CONVECTION

Course Contents
7.1 Introduction to Convection
7.2 Newton-Rikhman Law
7.3 Free and Forced Convection
7.4 Dimensional Analysis
7.5 Dimensionless Numbers &
Their Physical Significance
7.6 Dimensional Analysis Applied
to Forced Convection
7.7 Dimensional Analysis Applied
to Free Convection
7.8 Empirical Co-relations for Free
& Forced Convection
7.9 Thermal and Hydrodynamic
Boundary Layer
7.10 Derivation of Differential
Convection Equations
A. Continuity Equation
B. Momentum Equation
C. Energy Equation
7.11 Von-Karman Integral
Momentum Equation
7.12 Solution for Velocity Boundary
Layer
7.13 Solved Numerical
7.14 References
7. Convection Heat Transfer (2151909)

7.1 Introduction to Convection

 Thermal convection occurs when a temperature difference exists between a solid
surface and a fluid flowing past it.

𝒖∞ Free Stream
𝑻∞
𝒖 𝒒
𝑻𝒘

Heated Wall

Fig. 7.1 Convection Phenomena

 It is well known that a hot plate of metal will cool faster when placed in front of a fan
than when exposed to still air.

 For example,

We know that the velocity at which the air blows over the hot plate obviously
influences the heat transfer rate. But does it influence the cooling in a linear way?
i.e. if the velocity is doubled, will the heat transfer rate doubled?

Relation with conduction:

 As shown in Fig. 7.1 the velocity of fluid layer at the wall will be zero, the heat must
be transferred by conduction at that point.

 Thus we might compute the heat transfer using Fourier’s equation of conduction i.e.
𝜕𝑡
𝑞 = −𝐾𝐴 𝜕𝑥 with the thermal conductivity of fluid and the fluid temperature
gradient at wall.

 Why then, if the heat flows by conduction in this layer, do we speak of “Convection”
heat transfer and need to consider the velocity of the fluid?

 The answer is that the temperature gradient is dependent on the rate at which the
fluid carries the heat away; a high velocity produces a large temperature gradient,
and so on.

 It must be remembered that the physical mechanism of heat transfer at the wall is a
conduction process.
Heat Transfer (2151909) 7. Convection

7.2 Newton-Rikhman Law OR Newton’s Law of Cooling OR

Convection Rate Equation
 The appropriate convection rate equation for the convective heat transfer between
a surface and an adjacent fluid is given by Newton’s law of cooling:
𝑄 = ℎ𝐴(𝑡𝑠 − 𝑡∞ ) − − − − − − − −(7.1)
Where,
𝑄 = Convective heat flow rate
𝐴 = Surface area exposed to heat transfer
𝑡𝑠 = Surface temperature of solid and
𝑡∞ = Temperature of the fluid (Stagnant or Undisturbed)
ℎ = The Convective heat transfer co-efficient or The film co-efficient or The surface
conductance
 The heat transfer co-efficient is sometimes called the film conductance or surface
conductance because of its relation to the conduction process in the thin stationary
layer of fluid at the wall surface.

 Unit of Convective heat transfer co-efficient: 𝑊⁄𝑚2 𝐾 or 𝐶𝑎𝑙⁄𝑚2 ℎ𝑟𝐾 or 𝑊⁄𝑚2 °𝐶

 The value of film co-efficient is dependent upon:

1. Surface conditions: Roughness & Cleanliness
2. Geometry and orientation of surface: Plate, Tube and Cylinder placed
horizontally or vertically.
3. Thermo-physical properties of the fluid: Density, Viscosity, Specific heat, Co-
efficient of expansion and thermal conductivity.
4. Nature of fluid flow: Laminar or Turbulent
5. Boundary layer configuration
6. Existing thermal conditions.

The film co-efficient (𝒉) depends on viscosity of fluid because………

The viscosity influences the velocity profile and correspondingly the energy
transfer rate in the region near the wall.
7. Convection Heat Transfer (2151909)

7.3 Free and Forced Convection

 With respect to the cause of fluid flow, two types of convection are distinguished:
1. Free Convection or Natural Convection and
2. Forced Convection.
1. Free Convection or Natural Convection
 When a surface is maintained in still fluid at a temperature higher or lower than that
of the fluid, a layer of fluid adjacent to the hot or cold surface gets heated or cooled
by conduction.
 A density difference is created between this adjacent layer and the still fluid
surrounding it.
 The density difference introduces a buoyant force causing flow of fluid near the
surface.
 Heat transfer under such conditions is known as Free or Natural Convection.
 Thus, “Free or Natural convection is the process of heat transfer which occurs due
to movement of the fluid particles by density changes associated with temperature
differential in a fluid.”
 This mode of heat transfer occurs very commonly, some of the examples are:
I. House heating system
II. The cooling of transmission lines, electric transformers and rectifiers.
2. Forced Convection
 Flow of fluid is caused by a pump, a fan or by the atmospheric winds.
 These mechanical devices speeds up the heat transfer rate.
 In free convection flow velocities encountered are lower compared to flow velocities
in forced convection, consequently the value of convection co-efficient is lower, and
for a given rate of heat transfer larger area could be required.
 Examples of forced convection are: cooling of I.C. Engines, Air conditioner, Heat
exchangers, etc.
 The rate of heat transfer is calculated using the equation 7.1.
Table 7.1 Typical values of convective co-efficient
Sr. No. Free Convection Forced convection
Air & Super heated steam – 30 to
1 Air – 3 to 7 W/m2K
300 W/m2K
2 Gases – 2 to 20 W/m2K Oil – 60 to 3000 W/m2K
3 Liquids – 30 to 300 W/m2K Water – 3000 to 10000 W/m2K
Heat Transfer (2151909) 7. Convection

7.4 Dimensional Analysis

 “Dimensional analysis is a mathematical technique which makes use of the study of
the dimensions for solving several engineering problems.”
 Dimensional analysis has become an important tool for analyzing fluid flow
problems. It is especially useful in presenting experimental results in a concise form.
 There are two methods are used in dimensional analysis: 1) Rayleigh’s Method and
2) Buckingham’s 𝜋-Theorem.
Buckingham’s 𝝅-Theorem
“If there are 𝑛 variables (independent and dependent variables) in a physical
phenomenon and if these variables contain 𝑚 fundamental dimensions, then the
variables are arranged into (𝑛 − 𝑚) dimensionless terms; each terms are called 𝜋-
terms.”

System of Dimensions:
 In the area of heat transfer, two more dimensions namely the temperature
difference (𝜃) and the heat (𝐻) are also taken as fundamental quantities.
 Here heat (𝐻) can be expressed in terms of MLT. So the fundamental quantities are
mass, length, time and temperature; designated by the M,L,T,θ respectively.
 Temperature is specially used in compressible flow and heat transfer phenomena.

Table 7.2 Quantities used in fluid mechanics and heat transfer & their dimensions

Sr. Dimensions Dimensions

Quantity Symbol Units (SI)
No. (MLTθ System) (MLTθH System)
A Fundamental
1 Mass M Kg M1L0T0θ0 M1L0T0θ0H0
2 Length L m M0L1T0θ0 M0L1T0θ0H0
3 Time T Sec M0L0T1θ0 M0L0T1θ0H0
4 Temperature θ K M0L0T0θ1 M0L0T0θ1H0
5 Heat Q, H Joule M1L2T-2 M0L0T0θ0H1
B Geometric
1 Area A m2 L2 L2
2 Volume V m3 L3 L3
C Kinematic
1 Linear Velocity u, v m/s L1T-1 L1T-1
7. Convection Heat Transfer (2151909)

2 Angular Velocity ω rad/s T-1 T-1

3 Acceleration a m/s2 L1T-2 L1T-2
4 Angular α rad/s2 T-2 T-2
Acceleration
5 Discharge Q m3/sec L3T-1 L3T-1
6 Kinematic Viscosity ν m2/sec L2T-1 L2T-1
D Dynamic
1 Force / Resistance F/R N (kg-m/s2) M1L1T-2 M1L1T-2
2 Density ρ Kg/ m3 M1L-3 M1L-3
3 Specific Weight w N/ m3 M1L-2T-2 M1L-2T-2
4 Dynamic Viscosity μ Kg/m-sec M1L-1T-1 M1L-1T-1
5 Work, Energy W, E N-m (Joule) M1L2T-2 H1
6 Power P Watt (J/sec) M1L2T-3 T-1H1
E Thermodynamic
1 Thermal K W/m-K M1L1T-3θ-1 L-1T-1θ-1H1
Conductivity
2 Specific Heat Cp, Cv kJ/kg-K L2T-2θ-1 M-1θ-1H1
3 Heat Transfer Co- h W/m2-K M1T-3θ-1 L-2T-1θ-1H1
efficient
4 Gas Constant R J/kg-K L2T-2θ-1 M-1θ-1H1
5 Thermal Diffusivity α m2/sec L2T-1 L2T-1

7.5 Dimensionless Numbers & Their Physical Significance

1. Reynolds Number (Re)
 It is defined as a ratio of inertia force to viscous force.
𝑣 𝑣
𝐼𝑛𝑒𝑟𝑡𝑖𝑎 𝐹𝑜𝑟𝑐𝑒 = 𝑚𝑎 = 𝜌𝑉 × = 𝜌𝐴𝐿 × = 𝜌𝐴𝑣 2
𝑡 𝑡
𝑑𝑢
𝑉𝑖𝑠𝑐𝑜𝑢𝑠 𝐹𝑜𝑟𝑐𝑒 = 𝜏𝐴 = 𝜇 𝐴 = 𝜇𝑣𝐿
𝑑𝑦
𝐼𝑛𝑒𝑟𝑡𝑖𝑎 𝐹𝑜𝑟𝑐𝑒 𝜌𝐴𝑣 2 𝜌𝑣𝐿 𝑣𝐿
𝑅𝑒 = = = = − − − − − − − −(7.2)
𝑉𝑖𝑠𝑐𝑜𝑢𝑠 𝐹𝑜𝑟𝑐𝑒 𝜇𝑣𝐿 𝜇 𝜈
 It indicates the relative importance of the inertial and viscous effects in a fluid
motion.
Heat Transfer (2151909) 7. Convection

 At low Reynolds number, the viscous effect dominates and the fluid motion is
laminar.
 At high Reynolds number, the inertial effects lead to turbulent flow.
 Reynolds number constitutes an important criterion of kinematic and dynamic
similarity in forced convection heat transfer.

2. Prandtl Number (Pr)

“It is the ratio of kinematic viscosity to thermal diffusivity of the fluid”.
𝜇𝐶𝑝 𝜌𝜈𝐶𝑝 𝜈
𝑃𝑟 = = =
𝐾 𝐾 (𝐾⁄𝜌𝐶 )
𝑝

𝜈 𝐾𝑖𝑛𝑒𝑚𝑎𝑡𝑖𝑐 𝑉𝑖𝑠𝑐𝑜𝑠𝑖𝑡𝑦
𝑃𝑟 = = − − − − − − − −(7.3)
𝛼 𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝐷𝑖𝑓𝑓𝑢𝑠𝑖𝑣𝑖𝑡𝑦
 The kinematic viscosity represents the momentum transport by molecular friction
and thermal diffusivity represents the heat energy transport through conduction.
 Pr provides a measure of the relative effectiveness of momentum and energy
transport by diffusion.
 For highly viscous oils, Pr is quite large (100 to 10000) and that indicates rapid
diffusion of momentum by viscous action compared to the diffusion of energy.
 For gases, Pr is about 1, which indicates that both momentum and heat dissipate
through the field at about the same rate.
 The liquid metals (liquid sodium or liquid potassium) have Pr = 0.003 to 0.01 and
that indicates more rapid diffusion of energy compared to the momentum diffusion
rate.
 The Prandtl number is connecting link between the velocity field and the
temperature field, and its value strongly influences relative growth of velocity and
thermal boundary layers.
 Mathematically,
𝛿
≅ (𝑃𝑟)𝑛 − − − − − − − −(7.4)
𝛿𝑡
Where,
𝛿 = Thickness of velocity boundary layer
𝛿𝑡 = Thickness of thermal boundary layer
For,
Oil − 𝛿𝑡 << 𝛿 Gases − 𝛿𝑡 ≅ 𝛿 Liquid Metals − 𝛿𝑡 >> 𝛿
7. Convection Heat Transfer (2151909)

3. Nusselt Number (Nu)

 Nu established the relation between convective film co-efficient (ℎ), thermal
conductivity of the fluid (𝐾) and a significant length parameter (𝑙) of the physical
system.
ℎ𝑙
𝑁𝑢 = − − − − − − − −(7.5)
𝐾
 To understand the physical significance of the Nu, consider a fluid layer of thickness 𝑙
and temperature difference ∆𝑇 = 𝑇2 − 𝑇1 as shown in Fig. 7.2.

𝑻𝟐

Fluid 𝒒̇
𝒍 ∆𝑻 = 𝑻𝟐 − 𝑻𝟏
Layer

𝑻𝟏
Fig. 7.2 Heat transfer through the fluid layer
 Heat transfer through the fluid layer is by convection when the fluid involves some
motion and by conduction when the fluid layer is motionless.
 Heat flux (The rate of heat transfer per unit surface area) in either case is,
∆𝑇
𝑞̇ 𝑐𝑜𝑛𝑣 = ℎ∆𝑇 𝑎𝑛𝑑 𝑞̇ 𝑐𝑜𝑛𝑑 = 𝐾
𝑙
Taking their ratios,
𝑞̇ 𝑐𝑜𝑛𝑣 ℎ∆𝑇 ℎ𝑙
= = = 𝑁𝑢
𝑞̇ 𝑐𝑜𝑛𝑑 𝐾 ∆𝑇 𝐾
𝑙

 The Nusselt number is a convenient measure of the convective heat transfer co-
efficient.
 The larger the Nusselt number, the more effective the convection.
 The 𝑁𝑢 = 1 for a fluid layer represents heat transfer across the layer by pure
conduction.
 For a given Nu, h is directly proportional to thermal conductivity of the fluid and
inversely proportional to the significant length parameter.

4. Grashoff Number (Gr)

 It indicates the relative strength of the buoyant to viscous forces.
𝛽𝑔∆𝑇𝜌2 𝑙 3
𝐺𝑟 = − − − − − − − −(7.6)
𝜇2
Heat Transfer (2151909) 7. Convection

𝜌
∴ 𝐺𝑟 = (𝜌𝑙 3 𝛽𝑔∆𝑇) ×
𝜇2
𝜌𝑣 2 𝑙 2
∴ 𝐺𝑟 = (𝜌𝑙 3 𝛽𝑔∆𝑇) ×
(𝜇𝑣𝑙)2
𝐼𝑛𝑒𝑟𝑡𝑖𝑎 𝐹𝑜𝑟𝑐𝑒
∴ 𝐺𝑟 = 𝐵𝑢𝑜𝑦𝑎𝑛𝑡 𝐹𝑜𝑟𝑐𝑒 ×
(𝑉𝑖𝑠𝑐𝑜𝑢𝑠 𝐹𝑜𝑟𝑐𝑒)2
 Obviously the Grashoff number represents the ratio of Buoyant force and Inertia
force to the square of the Viscous force.
 Grashoff number has a role in free convection.
 Free convection is usually suppressed at sufficiently small Gr, begins at some
critical value of Gr and then becomes more and more effective as Gr increases.

5. Stanton Number (St)

 “It is the ratio of heat transfer co-efficient to the flow of heat per unit temperature
rise due to the velocity of fluid”.
ℎ
𝑆𝑡 =
𝜌𝑣𝐶𝑝
ℎ𝑙
(𝐾 )
= 𝜌𝑣𝑙 𝜇𝐶𝑝
( )( )
𝜇 𝐾

𝑁𝑢
∴ 𝑆𝑡 = − − − − − − − −(7.7)
𝑅𝑒 × 𝑃𝑟
 It should be noted that Stanton number can be used only in co-relating forced
convection data (since the expression contains velocity, 𝑣).

6. Peclet Number (Pe)

 “It is the ratio of mass heat flow rate by convection to the flow rate by conduction
under an unit temperature gradient and through a thickness 𝑙”.
𝑄𝑐𝑜𝑛𝑣 (𝜌𝐴𝑣)𝐶𝑝 ∆𝑇 𝜌𝐶𝑝
𝑃𝑒 = = = × 𝑙𝑣
𝑄𝑐𝑜𝑛𝑑 𝐾𝐴∆𝑇⁄ 𝐾
𝑙
𝑙𝑣
∴ 𝑃𝑒 = − − − − − − − −(7.8)
𝛼
𝑶𝑹
𝜌𝐶𝑝 𝜌𝑣𝑙 𝜇𝐶𝑝
𝑃𝑒 = × 𝑙𝑣 = ×
𝐾 𝜇 𝐾
∴ 𝑃𝑒 = 𝑅𝑒 × 𝑃𝑟 − − − − − − − −(7.9)
 The Peclet number is a function of Reynolds number and Prandtl number.
7. Convection Heat Transfer (2151909)

7. Graetz Number (G)

 “It is the ratio of heat capacity of fluid flowing through the pipe per unit length to the
conductivity of pipe material.”
𝑚𝐶𝑝
⁄
𝐺= 𝑙 = 𝑚𝐶𝑝
𝐾 𝐾𝑙
(𝜌𝐴𝑣)𝐶𝑝
∴𝐺=
𝐾𝑙
𝜋
(𝜌 4 𝑑 2 𝑣) 𝐶𝑝
∴𝐺=
𝐾𝑙
𝜋 𝜌𝑣𝑑 𝜇𝐶𝑝 𝑑
∴𝐺= × × ×
4 𝜇 𝐾 𝑙
𝜋 𝑑
∴𝐺= × (𝑅𝑒 × 𝑃𝑟) × − − − − − − − −(7.10)
4 𝑙
Where,
𝑑 and 𝑙 are the diameter and length of pipe respectively.
𝑶𝑹
𝑚𝐶𝑝 (𝜌𝐴𝑣)𝐶𝑝 𝐴𝑣 𝜋 2 𝑣
𝐺= = = = 𝑑
𝐾𝑙 𝐾𝑙 𝛼𝑙 4 𝛼𝑙
𝑣𝑑 𝜋𝑑
∴𝐺= ( )
𝛼 4𝑙
𝜋𝑑
∴ 𝐺 = 𝑃𝑒 × ( ) − − − − − − − −(7.11)
4𝑙
 Graetz number is merely a product of a constant and the Peclet number.

7.6 Dimensional Analysis Applied to Forced Convection

 Let us now consider the case of a fluid flowing across a horizontal heated tube.
 The heat transfer co-efficient is a function of the following variables:
ℎ = 𝑓(𝐷, 𝑣, 𝜌, 𝜇, 𝐶𝑝 , 𝐾) − − − − − − − −(7.12)

∴ 𝑓1 (ℎ, 𝐷, 𝑣, 𝜌, 𝜇, 𝐶𝑝 , 𝐾) = 0 − − − − − − − −(7.13)
𝑊
ℎ = Heat transfer co-efficient = 𝑚2 𝐾 = 𝑀1 𝑇 −3 𝜃 −1

𝐷 = Tube diameter = 𝑚 = 𝐿1
𝑚
𝑣 = Fluid velocity = 𝑠𝑒𝑐 = 𝐿1 𝑇 −1
𝑘𝑔
𝜌 = Fluid density = 𝑚3 = 𝑀1 𝐿−3
Heat Transfer (2151909) 7. Convection

𝑁−𝑠𝑒𝑐
𝜇 = Fluid viscosity = = 𝑀1 𝐿−1 𝑇 −1
𝑚2
𝐽
𝐶𝑝 = Specific heat = 𝑘𝑔−𝐾 = 𝐿2 𝑇 −2 𝜃 −1
𝑊
𝐾 = Thermal conductivity = 𝑚−𝐾 = 𝑀1 𝐿1 𝑇 −3 𝜃 −1

 Total number of variables, 𝑛 = 7

Number of fundamental dimensions, 𝑚 = 4 (i.e. M, L, T, θ)
Total number of 𝜋 − 𝑡𝑒𝑟𝑚𝑠 = 𝑛 − 𝑚 = 7 − 4 = 𝟑
 Hence equation 7.13 may be written as,
𝑓1 (𝜋1 , 𝜋2 , 𝜋3 ) = 0 − − − − − − − −(7.14)
 Selecting 𝑫, 𝝆, 𝝁, 𝑲 as a repeating variables.
∴ 𝜋1 = 𝐷𝑎1 𝜌𝑏1 𝜇 𝑐1 𝐾 𝑑1 𝑣
∴ 𝜋2 = 𝐷𝑎2 𝜌𝑏2 𝜇 𝑐2 𝐾 𝑑2 𝐶𝑝

∴ 𝜋3 = 𝐷𝑎3 𝜌𝑏3 𝜇 𝑐3 𝐾 𝑑3 ℎ
𝝅𝟏 − 𝑻𝒆𝒓𝒎:
𝜋1 = 𝐷𝑎1 𝜌𝑏1 𝜇 𝑐1 𝐾 𝑑1 𝑣
∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎1 (𝑀1 𝐿−3 )𝑏1 (𝑀1 𝐿−1 𝑇 −1 )𝑐1 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑1 (𝐿1 𝑇 −1 )
𝑴: − 0 = 𝑏1 + 𝑐1 + 𝑑1
𝑳: − 0 = 𝑎1 − 3𝑏1 − 𝑐1 + 𝑑1 + 1
𝑻: − 0 = −𝑐1 − 3𝑑1 − 1
𝜽: − 0 = −𝑑1
By solving above equations, we get,
𝒂𝟏 = 𝟏; 𝒃𝟏 = 𝟏; 𝒄𝟏 = −𝟏; 𝒅𝟏 = 𝟎
∴ 𝜋1 = 𝐷1 𝜌1 𝜇 −1 𝐾 0 𝑣
𝝆𝒗𝑫
∴ 𝝅𝟏 = = 𝑹𝒆 (𝑹𝒆𝒚𝒏𝒐𝒍𝒅𝒔 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟏𝟓)
𝝁
𝝅𝟐 − 𝑻𝒆𝒓𝒎:
𝜋2 = 𝐷 𝑎2 𝜌𝑏2 𝜇 𝑐2 𝐾 𝑑2 𝐶𝑝

∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎2 (𝑀1 𝐿−3 )𝑏2 (𝑀1 𝐿−1 𝑇 −1 )𝑐2 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑2 (𝐿2 𝑇 −2 𝜃 −1 )
𝑴: − 0 = 𝑏2 + 𝑐2 + 𝑑2
𝑳: − 0 = 𝑎2 − 3𝑏2 − 𝑐2 + 𝑑2 + 2
𝑻: − 0 = −𝑐2 − 3𝑑2 − 2
7. Convection Heat Transfer (2151909)

𝜽: − 0 = −𝑑2 − 1
By solving above equations, we get,
𝒂𝟐 = 𝟎; 𝒃𝟐 = 𝟎; 𝒄𝟐 = 𝟏; 𝒅𝟐 = −𝟏
∴ 𝜋2 = 𝐷0 𝜌0 𝜇1 𝐾 −1 𝐶𝑝
𝝁𝑪𝒑
∴ 𝝅𝟐 = = 𝑷𝒓 (𝑷𝒓𝒂𝒏𝒅𝒕𝒍 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟏𝟔)
𝑲
𝝅𝟑 − 𝑻𝒆𝒓𝒎:
∴ 𝜋3 = 𝐷𝑎3 𝜌𝑏3 𝜇 𝑐3 𝐾 𝑑3 ℎ
∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎3 (𝑀1 𝐿−3 )𝑏3 (𝑀1 𝐿−1 𝑇 −1 )𝑐3 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑3 (𝑀1 𝑇 −3 𝜃 −1 )
𝑴: − 0 = 𝑏3 + 𝑐3 + 𝑑3 + 1
𝑳: − 0 = 𝑎3 − 3𝑏3 − 𝑐3 + 𝑑3
𝑻: − 0 = −𝑐3 − 3𝑑3 − 3
𝜽: − 0 = −𝑑3 − 1
By solving above equations, we get,
𝒂𝟑 = 𝟏; 𝒃𝟑 = 𝟎; 𝒄𝟑 = 𝟎; 𝒅𝟑 = −𝟏
∴ 𝜋3 = 𝐷1 𝜌0 𝜇 0 𝐾 −1 ℎ
𝒉𝑫
∴ 𝝅𝟑 = = 𝑵𝒖 (𝑵𝒖𝒔𝒔𝒆𝒍𝒕 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟏𝟕)
𝑲
Put the values of π1 , π2 and π3 in equation 7.14, we get,
𝑓1 (𝜋1 , 𝜋2 , 𝜋3 ) = 0
𝜌𝑣𝐷 𝜇𝐶𝑝 ℎ𝐷
∴ 𝑓1 ( , , )=0
𝜇 𝐾 𝐾
ℎ𝐷 𝜌𝑣𝐷 𝜇𝐶𝑝
∴ = 𝜑( , )
𝐾 𝜇 𝐾
∴ 𝑵𝒖 = 𝝋(𝑹𝒆, 𝑷𝒓) − − − − − − − −(𝟕. 𝟏𝟖)
 Hence Nusselt number is a function of Reynolds number and Prandtl number for
forced convection.

7.7 Dimensional Analysis Applied to Free Convection

 Let us now consider the case of natural convection from a horizontal heated tube to
an adjacent fluid.
 The free convection heat transfer co-efficient (ℎ) depends upon the variables;
𝑣, 𝜌, 𝐾, 𝜇, 𝐶𝑝 𝑎𝑛𝑑 𝐿 𝑜𝑟 𝐷.
Heat Transfer (2151909) 7. Convection

 Since the fluid circulation in free convection is due to the difference in density
between the various fluid layers due to temperature gradient and not by external
agency. Therefore, velocity (𝒗) is no longer an independent variable but depends
upon the following factors:
(i) 𝛽 (The co-efficient of thermal expansion of the fluid)
(ii) 𝑔 (Acceleration due to gravity)
(iii) ∆𝑡 (The difference of temperature between the heated surface and the
undisturbed fluid)
 Thus, heat transfer co-efficient (ℎ) can be expressed as follows:
ℎ = 𝑓(𝐷, 𝛽𝑔∆𝑡, 𝜌, 𝜇, 𝐶𝑝 , 𝐾, ) − − − − − − − −(7.19)

𝑓1 (ℎ, 𝐷, 𝛽𝑔∆𝑡, 𝜌, 𝜇, 𝐶𝑝 , 𝐾, ) = 0 − − − − − − − −(7.20)

𝑊
ℎ = Heat transfer co-efficient = 𝑚2 𝐾 = 𝑀1 𝑇 −3 𝜃 −1

𝐷 = Pipe Diameter = 𝑚 = 𝐿1
𝑚
𝛽𝑔∆𝑡 = Buoyant force = 𝑠𝑒𝑐 2 = 𝐿1 𝑇 −2
𝑘𝑔
𝜌 = Fluid density = 𝑚3 = 𝑀1 𝐿−3
𝑁−𝑠𝑒𝑐
𝜇 = Fluid viscosity = = 𝑀1 𝐿−1 𝑇 −1
𝑚2
𝐽
𝐶𝑝 = Specific heat = 𝑘𝑔−𝐾 = 𝐿2 𝑇 −2 𝜃 −1
𝑊
𝐾 = Thermal conductivity = 𝑚−𝐾 = 𝑀1 𝐿1 𝑇 −3 𝜃 −1

 Total number of variables, 𝑛 = 7

Number of fundamental dimensions, 𝑚 = 4 (i.e. M, L, T, θ)
Total number of 𝜋 − 𝑡𝑒𝑟𝑚𝑠 = 𝑛 − 𝑚 = 7 − 4 = 𝟑
 Hence equation 7.20 may be written as,
𝑓1 (𝜋1 , 𝜋2 , 𝜋3 ) = 0 − − − − − − − −(7.21)
 Selecting 𝑫, 𝝆, 𝝁, 𝑲 as a repeating variables.
∴ 𝜋1 = 𝐷𝑎1 𝜌𝑏1 𝜇 𝑐1 𝐾 𝑑1 𝛽𝑔∆𝑡
∴ 𝜋2 = 𝐷𝑎2 𝜌𝑏2 𝜇 𝑐2 𝐾 𝑑2 𝐶𝑝

∴ 𝜋3 = 𝐷𝑎3 𝜌𝑏3 𝜇 𝑐3 𝐾 𝑑3 ℎ
𝝅𝟏 − 𝑻𝒆𝒓𝒎:
𝜋1 = 𝐷 𝑎1 𝜌𝑏1 𝜇 𝑐1 𝐾 𝑑1 𝛽𝑔∆𝑡
∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎1 (𝑀1 𝐿−3 )𝑏1 (𝑀1 𝐿−1 𝑇 −1 )𝑐1 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑1 (𝐿1 𝑇 −2 )
7. Convection Heat Transfer (2151909)

𝑴: − 0 = 𝑏1 + 𝑐1 + 𝑑1
𝑳: − 0 = 𝑎1 − 3𝑏1 − 𝑐1 + 𝑑1 + 1
𝑻: − 0 = −𝑐1 − 3𝑑1 − 2
𝜽: − 0 = −𝑑1
By solving above equations, we get,
𝒂𝟏 = 𝟑; 𝒃𝟏 = 𝟐; 𝒄𝟏 = −𝟐; 𝒅𝟏 = 𝟎
∴ 𝜋1 = 𝐷3 𝜌2 𝜇 −2 𝐾 0 𝛽𝑔∆𝑡
(𝜷𝒈∆𝒕)𝝆𝟐 𝑫𝟑
∴ 𝝅𝟏 = = 𝑮𝒓 (𝑮𝒓𝒂𝒔𝒉𝒐𝒇𝒇 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟐𝟐)
𝝁𝟐
𝝅𝟐 − 𝑻𝒆𝒓𝒎:
𝜋2 = 𝐷 𝑎2 𝜌𝑏2 𝜇 𝑐2 𝐾 𝑑2 𝐶𝑝

∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎2 (𝑀1 𝐿−3 )𝑏2 (𝑀1 𝐿−1 𝑇 −1 )𝑐2 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑2 (𝐿2 𝑇 −2 𝜃 −1 )
𝑴: − 0 = 𝑏2 + 𝑐2 + 𝑑2
𝑳: − 0 = 𝑎2 − 3𝑏2 − 𝑐2 + 𝑑2 + 2
𝑻: − 0 = −𝑐2 − 3𝑑2 − 2
𝜽: − 0 = −𝑑2 − 1
By solving above equations, we get,
𝒂𝟐 = 𝟎; 𝒃𝟐 = 𝟎; 𝒄𝟐 = 𝟏; 𝒅𝟐 = −𝟏
∴ 𝜋2 = 𝐷0 𝜌0 𝜇1 𝐾 −1 𝐶𝑝
𝝁𝑪𝒑
∴ 𝝅𝟐 = = 𝑷𝒓 (𝑷𝒓𝒂𝒏𝒅𝒕𝒍 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟐𝟑)
𝑲
𝝅𝟑 − 𝑻𝒆𝒓𝒎:
∴ 𝜋3 = 𝐷𝑎3 𝜌𝑏3 𝜇 𝑐3 𝐾 𝑑3 ℎ
∴ 𝑀0 𝐿0 𝑇 0 𝜃 0 = (𝐿1 )𝑎3 (𝑀1 𝐿−3 )𝑏3 (𝑀1 𝐿−1 𝑇 −1 )𝑐3 (𝑀1 𝐿1 𝑇 −3 𝜃 −1 )𝑑3 (𝑀1 𝑇 −3 𝜃 −1 )
𝑴: − 0 = 𝑏3 + 𝑐3 + 𝑑3 + 1
𝑳: − 0 = 𝑎3 − 3𝑏3 − 𝑐3 + 𝑑3
𝑻: − 0 = −𝑐3 − 3𝑑3 − 3
𝜽: − 0 = −𝑑3 − 1
By solving above equations, we get,
𝒂𝟑 = 𝟏; 𝒃𝟑 = 𝟎; 𝒄𝟑 = 𝟎; 𝒅𝟑 = −𝟏
∴ 𝜋3 = 𝐷1 𝜌0 𝜇 0 𝐾 −1 ℎ
Heat Transfer (2151909) 7. Convection

𝒉𝑫
∴ 𝝅𝟑 = = 𝑵𝒖 (𝑵𝒖𝒔𝒔𝒆𝒍𝒕 𝑵𝒖𝒎𝒃𝒆𝒓) − − − − − − − −(𝟕. 𝟐𝟓)
𝑲
Put the values of π1 , π2 and π3 in equation 7.21, we get,
𝑓1 (𝜋1 , 𝜋2 , 𝜋3 ) = 0
(𝛽𝑔∆𝑡)𝜌2 𝐷3 𝜇𝐶𝑝 ℎ𝐷
∴ 𝑓1 ( , , )=0
𝜇2 𝐾 𝐾
ℎ𝐷 (𝛽𝑔∆𝑡)𝜌2 𝐷3 𝜇𝐶𝑝
∴ = 𝜑( , )
𝐾 𝜇2 𝐾
∴ 𝑵𝒖 = 𝝋(𝑮𝒓, 𝑷𝒓) − − − − − − − −(𝟕. 𝟐𝟔)
 Hence Nusselt number is a function of Grashoff number and Prandtl number for
natural or free convection.

Key Notes:

 In natural or free convection, the flow is produced by buoyant effects

resulting from temperature difference. These effects are included in the
Grashoff number.
 Reynolds number is important in the case of forced convection and
similarly the Grashoff number is important in the case of free
convection.

7.8 Empirical Co-relations for Free & Forced Convection

 Mathematical analysis of convective heat problems is complicated due to the large
number of variables involved.
 Majority of the convective problems are, therefore, analysed through the technique
of dimensional analysis supported by experimental investigations. The dimensional
analysis helps to develop certain correlations for the convective coefficient.
 The constants and exponents appearing in these correlations for a particular
situation are worked out through experiments.
 Use “Heat & Mass Transfer by Dr. D. S. Kumar” to see different empirical co-relations
for free and forced convection for different cases. (Equations should be given in
examination so no need to remember)
 Some of the important terminology associated with this topic is explained below:

Bulk Temperature & Mean Film Temperature

 The physical properties (µ, ρ, Cp, k) of a fluid are temperature dependent.
7. Convection Heat Transfer (2151909)

 The accuracy of the results obtained by using theoretical relations and the
dimensionless empirical co-relations would depend upon the temperature chosen
for the evaluation of these properties.
 No uniform procedure has been attained in the selection of this reference
temperature.
 However, it is customary to evaluate the fluid properties either on the basis of bulk
temperature or the mean film temperature.
Mean Bulk Temperature:
 The mean bulk temperature (tb) denotes the equilibrium temperature that would
result if the fluid at a cross section was thoroughly mixed in an adiabatic container.
 For internal flow (Heat exchangers), the fluid flowing through the tubes may be
heated or cooled during its flow passage. The bulk temperature is then taken to be
the arithmetic mean of the temperatures at inlet to and at exit from the heat
exchanger tube; i.e.
𝑡𝑖 + 𝑡𝑜
𝑡𝑏 = − − − − − − − −(7.27)
2
Mean Film Temperature:
 It is the arithmetic mean of the surface temperature (𝑡𝑠 ) of a solid and the
undisturbed temperature (𝑡∞ ) of the fluid which flows over the surface. i.e.
𝑡𝑠 + 𝑡∞
𝑡𝑓 = − − − − − − − −(7.28)
2
Characteristic Length OR Equivalent Diameter
 Characteristic length (𝐿) or Diameter (𝐷) has appeared in the dimensionless
numbers discussed in the Art. 7.5.
 The pipe and the flat plate are the simplest geometries for the occurrence of a flow.
However in many instances some complicated geometries are also used and hence
all the calculations of convective heat transfer become much more complicated and
difficult.
 In order to avoid such difficulties, the concept of an equivalent circular tube is used.
This is a tube which would present the same resistance against the flow or would
secure the same heat transfer as the duct usually used under comparable conditions.
 The diameter of an equivalent tube is known as equivalent diameter (𝐷𝑒 ) or
characteristic length (𝐿𝑒 ). The equivalent diameter is usually defined as;
4𝐴𝑐
𝐷𝑒 = − − − − − − − −(7.29)
𝑃
Where,

Prepared By: Bhavin J. Vegada Department of Mechanical Engineering

Page 7.16 Darshan Institute of Engineering & Technology, Rajkot
Heat Transfer (2151909) 7. Convection

𝐴𝑐 = Cross-sectional area and

𝑃 = Perimeter
 The equivalent diameter or characteristic length of few geometries are given below:

For Rectangular Duct:

4𝐴𝑐
𝐷𝑒 =
𝑃 𝒃
4𝑙𝑏
∴ 𝐷𝑒 =
2(𝑙 + 𝑏)
2𝑙𝑏
∴ 𝐷𝑒 = 𝒍
𝑙+𝑏
For Rectangular Annulus:
4𝐴𝑐
𝐷𝑒 =
𝑃
4 × (𝑙1 𝑏1 − 𝑙2 𝑏2 )
∴ 𝐷𝑒 =
2[(𝑙1 + 𝑏1 ) + (𝑙2 + 𝑏2 )] 𝒃𝟐 𝒃𝟏
2 × (𝑙1 𝑏1 − 𝑙2 𝑏2 )
∴ 𝐷𝑒 =
[(𝑙1 + 𝑏1 ) + (𝑙2 + 𝑏2 )]
𝒍𝟐
When, 𝑙1 = 𝑏1 and 𝑙2 = 𝑏2 ,
2 × (𝑙12 − 𝑙22 ) 𝒍𝟏
∴ 𝐷𝑒 =
2𝑙1 + 2𝑙2
For Annulus: (Refer Fig.) 𝒅
4𝐴𝑐
𝐷𝑒 =
𝑃
𝜋
4 (𝑙𝑏 − 4 𝑑2 ) 𝒃
∴ 𝐷𝑒 =
[2(𝑙 + 𝑏) + 𝜋𝑑]

𝒍
For Annulus: (Refer Fig.) 𝒅
4𝐴𝑐
𝐷𝑒 =
𝑃
𝜋
4 × 4 (𝐷2 − 𝑑2 )
∴ 𝐷𝑒 =
𝜋(𝐷 + 𝑑)
∴ 𝐷𝑒 = 𝐷 − 𝑑

𝑫
7. Convection Heat Transfer (2151909)

7.9 Thermal and Hydrodynamic Boundary Layer

 The concept of boundary layer was first introduced by L. Prandtl in 1904 and since
then it has been applied to several fluid flow problems.

A. Hydrodynamic Boundary Layer: Flat Plate

 “When a fluid flows around an object, their exist a thin layer of fluid close to the
solid surface within which shear stresses significantly influence the velocity
distribution. The fluid velocity varies from zero at the solid surface to the velocity of
free stream flow at a certain distance away from the solid surface. This thin layer of
changing velocity has been called the hydrodynamic boundary layer.”
 Consider the parallel flow of a fluid over a flat plate as shown in Fig. 7.3.

𝒖 = 𝟎. 𝟗𝟗𝒖∞
𝑹𝒆 = 𝟓 × 𝟏𝟎𝟓
𝑹𝒆 = 𝟑 × 𝟏𝟎𝟓
𝒖∞

Fig. 7.3 Development of a boundary layer on a flat plate

 The edge facing the direction of flow is called leading edge. The rear edge is called
the trailing edge.
 The 𝑥 − coordinate is measured along the plate surface from the leading edge of the
plate in the direction of flow, and 𝑦 is measured from the surface in the normal
direction.
 The fluid approaches the plate in the 𝑥 − direction with a uniform velocity 𝑢∞ , which
is practically identical to the free stream velocity of the fluid.
 The velocity of the fluid particles in the first fluid layer adjacent to the plate becomes
zero because of the no – slip condition.
 This motionless layer slows down the particles of the neighboring fluid layer as a
result of friction between the particles of these two adjoining fluid layers at different
velocities.
 This fluid layer then slows down the molecules of the next layer and so on.
Heat Transfer (2151909) 7. Convection

 Thus the presence of the plate is felt up to some normal distance 𝛿 (thickness of
velocity boundary layer) from the plate beyond which the free stream velocity
remains unchanged.
 As a result, the 𝑥 − component of the fluid velocity 𝑢 varies from 0 at 𝑦 = 0 to
nearly 𝑢∞ at 𝑦 = 𝛿.
 The region of the flow above the plate bounded by 𝛿 in which the effects of the
viscous shearing forces caused by fluid viscosity are felt is called the velocity or
hydrodynamic boundary layer.
 The thickness of boundary layer (𝛿) increases with distance from the leading edge;
as more and more fluid is slowed down by the viscous effects, becomes unstable and
breaks into turbulent boundary layer.
 In turbulent boundary layer, a very thin layer near the smooth surface remains
laminar, called laminar sub-layer.
 For the flow over a flat surface, if Reynolds No. is less than 5 X 10 5, the flow is
laminar and velocity distribution is parabolic.
 The boundary layer thickness (𝜹):
“It is arbitrarily defined as that distance from the plate surface in which the velocity
reaches 99% of the velocity of the free stream (𝑢 = 0.99𝑢∞ )”
The hypothetical line of 𝑢 = 0.99𝑢∞ divides the flow over a plate into two regions:
(a) The boundary layer region, in which the viscous effects and the velocity changes
are significant and (b) The irrotational flow region, in which the frictional effects are
negligible and the velocity remains essentially constant.

B. Thermal Boundary Layer

 Whenever a flow of fluid takes place over a heated or cold surface, a temperature
field is set-up in the field next to the surface. The zone or thin layer wherein the
temperature field exists is called the thermal boundary layer.
 The temperature gradient results due to heat exchange between the plate and the
fluid.
Cold fluid flowing over a hot plate:
 Consider the flow of a fluid at a uniform temperature of 𝑡∞ over a hot flat plate at
temperature 𝑡𝑠 as shown in Fig. 7.4.
 The fluid particles in the layer adjacent to the surface will reach thermal equilibrium
with the plate and assume the surface temperature t s . These fluid particles will then
exchange energy with the particles in the adjoining fluid layer and so on.
7. Convection Heat Transfer (2151909)

 As a result, a temperature profile will develop in the flow field that ranges from t s at
the surface to t ∞ sufficiently far from the surface.

𝒖 ∞ , 𝒕∞

Fig. 7.4 Thermal boundary layer during flow of cold fluid over a warm plate
 The flow region over the surface in which the temperature variation in the direction
normal to the surface is significant is the thermal boundary layer.
 The thickness of the thermal boundary layer 𝜹𝒕 at any location along the surface is
defined as the distance from the surface at which the temperature difference
(𝑡𝑠 − 𝑡) equals 0.99(𝑡𝑠 − 𝑡∞ ).
 The thickness of the thermal boundary layer increases in the flow direction, since the
effects of heat transfer are felt at greater distances from the surface further
downstream.
Hot fluid flowing over a cold plate:
 If the approaching free stream temperature 𝑡∞ is above the plate surface
temperature 𝑡𝑠 , the thermal boundary layer will have the shape as depicted in Fig.
7.5.

Fig. 7.5 Temperature profile in T.B.L. when warm fluid flows over a cold plate
 The temperature of the fluid changes from a minimum at the plate surface to the
temperature of the main stream at a certain distance from the surface.
 At point A, the temperature of the fluid is the same as the surface temperature 𝑡𝑠 .
Heat Transfer (2151909) 7. Convection

 The fluid temperature increases gradually until it acquires the free stream
temperature 𝑡∞ .
 The distance 𝛿𝑡ℎ measured perpendicularly to the plate surface, denotes the
thickness of thermal boundary layer at a distance 𝑥 from the leading edge of the
plate.

Relation between Thermal & Velocity Boundary Layer

 The velocity profile of the velocity boundary layer is
dependent primarily upon the viscosity of the fluid.
 The temperature profile of the thermal boundary layer is
depends upon the flow velocity, specific heat, viscosity and
thermal conductivity of the fluid.
 The thermo-physical properties of the fluid affect the relative
magnitude of 𝛿 and 𝛿𝑡 , and the non-dimensional Prandtl
𝜇𝐶𝑝
number (𝑃𝑟 = ) constitutes the governing parameter:
𝐾

(i) When 𝑃𝑟 = 1 𝛿𝑡 = 𝛿
(ii) When 𝑃𝑟 > 1 𝛿𝑡 < 𝛿
(iii) When 𝑃𝑟 < 1 𝛿𝑡 > 𝛿

Fig. 7.6 Relation between thermal and hydrodynamic boundary

layer for different Prandtl number