338 = 85 K
∆T2 = Th2 – Tc1 = 363 – 308 = 55 K
∆T1 – ∆T2
LMTD for counter current flow =
ln (∆T1/∆T2)
85 – 55
= ln (85/55) = 68.91 K

Since the LMTD for the counter current flow is more than that for the co-current /
parallel flow, the fluids should be directed in the counter current fashion to have a high rate
of heat transfer. … Ans.
Example 3.9 : A cold fluid is flowing through a double pipe heat exchanger at a rate
15 m3/h. It enters at 303 K (30oC) and is to be heated to 328 K (55oC). Hot thermic fluid is
available at the rate of 21 m3/h and at 383 K (110oC). Find the logarithmic mean
temperature difference for co-current and counter-current type of flow.
Data : Sp. heat of thermic fluid = 2.72 kJ/(kg·K)
Density of thermic fluid = 0.95 g/cm3.
Density of water = 1 g/cm3, Specific heat of water = 4.187 kJ/(kg·K)
Unit Operations – II 3.45 Convection

Cold fluid Cold fluid

Solution : 303 K → 328 K (t1) 303 K → 328 K (t2)
Thermic fluid Thermic fluid
383 K → T2 K T2 ← 383 K (T1)
Co-current flow Counter current flow
Mass flow rate of water (cold fluid) = Volumetric flow rate × density
= 15 × 1000
·
mc = 15000 kg/h
Mass flow rate of thermic fluid = Volumetric flow rate × density
= 21 × 950
·
mt = 19950 kg/h
The heat balance over the exchanger is
Heat lost by the thermic fluid = Heat gained by the cold fluid
· ·
mt Cp (T1 – T2) = mc Cp (t2 – t1)
t c
T1 = 383 K, t2 = 328 K, t1 = 303 K
19950 × 2.72 (383 – T2) = 15000 × 4.187 × (328 – 303)
∴ T2 = 354 K (81o C)
∆T1 – ∆T2
LMTD for co-current flow : LMTD =
ln (∆T1 / ∆T2)
∆T1 = Th1 – Tc1 = T1 – t1 = 383 – 303 = 80 K
∆T2 = T2 – t2 = 354 – 328 = 26 K
LMTD for co-current flow :
∆T1 – ∆T2 80 – 26
LMTD = = ln (80/26) = 48 K
ln (∆T1 / ∆T2)
LMTD for counter current flow,
∆T1 = T1 – t2 = 383 – 328 = 55 K
∆T2 = T2 – t1 = 354 – 303 = 51 K
∆T1 – ∆T2 55 – 51
LMTD = = ln (55/51) = 53 K
ln (∆T1/∆T2)
LMTD for concurrent/co-current flow = 48 K
LMTD for counter current flow = 53 K … Ans.
Example 3.10 : Find the overall heat transfer coefficient if :
(i) Inside and outside film heat transfer coefficients are 12 and 11600 W/(m2·K)
respectively.
(ii) Inside and outside diameters are 25 mm and 29 mm respectively.
(iii) Thermal conductivity of metal = 34.9 W/(m·K).
Unit Operations – II 3.46 Convection

Solution : The relationship among U, ho, hi and k/x is

1 1 1 1
U = hi + ho + k/x … (1)
where hi = Inside heat transfer coefficient = 12 W/(m2·K)
ho = Outside heat transfer coefficient = 11600 W/(m2·K)
k = Thermal conductivity of pipe wall material = 34.9 W/(m·K)
Di = Inside diameter of pipe = 25 mm = 0.025 m
Do = Outside diameter of pipe = 29 mm = 0.029 m
x = Thickness of pipe wall
Do – Di 29 – 25
x = 2 = 2 = 2 mm = 0.002 m

Substituting the values of various parameters in Equation (1), we get

1 1 1 1
U = 12 + 11600 + 34.9/0.002
1
U = 0.0835
U = 11.97 W/(m2·K) … Ans.
Example 3.11 : Find the overall heat transfer coefficient from the following data :
(i) LMTD = 23 K for counter current flow
(ii) Heat transfer area = 1.5 m2
(iii) Rate of heat transfer = 116 kW
(iv) Correction factor for LMTD = 0.85
Solution : The rate of heat transfer in this case is given by
Q = U.A. FT . (LMTD) for counter current flow
where, Q = rate of heat transfer = 116 kW = 116 × 103 W
A = Area of heat transfer = 1.5 m2
FT = correction factor for LMTD = 0.85
LMTD = Log mean temperature difference for counter current flow = 23 K
U = Heat transfer coefficient (overall) in W/(m2·K)
116 × 103 = U × 1.5 × 0.85 × 23
116 × 103
U =
1.5 × 0.85 × 23
= 3956 W/(m2·K) … Ans.
Unit Operations – II 3.47 Convection

Example 3.12 : Find the inside heat transfer coefficient using Sieder-Tate equation for
turbulent flow.
Data : I.d. of tube = 20 mm, NRe = 15745
Viscosity of fluid at bulk mean temperature = 550 × 10–5 Pa·s
Viscosity of fluid at average wall temperature = 900 × 10–6 Pa·s
Prandtl number = 36, Thermal conductivity of fluid = k = 0.25 W/(m·K)
Solution : The Sieder-Tate equation is
hi Di  µ 0.14
k = 0.023 (NRe) (NPr)
0.8 1/3
µ  … (1)
 w
 µ 0.14
where, µ  = Sieder-Tate correction factor
 w
NRe = Reynolds number = 15745
NPr = Prandtl number = 36
hi = inside heat transfer coefficient, W/(m2·K)
Di = inside diameter of pipe = 20 mm = 0.002 m
k = Thermal conductivity of fluid = 0.25 W/(m·K)
µ = 550 × 10–6 Pa·s
µw = 900 × 10–6 Pa·s
Substituting the values in Equation (1), we get
hi (0.02) 550 × 10–60.14
0.8 (36)1/3 × 
= 0.023 × (15745) 
0.25 900 × 10–6
hi (0.02)
0.25 = 0.023 × 2278.84 × 3.3 × 0.933

hi (0.02)
0.25 = 161.37

hi = 2017
Inside heat transfer
coefficient = 2017 W/(m2·K) … Ans.

Example 3.13 : Water enters a two-fluid heat exchanger at 328 K and leaves at 358 K.
Hot gases enter at 578 K and leave at 433 K. If the total heat transfer area is 500 m2 and the
overall heat transfer coefficient is 700 W/(m2·K), determine the total heat transferred for
(i) parallel flow and
(ii) counter current flow of the two fluids.
Unit Operations – II 3.48 Convection

Solution : (I) Parallel flow

Hot gas
578 K → 433 K
Water
328 K → 358 K
∆T1 = Th1 – Tc1 = 578 – 328 = 250 K
∆T2 = Th2 – Tc1 = 433 – 358 = 75 K
250 – 75
∆Tlm = 250 = 145.35 K

ln 75
 
The rate of heat transfer for parallel flow is
Q = UA ∆Tlm
= 700 × 500 × 145.35
= 50.87 × 103 kW … Ans. (i)
(II) Counter current flow
Hot gas
578 K → 433 K
Water
358 K ← 328 K
∆T1 = Th1 – Tc2 = 578 – 358 = 220 K

∆T2 = Th2 – Tc1 = 433 – 328 = 105 K

220 – 105
∆Tlm = 220 = 155.5 K
ln 105
 
The rate of heat transfer for counter current flow is
Q = U.A. ∆Tlm
= 700 × 500 × 155.5
= 54.42 × 103 kW … Ans. (ii)
Example 3.14 : Crude oil flows at a rate of 1000 kg/h through the inside pipe of a
double pipe heat exchanger and is heated from 303 K (30oC) to 363 K (90oC). The heat is
supplied by kerosene initially at 473 K (200oC) flowing through the annular space. If the
temperature of approach (minimum temperature difference) is 10 K, determine the heat
transfer area for co-current flow and the kerosene flow rate.
Cp for crude oil = 2.1 kJ/(kg·K)
Cp for kerosene = 2.51 kJ/(kg·K)
Uo = 465 W/(m2·K)
Unit Operations – II 3.49 Convection

Solution : Let T be the outlet temperature of kerosene.

For co-current flow, we have
Crude oil
(t2) 363 K ← 303 K (t1)
Kerosene
T2, K ← 473 K (T1)
Given : Temperature of approach = 10 K. Approach at the outlet of the exchanger is the
difference in the temperatures of kerosene and crude oil (T2 – t2).
Hence, the outlet temperature of kerosene is :
T2 = 363 + 10 = 373 K
For co-current flow :
∆T1 = T1 – t1 = 473 – 303 = 170 K
∆T2 = T2 – t2 = 373 – 363 = 10 K
170 – 10
∆Tlm = 170 = 56.47 K
ln  10 
 

Given :  the crude oil  = mc = 1000 kg/h = 0.278 kg/s

Mass flow rate of ·

Cpc = 2.1 kJ/(kg·K) = 2.1 × 103 J/(kg·K)

Uo = 465 W/(m2·K)
The rate of heat transfer is given by
Q = UA ∆Tlm
The rate of heat transfer to crude oil is given by
⋅
Q = mc Cpc (t2 – t1)

∴
⋅
UA ∆Tlm = mc Cpc (t2 – t1)
· 0.278 × 2.1 × 103 × (363 – 303)
Ao = mc Cpc ∆Tc/Uo ∆Tlm =
465 × 56.47
= 1.33 m 2

A heat balance yields :

Heat gained by crude oil = Heat removed from kerosene
· ·
mc Cpc ∆Tc = mk Cpk ∆Tk
where Cpk = 2.51 kJ/(kg·K)
·
mc = 1000 kg/h, ∆Tc = t2 – t1 = 363 – 303 = 60 K
·
mk = mass flow rate of kerosene
∆Tk = T1 – T2 = 473 – 373 = 100 K
Unit Operations – II 3.50 Convection

⋅
· m Cpc ∆Tc 1000 × 2.1 × 60
mk = =
Cpk ∆Tk 2.51 × 100
= 502 kg/h
Mass flow rate of kerosene = 502 kg/h … Ans.
Example 3.15 : Methyl alcohol flowing in the inner pipe of a double pipe heat exchanger
is cooled with water flowing in the outer pipe. The inside and outside diameters of the inner
pipe are 26 mm and 35 mm, respectively. The thermal conductivity of steel is 50 W/(m·K).
The individual film coefficients and fouling factors are :
Alcohol coefficients = 250 W/(m2·K)
Water coefficients = 500 W/(m2·K)
Inside fouling factor = 0.86 × 10–3 (m2·K)/W
Outside fouling factor = 1.7 × 10–3 (m2·K)/W
Calculate the overall coefficients based on the outside area of the inner pipe including
dirt factors and excluding dirt factors.
Solution :
Inside diameter = Di = 26 mm = 0.026 m
Outside diameter = Do = 35 mm = 0.035 m
35 – 26
xw = 2 = 4.5 mm = 0.0045 m
Dw = log mean diameter
Do – Di
=
Do
ln  D 
 i
0.035 – 0.026
= 0.035 = 0.0303 m

ln 0.026
 
k for steel = 50 W/(m·K)
Overall heat transfer coefficient based on the outside area of inner pipe (Uo) is given by
1 1 1 Do xw · Do Do
= + · + + Rd o + Rd i D 
U ho hi Di k · Dw  i
This equation includes dirt factors.

–3 
1 1 0.035 1 0.0045 0.035 0.035
Uo = 500 + 0.026 × 250 + 50 × 0.0303 + 1.7 × 10 + 0.86 × 10 0.026
–3

Uo (inclusive of dirt factors) = 96.65 W/(m2·K) … Ans.

Uo excluding dirt factors is
1 1 1 Do xw · Do
Uo = +
ho hi × Di + k · Dw
Unit Operations – II 3.51 Convection

1 1 1 0.035 0.0045  0.035 

Uo = 500 + 250 · 0.026 + 50 0.0303
Uo = 133.56 W/(m2·K)
Uo (excluding dirt factors) = 133.56 W/(m2·K) … Ans.
Example 3.16 : Heat is transferred from one fluid to a second fluid across a heat
transfer surface. If the heat transfer coefficients for the two fluids are 1.0 and 1.5 kW/(m2·K),
respectively, the metal wall is 6 mm thick and the scale coefficient is equivalent to
850 W/(m2·K), calculate the overall heat transfer coefficient.
Take thermal conductivity of metal as 20 W/(m·K).
Solution : The relationship among U, hi, ho, etc. is given by
1 1 xw 1
U = ho + k + hi + Rd
where ho = 1 kW/(m2·K) = 1000 W/(m2·K)
hi = 1.5 kW/(m2·K) = 1500 W/(m2·K)
xw = 6 mm = 0.006 m
k = 20 W/(m·K)
Rd = 1/850 (m2·K)/W
1 1 0.006 1 1
U = 1000 + 20 + +
1500 850
= 0.001 + 0.003 + 0.00118 + 0.00067 = 0.00315
∴ U = 317.5 W/(m2·K) = 0.318 kW/(m2·K) … Ans.
Example 3.17 : 27 t/h of pure isobutane is to be condensed at 332 K in a horizontal
tubular heat exchanger using water as a cooling medium. Water enters at 300 K and leaves
the exchanger at 315 K. Calculate the log mean temperature difference, heat load, and the
mass flow rate of cooling water.
Data : Latent heat of vaporisation of isobutane is 286 kJ/kg and specific heat of water is
4.187 kJ/(kg·K).
Solution :
.
Mass flow rate of isobutane (m1) = 27 t/h
= 7.5 kg/s
λ for isobutane = 286 kJ/kg
The heat duty/heat load of condenser is given by
·
Q = m1 λ
= 7.5 × 286 = 2145 kJ/s
= 2145 kW
= 2145 × 103 W … Ans.
Unit Operations – II 3.52 Convection

A heat balance over the condenser yields

Heat removed from vapours = Heat gained by water
· ·
Q = m1λ = m2 Cp2 (t2 – t1)
·
where m2 = ?, t2 = 315 K, t1 = 300 K
Cp2 = 4.187 kJ/(kg·K)
·
2145 = m2 × 4.187 × (315 – 300)
·
m2 = 34.15 kg/s
Mass flow rate of cooling water = 34.15 kg/s … Ans.
∆Tlm = log mean temperature difference
For the condenser, we have
Isobutane
332 K → 332 K
Water
300 K → 315 K
∆T1 = 332 – 300 = 32 K, ∆T2 = 332 – 315 = 17 K
∆T1 – ∆T2
∆Tlm =
∆T1
ln  
∆T2
(32 – 17)
=
ln 17
32
 
= 23.8 K (23.8oC) … Ans.
Example 3.18 : Determine the heat transfer coefficient for water flowing in a tube of
16 mm diameter at a velocity of 3 m/s. The temperature of the tube is 297 K (24oC) and the
water enters at 353 K (80oC) and leaves at 309 K (36oC). Using (i) Dittus - Boelter equation
and (ii) Sieder-Tate equation.
Data : Properties of water at 331 K (58oC), i.e., at arithmetic mean-bulk temperature
are : ρ = 984.1 kg/m3, Cp = 4187 J/(kg·K), µ = 485 × 10–6 Pa·s, k = 0.657 W/(m·K)
Viscosity of water at 297 K (24oC), µw = 920 × 10–6 Pa·s
Solution : Let us find NRe.
Duρ
NRe =
µ
where D = 16 mm = 0.016 m, u = 3 m/s
µ = 485 × 10–6 Pa·s or (N·s)/m2 = 485 × 10–6 kg/(m·s)
ρ = 984.1 kg/m3 at arithmetic mean bulk temperature
0.016 × 3 × 984.1
NRe = = 97395
485 × 10–6
Unit Operations – II 3.53 Convection

Let us find NPr.

Cp µ
NPr = k
where k = 0.657 W/(m·K)
Cp = 4187 J/(kg·K)
4187 × 485 × 10–6
NPr = 0.657 = 3.09
(i) The Dittus-Boelter equation for cooling is
NNu = 0.023 (NRe)0.8 (NPr)0.3
hD
k = 0.023 (NRe) (NPr)
0.8 0.3

where h is the film heat transfer coefficient.

k
h = 0.023 × (NRe)0.8 (NPr)0.3 × D

h = 0.023 (97395)0.8 (3.09)0.3 × 0.016

0.657
 
h = 12972.6 W/(m ·K)2 … Ans.
(ii) The Sieder-Tate equation is
hD
NNu = k = 0.023 (NRe)0.8 (NPr)1/3 (µ/µw)0.14
h = 0.023 (NRe)0.8 (NPr)1/3 (µ/µw)0.14 (k/D)
where µ = 485 × 10–6 Pa·s, µw = 920 × 10–6 Pa·s
485 × 10–60.14 0.657
h = 0.023 (97395)0.8 (3.09)1/3   × 0.016
920 × 10–6
= 12267.7 W/(m2·K) … Ans.
Example 3.19 : A heat exchanger is required to cool 20 kg/s of water from 360 K (87oC)
to 340 K (67oC) by means of 25 kg/s of water entering at 300 K (27oC). If the overall
coefficient of heat transfer is constant at 2000 W/(m2·K), calculate the surface area required
in
(a) a counter current concentric tube exchanger, and
(b) a co-current flow concentric tube exchanger.
Data : Take Cp for water = 4.187 kJ/(kg·K)
.
Solution : Mass flow rate of hot water (mhw) is 20 kg/s.
The rate of heat removal from hot water (heat load) is given by
·
Q = mhw · Cp · (T1 – T2) = 20 × 4.187 (360 – 340)
= 1674.8 kJ/s = 1674.8 × 103 W
Let t2 be the cooling water outlet temperature.
Unit Operations – II 3.54 Convection

The rate of heat gained by cold water is given by

·
Q = mcw Cp (t2 – t1)
But rate of heat removal = rate of heat gained
∴ 1674.8 = 25 × 4.187 (t2 – 300)
t2 = 316 K (43o C)
(I) Counter current flow :
(360 K) T1 ––→ T2 (340 K)
(316 K) t2 ←––– t1 (300 K)
∆T1 = T1 – t2 = 360 – 316 = 44 K
∆T2 = T2 – t1 = 340 – 300 = 40 K
∆Tlm = (∆T1 – ∆T2) / ln (∆T1/∆T2)
(44 – 40)
∆Tlm = ln (44/40) = 41.9 K

The rate of heat transfer is given by

Q = U.A. ∆Tlm
Q
A =
U · ∆Tlm
1674.8 × 103
=
2000 × 41.9
= 19.98 m2 … Ans. (a)
(II) Parallel / Co-current flow :
(360 K) T1 –––→ T2 (340 K)
(300 K) t1 –––→ t2 (316 K)
∆T1 = 360 – 300 = 60 K
∆T2 = 340 – 316 = 24 K
∆T1 – ∆T2
∆Tlm =
∆T1
ln  
∆T2
60 – 24
=
ln 24
60
 
= 39.3 K
Unit Operations – II 3.55 Convection

Heat transfer area :

Q
A =
U ∆Tlm
1674.8 × 103
=
2000 × 39.3
= 21.31 m2 … Ans. (b)
Thus, the heat transfer area required for co-current flow arrangement is more than that
required for counter current flow arrangement.
Example 3.20 : Fuel oil in a tank is to be heated from 288.5 K to 316 K by means of a
steam coil. The oil has a pour point below 288.5 K. This oil stands in the tank and
circulates only by natural convection. Exhaust steam is used and it leaves the coil at
atmospheric pressure but not cooled below 373 K. The film heat transfer coefficient for fuel
oil is 18.26 and the heat transfer coefficient for condensing steam is 4070 W/(m2·K). The
fouling factor for steam is 0.215 (m2·K)/W and for fuel oil is 0.123 (m2·K)/W. The coil is
required to heat 2250 kg/h of fuel oil of specific heat 3.35 kJ/(kg·K). The inside diameter of
the coil is 40 mm and the outside diameter is 48 mm. Calculate :
(a) the overall co-efficient of heat transfer,
(b) the heat transfer surface area.
Solution : The rate of sensible heat transfer to oil is given by
.
Q = m Cp ∆T
.
where m = 2250 kg/h
Cp = 3.35 kJ/(kg·K)
∆T = (316 – 288.5) = temperature rise of oil
Q = 2250 × 3.35 × (316 – 288.5)
= 20.73 × 104 kJ/h
= 57578 J/s ≡ 57578 W
Given : Di = 40 mm = 0.04 m
Do = 48 mm = 0.048 m
hi (for steam) = 4070 W/(m2·K)
ho (for oil) = 18.26 W/(m2·K)
Rdo = 0.123 (m2·K)/W

Rdi = 0.215 (m2·K)/W

Unit Operations – II 3.56 Convection

The relationship among U, hi, ho etc., neglecting the metal wall resistance, based on the
outside diameter is
1 1 1 Do Do
Uo = +
ho hi Di + R d o
+ R d  
i  Di 

1 1 1 0.0480 0.0480
Uo = 18.26 + 4070  0.040  + 0.123 + 0.215  0.040 
Uo = 2.3 W/(m2·K)
Steam temperature is constant and is equal to 373 K. Therefore,
∆T1 = 373 – 288.5 = 84.5 K
∆T2 = 373 – 316 = 57 K
∆T1 – ∆T2
∆Tlm =
ln (∆T1/∆T2)
84.5 – 57
= ln (84.5/57)

= 69.85 K
The rate of heat transfer is given by
Q = Uo Ao ∆Tlm
Q
Ao =
Uo ∆Tlm
57578
Ao = = 358.4 m2
2.3 × 69.85
Heat transfer area = Ao = 358.4 m2 … Ans.
Example 3.21 : A heat exchanger is to be designed to heat 1720 kg/h of water from
293 K to 318 K with steam condensing on the outside surface of brass tubes of o.d. 25 mm
and i.d. 22.5 mm and 4 m long. The water velocity is 1.02 m/s, find the number of tubes.
ktube material = 111.65 W/(m·K)
Weight of steam condensed = 4500 kg/h
Latent heat of vaporization of water = 2230 kJ/kg
Temperature of steam = 383 K
Steam side film coefficient = 4650 W/(m2·K)
Physical properties of water at the mean temperature are as follows :
ρ = 995.7 kg/m3
Cp = 4.174 kJ/(kg·K)
k = 0.617 W/(m·K)
(kinematic viscosity) ν = 0.659 × 10–6 m2/s
Unit Operations – II 3.57 Convection

Solution :
.
Mass flow rate of water = m = 1720 kg/h
Cp of water = 4.174 kJ/(kg·K)
T1 = Initial temperature of water = 293 K
T2 = Final temperature of water = 318 K
Rate of heat transfer is equal to the rate of heat gained by water and is given by
.
Q = m Cp ∆T
= 1720 × 4.174 (318 – 293)
Q = 179482 kJ/h = 49856 J/s ≡ 49856 W
Let us calculate NRe.
Di u ρ Di u
NRe = =
µ ν
 µ
as ν = , where ν is the kinematic viscosity
 ρ
Di u
NRe =
ν
where Di = 22.5 mm = 0.0225 m
u = 1.2 m/s
ν = 0.659 × 10–6 m2/s
0.0225 × 1.2
NRe = = 40970
0.659 × 10–6
As NRe > 10000, the flow is turbulent and Dittus-Boelter equation is applicable.
Let us calculate NPr.
Cp = 4.174 kJ/(kg·K)
= 4.174 × 103 J/(kg·K)
ν = Kinematic viscosity of water
µ = Viscosity of water
ν = µ/ρ
µ = ν×ρ
= 0.659 × 10–6 × 995.7
= 0.656 × 10–3 kg/(m·s)
Unit Operations – II 3.58 Convection

NPr = Prandtl number

Cp µ
= k
where k = 0.617 W/(m·K)
4.174 × 103 × 0.656 × 10–3
∴ NPr = 0.617 = 4.44

The Dittus-Boelter equation for heating is

NNu = 0.023 (NRe)0.8 (NPr)0.4
= 0.023 (40970)0.8 (4.44)0.4 = 204.5
hi Di
NNu = k = 204.5
0.617
hi = 204.5 × 0.0225 = 5607 W/(m2·K)

Di = 22.5 mm = 0.0225 m
Do = 25 mm = 0.025 m
Dw = log mean diameter
Do – Di
= ln (D /D )
o i

0.025 – 0.0225
= 0.025 = 0.0237 m
ln 0.0225

The overall heat transfer coefficient based on the outside diameter [Uo] is given by
1 1 1 Do xw Do
Uo = ho + hi × Di + k Dw

where hi = 5607 W/(m2·K)

ho = 4650 W/(m2·K)

k = 111.65 W/(m·K)
xw = (Do – Di)/2
0.025 – 0.0225
= 2

= 1.25 × 10–3 m
Unit Operations – II 3.59 Convection

1 1 1  0.025  1.25 × 10–3  0.025 

Uo = 4650 + 5607 0.0225 + 111.65 0.0237
∴ Uo = 2353 W/(m2·K)
Le us calculate ∆Tlm.
Temperature of condensing steam = 373 K. Therefore,
∆T1 = 373 – 293 = 90 K
∆T2 = 373 – 318 = 65 K
90 – 65
∆Tlm = ln 90/65 = 76.8 K

The rate of heat transfer is given by

Q = Uo Ao ∆Tlm
49856
∴ Ao = = 0.276 m2
2353 × 76.8
We have : Area of heat transfer = Ao = n π Do L
where n = number of tubes
∴ n = Ao/π Do L
0.276
= = 0.88
π × 0.025 × 4
~
– 1
∴ Number of tubes required = 1 … Ans.
Example 3.22 : Estimate the average value of convective film coefficient on the inside
surface of a tube of 50 mm i.d. meant for heating water. The mass flow rate of water is
25000 kg/h, which enters at a temperature of 293 K (20oC) and leaves at 333 K (60oC).
Calculate also the heat transferred per unit length of the tube if the wall temperature is 10 K
above the bulk temperature of the fluid.
Physical properties of water at the mean temperature of 313 K (40o C) are given below :
ρ = 992.2 kg/m3
k = 0.634 W/(m·K)
ν = 0.659 × 10–6 m2/s
and NPr = 4.31
⋅
Solution : Mass flow rate of water = m = 25000 kg/h
Density of water = ρ = 992.2 kg/m3
⋅ 25000
Volumetric flow rate of water = m/ρ = 992.2 = 25.2 m3/h
Unit Operations – II 3.60 Convection

Inside diameter of tube = Di = 50 mm = 0.050 m

25.2 1
Velocity of water = u = 3600 ×
π
4 (0.05)
2

= 3.56 m/s
Di u ρ Di u µ
NRe = = (as ν = )
µ ν ρ
Given : ν = 0.659 × 10–6 m2/s
0.05 × 3.56
NRe = = 270106
0.659 × 10–6
NRe > 10000, the flow is in turbulent and thus Dittus-Boelter equation is applicable.
The Dittus-Boelter equation for heating is
hi Di
k = 0.023 (NRe) (NPr)
0.8 0.4

hi × 0.05
0.634 = 0.023 (270106) (4.31)
0.8 0.4

hi = 11,584 W/(m2·K) … Ans.

Heat transfer per unit = h π D ∆T
length of tube  i i

∴ Q/L = hi π Di ∆T
Since the wall is at a temperature of 10 K above the bulk temperature of fluid, ∆T = 10 K.
∴ Q/L = 11584 × π × 0.05 × 10
= 18196 W/m = 18.2 kW/m … Ans.
Example 3.23 : Water at 303 K (30oC) enters a 25 mm i.d. tube at the rate of 1200 litres
per hour. Steam condenses on the outside surface of the tube, (28 mm o.d.) at a temperature
of 393 K (120oC) and its film heat transfer coefficient may be taken as 5800 W/(m2·K). Find
the length of tube required to heat the water to 343 K (70oC). Assume the thermal
conductivity of metal wall as 950 W/(m·K) and of water as 0.63 W/(m·K). The average
density and viscosity of water may be taken as 0.98 g/cm3 and 0.0006 (Pa·s), respectively.
Solution : Water flow rate = 1200 l/h
density of water = 0.980 g/cm3 = 0.980 kg/l
.
Mass flow rate of water = m = 1200 × 0.98 = 1176 kg/h = 0.33 kg/s
DG
NRe =
µ
Di .
= (m/Ai)
µ
Unit Operations – II 3.61 Convection

Di = 25 mm = 0.025 m
µ = 0.0006 Pa·s = [(N·s)/m2]
= 0.0006 kg/(m·s)
0.025 0.33
NRe = 0.0006 × = 28011
π
4 (0.025) 2

Cp µ
NPr = k
4.187 × 103 × 0.0006
= 0.63 = 4.0
Since NRe > 10,000, the flow is turbulent
The Dittus-Boelter equation for heating is
hi Di
k = 0.023 (NRe)0.8 (NPr)0.4
0.63
hi = 0.023 (28011)0.8 (4)0.4 × 0.025
∴ hi = 3646 W/(m2·K)
Given : Do = 28 mm = 0.028 m
Di = 0.025 m
0.028 – 0.025
∴ xw = 2 = 0.0015 m
0.028 – 0.025
Dw = = 0.0265 m
ln 0.025
0.028
 
k for the metal wall = 50 W/(m·K)
Let us obtain U using the following relation.
1 1 1 Do xw Do
Uo = ho hi Di + k · Dw
+

1 1 1 0.028 0.0015 0.028

Uo = + ×
5800 3646 0.025 + 50 × 0.0265
Uo = 1956 W/(m2·K)
Let us calculate ∆Tlm.
∆T1 = 393 – 303 = 90 K
∆T2 = 393 – 343 = 50 K
(90 – 50)
∆Tlm = ln (90/50)

= 68.05 K
Unit Operations – II 3.62 Convection

Rate of heat gained by water is

.
Q = m Cp ∆T
= 1176 × 4.187 (343 – 303)
= 196957 kJ/h
= 54710 J/s ≡ 54710 W
Let the heat transfer area be Ao m2.
Rate of heat transfer from steam to water is given by
Q = Uo Ao ∆Tlm
Q
∴ Ao =
Uo ∆Tlm
54710
= = 0.411 m2
1956 × 68.05
Ao = π Do L = 0.411 m2
L = 0.411/π Do
0.411
=
π × 0.028
= 4.67 m
Length of the tube ~
– 5m … Ans.
Example 3.24 : An oil is cooled from 353 K (80oC) to 313 K (40oC) in an oil cooler.
The inlet temperature of water is 303 K (30oC). Calculate the temperature of cooling water
leaving the cooler and logarithmic mean temperature difference assuming flow to be counter
current, if the mass flow rate of oil and water are 1.4 kg/s and 2.9 kg/s respectively.
Cp for oil = 2.135 kJ/(kg·K)
Cp for water = 4.187 kJ/(kg·K)
Solution : At steady state : heat gained by water = heat given by oil
. .
∴ Q = mw Cpw (t2 – t1) = mo Cpo (T1 – T2)
.
mw = mass flow rate of water = 2.9 kg/s
.
mo = mass flow rate of oil = 1.4 kg/s
Cpw = 4.187 kJ/(kg·K), Cpo = 2.135 kJ/(kg·K)
t2 = ?, t1 = 303 K, T1 = 353 K, T2 = 313 K
Q = 2.9 × 4.187 (t2 – 303) = 1.4 × 2.135 (353 – 313)
∴ t2 = 312.9 K
Outlet temperature of water = 312.9 K (39.9oC) … Ans.
Unit Operations – II 3.63 Convection

Let us calculate ∆Tlm [LMTD] for counter-current flow.

oil
353 K ––––––→ 313 K
water
312.9 K ←––––– 303 K
∆T1 = 353 – 312.9 = 40.1 K, ∆T2 = 313 – 303 = 10 K
∆Tlm = (∆T1 – ∆T2) / ln (∆T1/∆T2)
= (40.1 – 10) / ln (40.1/10)
LMTD = 21.67 K (oC) … Ans.
Example 3.25 : Air is heated in an air heater from 303 K (30oC) to 383 K (110oC) by
hot gases. The hot gases thus cool from 483 K (210o C) to 418 K (145oC). Calculate the log
mean temperature difference between air and gas with the two fluids in (a) parallel flow, and
(b) counter current flow.
Solution : (a) Parallel flow :
Air
303 K ––––––→ 383 K
Hot gas
483 K ––––––→ 418 K
∆T1 = Th1 – Tc1 = 483 – 303 = 180 K
∆T2 = Th2 – Tc2 = 418 – 383 = 35 K
The log mean temperature difference is given by
∆T1 – ∆T2 (180 – 35)
∆Tlm = = ln (180/35)
ln (∆T1/∆T2)
= 88.54 K … Ans. (a)
(b) Counter current flow :
Air
303 K ––––––→ 383 K
Hot gas
418 K ←–––––– 483 K
∆T1 = 418 – 303 = 115 K
∆T2 = 483 – 383 = 100 K
(115 – 100)
∆Tlm = ln (115/100)

= 107.32 K (107.32oC) … Ans. (b)

Unit Operations – II 3.64 Convection

Example 3.26 : A cooling coil is incorporated in a reaction vessel of which the contents
are kept uniformly at a temperature of 360 K by means of a stirrer. The inlet and outlet
temperatures of the cooling water are 280 K and 320 K respectively. Find out the outlet
temperature of water if the length of cooling coil were increased 5 times the original
assuming that the overall heat transfer coefficient to be constant over the length of the tube
and is independent of water temperature.
Solution : The rate of heat transfer is given by
Q = U A ∆Tlm
I. For initial conditions :
∆T1 = 360 – 280 = 80 K, ∆T2 = 360 – 320 = 40 K
The rate of heat transfer is
.
Q1 = m1 Cp (T2 – T1) = U1 A1 ∆Tlm (I)
1
.
m1 × 4.187 (320 – 280) = U1 A1 [(80 – 40) ln (80/20)]
.
m1 / U1 A1 = 0.345
II. For final conditions :
. .
m2 = m1, U2 = U1 and A2 = 5 A1 (since it is given that L2 = 5 L1)
Let T be the outlet temperature of water in the final conditions.
Q2 = m2 × 4.187 (T – 280) = U2 A2 ∆Tlm (II) = U2 A2 
. (360 – 280) – (360 – T)

ln  360 – T 
360 – 280
 
   
∴ Q2 = m1 × 4.187 (T – 280) = 5 U1 A1 
. (360 – 280) – (360 – T)

ln  360 – T 
360 – 280
 
   
.
Substituting m1/U1A1 = 0.345 in the above equation yields
0.289 (T – 280) = (T – 280) / [ln 80/(360 – T)]
ln [80/(360 – T)] = 3.467
∴ T = outlet temperature of water = 357.5 K … Ans.
Example 3.27 : In an oil cooler, 60 g/s of hot oil enters a thin pipe of diameter of 25 mm.
An equal mass of cooling water flows through the annular space between pipe and large
concentric pipe. The oil and water are flowing in a counter current fashion. Calculate the
outlet temperature of water and area of heat transfer required from the following data :
Inlet temperature of oil = 420 K
Outlet temperature of oil = 320 K
Inlet temperature of water = 290 K
Oil side heat transfer coefficient = 1.6 kW/(m2·K)
Water side heat transfer coefficient = 3.6 kW/(m2·K)
Specific heat of oil = 2.0 kJ/(kg·K)
Specific heat of water = 4.18 kJ/(kg·K)
Metal wall resistance and scale resistance may be neglected.
Unit Operations – II 3.65 Convection

Solution :
.
Mass flow rate of oil = mo = 60 g/s
= 6.0 × 10–2 kg/s
Rate of heat flow = Heat removed from oil = Q
.
= mo Cp (T1 – T2)
o

= 6.0 × 10–2 × 2.0 × (420 – 320)

= 12 kJ/s
= 12 kW
Let us find the outlet temperature of water.
. .
Given : mw = mo = 6.0 × 10–2 kg/s
t1 = inlet temperature of water = 290 K, Cp = 4.18 kJ/(kg·K)
w
A heat balance yields
Heat removed from oil = Heat gained by water
.
Q = mw Cp (t2 – t1)
w

12 = 6.0 × 10–2 × 4.18 × (t2 – 290)

t2 = 338 K
Outlet temperature of water = 338 K … Ans.
Logarithmic mean temperature difference for the counter current flow is :
∆T1 = T1 – t2 = 420 – 338 = 82 K
∆T2 = T2 – t1 = 320 – 290 = 30 K
82 – 30
∆Tlm = ln (82/30) = 51.7 K

Overall heat transfer coefficient is given by

1 1 1
U = ho + hi
1 1
= 1.6 + 3.6 = 1.108 kW/(m2·K) = 1108 W/(m2·K)

Let us calculate A. The rate of heat transfer is given by

Q = UA ∆Tlm
where Q = 12 kW = 12000 W, U = 1108 W/(m2⋅K) and ∆Tlm = 51.7 K
Unit Operations – II 3.66 Convection

Q
∴ A =
U · ∆Tlm
12000
A =
1108 × 51.7
= 0.210 m2 … Ans.
Given : Do = 25 mm = 0.025 m
Let us find the length of tube, i.e., the length of heat transfer.
A = π Do L
L = A/π Do = 0.210 / (π × 0.025)
= 2.67 m … Ans.
Example 3.28 : Determine the heating surface of an economiser in which a primary
(hot gases) and a secondary fluid (water) are in counter current flow using the following
data :
Mass flow rate of gas = 1.22 kg/s
Specific heat of gas = 1.045 kJ/(kg·K)
Inlet temperature of gas = 693 K (420 oC)
Mass flow rate of water = 0.67 kg/s
Inlet temperature of water = 378 K (105 oC)
Rate of heat transfer from hot gas to water = 270 kW
Heat transfer coefficient from gases to water (i.e. U) = 80 W/(m2·K)
Solution : Let us find the outlet temperature of water.
Given : Q = 270 kW = 270 × 103 W = 270 kJ/s
.
For water : Rate of heat gained by water = Q = mw Cp (t2 – t1)
w
∴ 270 = 0.67 × 4.187 (t2 – 378)
t2 = outlet temperature of water = 474 K
Let us find the outlet temperature of the hot gases.
For hot gas : The rate of heat removal from the hot gases is
.
Q = mh Cp (T1 – T2)
h
270 = 1.22 × 1.045 (693 – T2)
T2 = outlet temperature of the hot gases = 481.2 K
Let us find ∆Tlm (LMTD) for counter current flow.
Hot gas
693 K –––––→ 481.2 K
Water
474 K ←––––– 378 K
∆T1 = 693 – 474 = 219 K, ∆T2 = 481.2 – 378 = 103.2 K
∆Tlm = (219 – 103.2) / ln (219/103.2) = 153.9 K
Unit Operations – II 3.67 Convection

The rate of heat transfer is given by

Q = U A ∆Tlm
270 × 103
∴ A = Q/(U · ∆Tlm) = = 21.93 m2
80 × 153.9
Heating surface of the economiser = 21.93 m2. … Ans.
Example 3.29 : Water is to be heated from 298 K (25oC) to 313 K (40oC) at a rate of
30 kg/s. Hot water is available at 353 K (80oC) at a rate of 24 kg/s for heating in a counter
current exchanger. Calculate the heat transfer area required, if the overall heat transfer
coefficient is 1220 W/(m2·K).
Solution : At steady state : Heat gained by cold water = Heat removed from hot water
. .
∴ Q = mc Cp (t2 – t1) = mh Cp (T1 – T2)
c h
.
mc = mass flow rate of cold water = 30 kg/s
.
mh = mass flow rate of hot water = 24 kg/s
Cp = Cp = 4.187 kJ/(kg·K)
c h
t2 = 313 K, t1 = 298 K, T1 = 353 K, T2 = ?
Q = 30 × 4.187 × (313 – 298) = 1884.15 kJ/s
= 1884.15 × 103 J/s ≡ 1884.15 × 103 W
Q = Heat gained by cold water = Heat lost by hot water
1884.15 = 24 × 4.187 (353 – T2)
T2 = 334.2 K
The rate of heat transfer is given by
Cold water
298 K ––––––––→ 313 K
Hot water
334.2 K ←–––––– 353 K
∆T1 = 353 – 313 = 40 K, ∆T2 = 334.2 – 298 = 36.2 K
∆Tlm for counter current flow = (40 – 36.2) / ln (40/36.2) = 38.1 K
The rate of heat transfer is given by
Q = U A ∆Tlm
1884.15 × 103
A = Q/U · ∆Tlm =
1220 × 38.1
= 40.53 m2
Heat transfer area required = 40.53 m2 … Ans.
Example 3.30 : Evaluate the inside heat transfer coefficient. Heavy oil flows at a rate of
0.5 kg/s through a tube of 19 mm inside diameter and is heated from 311 to 327 K by
condensing steam at 373 K.
Unit Operations – II 3.68 Convection

Data : Properties of oil at 319 K (mean/average oil temperature)

k for oil = 0.14 W/(m·K)
Cp for oil = 2.1 kJ/(kg·K)
µ for oil = 154 (mN·s)/m2
µ at mean wall temperature (346 K) = 87.0 (mN·s)/m2
Length of tube = 1.5 m
.
Solution : Mass flow rate of oil = m = 0.5 kg/s
Diameter of tube = 19 mm = 0.019 m
Mean temperature of oil = (311 + 327)/2 = 319 K
µ of oil at 319 K = 154 (mN·s)/m2
= 154 × 10–3 (N·s)/m2 [i.e. kg/(m·s)]
Cross sectional area of tube = π/4 × (0.019)2 = 2.835 × 10–4 m2
.
G = mass velocity = m /A = 0.5 / 2.835 × 10–4 = 1763.67 kg/(m2·s)
Let us calculate NRe.
Duρ DG 0.019 × 1763.67
NRe = = = = 217.6
µ µ 154 × 10–3
As NRe is less than 2100, the flow is laminar.
The Sieder-Tate equation for laminar flow is
m. C 1/3 0.14
hi Di  p µ
 k L  µw
k = 2.0

.
where m = 0.5 kg/s, L = 1.5 m, k = 0.14 W/(m·K)
µ = 154 × 10–3 kg/(m·s)
µw = viscosity at the mean wall temperature [(373 + 319)/2] i.e. at 346 K
µw = 87 (mN·s)/m2 = 87 × 10–3 (N·s)/m2 = 87 × 10–3 kg/(m·s)
Cp = 2.1 kJ/(kg·K) = 2.1 × 103 J/(kg·K)

0.5 × 2.1 × 103 154 × 10–3

1/3 0.14
hi × 0.019
= 2    
0.14  0.14 × 1.5   87 × 10–3 
∴ hi = 265.33 W/(m2·K)
Inside heat transfer coefficient = 265.33 W/(m2·K) … Ans.
Unit Operations – II 3.69 Convection

Example 3.31 : Pure iso-butane vapours are to be condensed at 331.7 K in a horizontal

tubular heat exchanger using water as a cooling medium. Determine the inside film heat
transfer coefficient based on the outside diameter of tube.
Outside diameter of tube = 19 mm
Wall thickness of tube = 1.6 mm
Inlet temperature of water = 301 K
Outlet temperature of water = 315 K
Water flow rate through each tube = 0.217 kg/s
Solution :
⋅
Water flow rate = m = 0.217 kg/s
i.d. of tube = (19.0 – 2 × 1.6) = 15.7 mm = 0.0157 m
Cross sectional area for flow = Ai = π/4 × (0.0157)2 = 0.000194 m2
.
Water velocity through tube = u = m /ρ · Ai
0.217
= = 1.12 m/s
1000 × 0.000194
Average temperature of water = T = (T1 + T2)/2 = (301 + 315)/2 = 308 K
The inside heat transfer coefficient (for water) may be calculated by
hi = 1063 (1 + 0.00293 × T) u0.8 / (di)0.20
= 1063 (1 + 0.00293 × 308) (1.12)0.8 / (0.0157)0.20
= 5084 W/(m2·K)
Inside heat transfer coefficient based on the outside diameter (hio) is calculated as
 Di 
hio = hi D  = 5084 ×  0.019 
0.0157
 o  
= 4201 W/(m ·K)2

= 4.20 kW/(m2 ·K) … Ans.

Example 3.32 : A heat exchanger is required to cool 20 kg/s of hot water from 360 K to
335 K by means of cold water flowing at 25 kg/s. Inlet temperature of cold water is 300 K.
Overall heat transfer coefficient is 2 kW/(m2·K). Calculate the area of heat transfer required
in (a) a counter current flow heat exchanger and (b) in a 1–2 pass heat exchanger (cold
water flowing twice through heat exchanger).
Data : Specific heat of cold and hot water may be taken as 4.187 kJ/(kg·K)
and LMTD correction factor = 0.94
Solution : The heat load of the exchanger is the heat to be removed from water and is
given by
.
Q = m1 Cp (T1 – T2)
1
Unit Operations – II 3.70 Convection

.
where m1 = 20 kg/s, Cp = 4.187 kJ/(kg·K)
1

T1 = 360 K, T2 = 335 K
Q = 20 × 4.187 × (360 – 335)
Q = 2093.5 kJ/s = 2093.5 kW … heat given out by hot water.
For cold water :
t1 = inlet temperature of cold water = 300 K
.
m2 = mass flow rate of cold water = 25 kg/s
t2 = outlet temperature of cold water
At steady state, heat given out by the hot water is equal to heat gained by the cold water.
.
Therefore, Q = m2 Cp (t2 – t1)
2

2093.5 = 25 × 4.187 (t2 – 300)

Solving for t2 gives
t2 = 320 K
I. Case (a) : Counter-current flow heat exchanger
Hot water
360 K ––––––→ 335 K
Cold water
320 K ←––––––– 300 K
∆T1 = T1 – t2 = 360 – 320 = 40 K
∆T2 = T2 – t1 = 335 – 300 = 35 K
(∆T1 – ∆T2)
∆Tlm for counter current flow =
ln (∆T1/∆T2)
∆Tlm = (40 – 35) / ln (40/35) = 37.4 K
Let us calculate the heat transfer area.
The rate of heat transfer is given by
Q = U A ∆Tlm
A = Q/U ∆Tlm
where Q = 2090 kW = 2090 × 103 W
U = 2.0 kW/(m2·K) = 2 × 103 W/(m2·K)
2090 × 103
∴ A =
2.0 × 103 × 37.4
= 27.94 m2 … Ans. (a)
II. Case (b) : 1-2 pass heat exchanger.
∆Tlm = 37.4 K
LMTD correction factor = FT = 0.94
The rate of heat transfer in this case is given by
Q = U A FT ∆Tlm
Unit Operations – II 3.71 Convection

Q
A =
U · FT · ∆Tlm
2090 × 103
A =
2.0 × 103 × 0.94 × 37.4
= 29.73 m2 … Ans. (b)
Example 3.33 : It is desired to heat 0.90 kg/s of air flowing through tubes from 283 to
366 K with the help of flue gas entering at 700 K and leaving at 366 K in a counter current
fashion to the air flow. Calculate the heat transfer area of a heat exchange equipment.
Data : Pipe diameter = 12 mm
Mass flow velocity = G = 19.9 kg/(m2·s)
k for air 324.5 K = 0.029 W/(m·K)
µ for air 324.5 K = 0.0198 (mN·s)/m2
Cp for air = 1.0 kJ/(kg·K)
ho = 232 W/(m2·K)
.
Solution : Given : mair = 0.90 kg/s, Di = 12 mm = 0.012 m, G = 19.9 kg/(m2·s)
Mean temperature of air = (283 + 366)/2 = 324.5 K
µ = 0.0198 (mN·s)/m2 = 1.98 × 10–5 (N·s)/m2 = 1.98 × 10–5 kg/(m·s)
Let us calculate NRe.
Di G
NRe =
µ
0.012 × 19.9
= = 1.206 × 104
1.98 × 10–5
Given : Cp = 1.0 kJ/(kg·K) = 1.0 × 103 J/(kg·K)
k = 0.029 W/(m·K)
Let us calculate NPr.
NPr = Cp µ / k
1.0 × 103 × 1.98 × 10–5
= 0.029 = 0.683

Since NRe > 10000, the flow is turbulent.

The Dittus-Boelter equation for heating is
hi Di
k = 0.023 (NRe) (NPr)
0.8 0.4

hi × 0.012
0.029 = 0.023 (1.206 × 104)0.8 (0.683)0.4
hi = 87.85 W/(m2·K)
ho = 232 W/(m2·K)
Unit Operations – II 3.72 Convection

The overall heat transfer coefficient is given by

1 1 1
U = hi + ho
1 1
= 87.85 + 232 = 0.0114 + 0.0043 = 0.0157

U = 63.7 W/(m2·K)
The rate of heat addition to air is given by
.
Q = (m × Cp)air (t2 – t1)
Q = 0.9 × 1.0 × (366 – 283) = 74.7 kJ/s
= 74.7 × 103 J/s ≡ 74.7 × 103 W
Let us calculate LMTD for counter current flow.
Air
283 K ––––––→ 366 K
Flue gas
366 K ←–––––– 700 K
∆T1 = T1 – t2 = 700 – 366 = 334 K
∆T2 = T2 – t1 = 366 – 283 = 83 K
(∆T1 – ∆T2)
∆Tlm =
ln (∆T1/∆T2)
(334 – 83)
= ln (334/83) = 180 K

Let us calculate the heat transfer area. The rate of heat transfer is given by
Q = U A ∆Tlm
A = Q/U · ∆Tlm
= 74.7 × 103 / (63.7 × 180)
= 6.52 m2 … Ans.
Example 3.34 : A light motor oil is to be heated from 338.5 K (65.5oC) to
394.1 K (121.1oC). The pipe wall is at 449.7 K (176.7oC). The light motor oil flows through
a 10 mm pipe, 5 m long at a rate of 41 kg/h. Calculate the heat transfer coefficient using the
following data :
Data : Properties of oil at the average bulk temperature are :
Cp = 2.01 kJ/(kg·K), k = 0.142 W/(m·K)
µ = 4.65 × 10–3 kg/(m·s)
Viscosity of oil at the wall temperature = µw = 1.37 × 10–3 kg/(m·s)
Unit Operations – II 3.73 Convection

⋅
Solution : Mass flow rate of oil = m = 41 kg/h = 0.0114 kg/s
Di = 10 mm = 0.01 m
2
Area of pipe = A = π/4 Di = π/4 (0.01)2 = 7.85 × 10–5 m2
⋅
m 0.0114
G = mass velocity of oil = a = = 145.15 kg/(m2·s)
7.85 × 10–6
Let us calculate NRe.
DiG 0.01 × 145.15
NRe = =
µ 4.65 × 10–3
= 312.15
As NRe is less than 2100, the flow is laminar.
The Sieder-Tate equation for laminar flow is
m. C 1/3
hD  p  µ 0.14
 kL  µw
k = 2

.
where k = 0.142 W/(m·K), m = 0.0114 kg/s
L = 5 m, Cp = 2.01 × 103 J/(kg·K)
µ = 4.65 × 10–3 kg/(m·s), µw = 1.37 × 10–3 kg/(m·s)
D = 0.01 m
Substituting the values yields
1/3 0.14
h × 0.01 0.0114 × 2.01 × 103 4.65 × 10–3
= 2    
0.142  0.142 × 5  1.37 × 10–3
h × 0.0704 = 7.56
h = 107.4 W/(m2·K)
Film heat transfer coefficient = 107.4 W/(m2·K) … Ans.
Example 3.35 : Ethylene glycol at 273 K enters a 40 mm diameter pipe at a velocity of
5 m/s. The tube wall is maintained at a temperature of 353 K. Calculate the heat transfer
coefficient using Sieder-Tate equation.
Data : Exit temperature is 313 K.
Properties of ethylene glycol at 293 K are as follows :
ρ = 1117 kg/m3, Cp = 2.39 kJ/(kg·K), ν = 19.18 × 10–6 m2/s
k = 0.249 W/(m·K)
µw = viscosity at 353 K = 3.21 × 10–3 kg/(m·s)
Unit Operations – II 3.74 Convection

Solution : Given : u = velocity of glycol through pipe = 5 m/s

Di = 40 mm = 0.04 m

ρ = 1117 kg/m3
ν = 19.18 × 10–6 m2/s = µ/ρ
Let us calculate NRe.
Duρ Du 0.04 × 5
NRe = = = = 10427.5
µ µ/ρ 19.18 × 10–6

Since NRe > 10000, the flow is turbulent.

µ = Viscosity at average temperature of 293 K
= ν × ρ = 19.18 × 10–6 × 1117 = 0.0214 kg/(m·s)
273 + 313
Average temperature of glycol = 2 = 293 K

At 293 K, we have :
µ = 0.0214 kg/(m·s)
k = 0.249 W/(m·K)
Cp = 2.39 kJ/(kg.K) = 2.39 × 10–3 J/(kg·K)
Cp µ 2.39 × 10–3 × 0.0214
NPr = k = 0.249 = 205.4

The Sieder-Tate equation is

NNu = 0.023 (NRe)0.8 (NPr)1/3 (µ/µw)0.14

NNu = 0.023 (10427.5)0.8 (205.4)1/3 

0.0214 0.14

3.21 × 10–3
= 290.06
hi Di
k = 290.06

where hi is the heat transfer coefficient.

k 0.249
hi = 290.06 × D = 290.06 × 0.04
i

= 1805.6 W/(m2·K) … Ans.

Unit Operations – II 3.75 Convection

Example 3.36 : A vertical plate, 30 by 30 cm, is exposed to steam at the atmospheric

pressure. The plate is at 371 K (98oC). Calculate the mean heat transfer coefficient, the heat
transfer rate and mass of steam condensed per hour.
Data : Properties of the condensate at the film temperature are :
ρ = 960 kg/m3, µ = 2.82 × 10–4 kg/(m·s), k = 0.68 W/(m·K)
λ = 2225 kJ/kg, Saturation temperature of steam = 373 K (100o C).
Assume that the condensate film is laminar.
Solution : ρ = 960 kg/m3, λ = 2225 × 103 J/kg, L = 0.3 m
∆T = 373 – 371 = 2 K
µ = 2.82 × 10–4 kg/(m·s), k = 0.68 W/(m·K)
g = 9.81 m/s2
The film coefficient for condensation on a vertical plate is given by

 ρ2 g λ k3 
1/4
hm = 0.943  
L µ (∆T)
(960)2 × 9.81 × 2225 × 103 × (0.68)31/4
= 0.943  
 (0.3) (2.82 × 10–4) (2) 
= 13150 W/(m2·K) … Ans.
The heat transfer rate is
Q = hm A (Tsat – Tw)

= 13150 × 0.3 × 0.3 (373 – 371) = 2367 W … Ans.

⋅
We have : Q = mλ
Q 2367
Mass flow of condensate = =
λ 2225 × 103

= 1.05 × 10–3 kg/s

= 3.78 kg/h … Ans.

Example 3.37 : Dry steam at 373 K (100oC) condenses on the outside surface of a
horizontal pipe of 25 mm O.D. The pipe surface is maintained at 357 K (84oC) by circulating
water through it. Determine the mean heat transfer coefficient, heat transfer per unit length
of pipe and condensate rate per unit length of pipe.
Unit Operations – II 3.76 Convection

Data : The properties of condensate at the film temperature of 350 K are :

µ = 306 × 10–6 N·s/m2, k = 0.668 W/(m·K), ρ = 974 kg/m3, λ = 2225 kJ/kg
Solution : hm = mean heat transfer coefficient

ρ = 974 kg/m3, λ = 2225 × 103 J/kg, k = 0.668 W/(m·K)

µ = 306 × 10–6 N·s/m2 , g = 9.81 m/s2
The film coefficient for condensation on a horizontal tube is given by

ρ2 g λ k3 1/4
hm = 0.725  
µ do ∆Tf
(974)2 × 9.81 × 2225 × 103 × (0.668)31/4
hm = 0.725  
 306 × 10–6 × 0.025 × 16 
= 10864 W/(m2·K) … Ans.
The heat transfer per unit length is
Q/L = hm πdo (Tsat – Ts)

= 10864 × π × 0.025 (373 – 357) = 13652 W/m … Ans.

The mass flow rate of the condensate is
⋅
m = Q/λ = 13652/2225 × 103
= 6.13 × 10–3 kg/s = 22.1 kg/h … Ans.

EXERCISES

1 . Fill in the blanks :

(a) Heat transfer in fluids occurs by mechanism known as ...............
Ans. convection
(b) Heat transfer from hot fluid to cold fluid in a heat exchange equipment takes
place by conduction and .................. modes. Ans. convection
(c) The unit of overall heat transfer coefficient in the SI system is ……….
Ans. W/(m2·K)
(d) When the hot and cold fluids flow in the same direction in a heat exchanger, then
the flow is called ........... flow. Ans. parallel / co-current
(e) When the hot and cold fluids flow in the opposite directions in a heat exchanger,
then the flow is called as ................ flow. Ans. counter current
Unit Operations – II 3.77 Convection

(f) Film heat transfer coefficients are higher for .......... condensation than for
filmwise condensation. Ans. dropwise
(g) In case of ........ boiling, vaporisation takes place directly from the surface.
Ans. nucleate
(h) The SI unit of film heat transfer coefficient is .......... Ans. W/(m2·K)
(i) The dirt factor / fouling factor has the units ........ in SI system. Ans. (m2·K)/W
(j) The ratio of Cp · µ to k is known as ....... number. Ans. Nusselt
(k) Thermal resistance in case of convection heat transfer has the units of ..........
Ans. (m2·K)/W
(l) In forced convection, the effect of liquid viscosity (for viscous liquids) is taken
into account in case of .......... equation. Ans. Sieder-Tate
2. Give the Dittus-Boelter equation for turbulent flow for heating.
3. Give the Dittus-Boelter equation for turbulent flow for cooling.
4. Give the Sieder-Tate equation used to calculate the film coefficient in case of
laminar flow.
5. Give the Sieder-Tate equation used to calculate the film coefficient in case of
turbulent flow.
6. What do you mean by film heat transfer coefficient ?
7. For a double pipe heat exchanger, give its equivalent diameter in terms of D1 and
D2, where D1 is the outside diameter of the inside pipe and D2 is the inside
diameter of the outside pipe.
8. What do you mean by natural convection ? Give example of heat transfer by
natural convection.
9. Define forced convection and give its example.
10. Compare natural convection with forced convection (Three points).
11. Compare parallel flow with counter current flow in heat exchangers (Three
points).
12. Define drop-wise condensation.
13. Compare drop-wise condensation with film-wise condensation (Three points).
14. Define co-current and counter current flow with neat sketches.
15. State the relationship between overall heat transfer coefficient and individual heat
transfer coefficients.
16. Write the relationship between U and hi, ho, xw/k and Rd.
17. Explain in brief heat transfer to boiling liquids.
18. Derive the relationship Q = UA ∆Tlm.