# Heat and Mass Transfer

Syllabus

V1/18.05.2004/1

Heat and Mass Transfer

Syllabus

Lecture Plan Module 1. Introduction 1. 2. 3. 4. 5. 6. 7. 8. 9.

Learning Units Modes of heat transfer Rate equations: conduction, convection and radiation Heat diffusion equation, boundary and initial conditions One dimensional steady state conduction Conduction with thermal energy generation Extended surface heat transfer Two-dimensional steady state conduction: analytical solutions. Conduction shape factor Finite difference and finite volume methods

Hours per topic 1 1 1 3 1 2 1 1 2 1 2

Total Hours 2

2. One Dimensional Steady State Heat Conduction 3. Multidimensional Steady State Heat Conduction 4. Unsteady State Heat Conduction

7

4

5. Convection

6. Heat Exchangers

7. Boiling and Condensation

10. Transient conduction: lumped capacity, Biot and Fourier numbers 11. One-dimensional transient conduction in slab and radial systems: exact and approximate solutions. 12. Finite difference methods: explicit and implicit formulations. 13. Flow over a body, velocity and thermal boundary layers, drag-co-efficient and heat transfer coefficient. 14. Use of various correlations in forced convection heat transfer, flow over a flat plate, and flow across a single cylinder and tube bundles. 15. Flow inside a duct; hydrodynamics and thermal entry lengths; fully developed and developing flow. 16. Free convection heat transfer from vertical surface and vertical cylinder, horizontal surface and horizontal cylinders. 17. Heat exchanger types, flow arrangements, overall heat transfer coefficient, fouling factor, LMTD for parallel flow and counter flow heat exchangers. 18. Effectiveness-NTU method, expression for effectiveness of a parallel flows and counter flow heat exchangers. Multi-pass and cross flow heat exchangers. 19. Different regimes of boiling 20. Mechanism of condensation, Nusselt’s theory of film condensation on a vertical surface, use of correlations in solving film wise condensation on plane surfaces, horizontal tubes and tube banks.

4

1 3

1

8

2

2

2 3

1

1 2 3

V1/18.05.2004/2

Heat and Mass Transfer

Syllabus

8. Radiation Heat Transfer

21. Definitions, concept of a black body, Kirchoff's law, Lambert's Cosine Law, Stefan-Boltzman's law, Plank's distribution law, Wein's displacement law, configuration factor. 22. Radiation heat exchange between two parallel plates, radiation shielding, 23. Radiation heat exchange in an enclosure 24. Fick's law of diffusion, Mass transfer coefficient 25. Evaporation of water into air, Schmidt number, Sherwood number.

2 4

1 1 1 1 2

9. Mass Transfer

V1/18.05.2004/3

Module 1: Learning objectives
• • • Overview: Although much of the material of this module will be discussed in greater detail, the objective of this module is to give you a reasonable overview of heat transfer. Heat transfer modes: You should be aware of the several modes of transfer modes of transfer and their physical origins. Physical insight: Given a physical situation, you should be able to perceive the relevant transport phenomena. The importance of developing this insight must not be underestimated. You will be investing a lot of time to acquire the tools needed to calculate heat transfer phenomena. However, before you can begin to use these tools to solve practical problems, you must have the intuition to determine what is happening physically. In short, you must be able to look at a problem and identify the pertinent transport phenomenon. The example and problems at the end of this module should help you to begin developing this intuition. Rate equations and conservation laws: You should also appreciate the significance of the rate equations and feel comfortable in using them to compute transport rates. You must also recognize the importance of the conservation laws and the need to carefully identify control volumes. With the rate equations, the conservation laws may be used to solve numerous heat transfer problems.

MODULE I

BASICS OF HEAT TRANSFER
1.1 Difference between heat and temperature
In describing heat transfer problems, we often make the mistake of interchangeably using the terms heat and temperature. Actually, there is a distinct difference between the two. Temperature is a measure of the amount of energy possessed by the molecules of a substance. It is a relative measure of how hot or cold a substance is and can be used to predict the direction of heat transfer. The usual symbol for temperature is T. The scales for measuring temperature in SI units are the Celsius and Kelvin temperature scales. On the other hand, heat is energy in transit. The transfer of energy as heat occurs at the molecular level as a result of a temperature difference. The usual symbol for heat is Q. Common units for measuring heat are the Joule and calorie in the SI system. What is Heat Transfer? “Energy in transit due to temperature difference.”

1.2 Difference between thermodynamics and heat transfer
Thermodynamics tells us: • how much heat is transferred (δQ) • how much work is done (δW) • final state of the system Heat transfer tells us: • how (with what modes) δQ is transferred • at what rate δQ is transferred • temperature distribution inside the body Heat transfer complementary Thermodynamics

1.3 Modes of Heat Transfer
• Conduction: An energy transfer across a system boundary due to a temperature difference by the mechanism of inter-molecular interactions. Conduction needs matter and does not require any bulk motion of matter. x A T1 q T2

k

Conduction rate equation is described by the Fourier Law:

r q = − kA∇T
where: q = heat flow vector, (W) k = thermal conductivity, a thermodynamic property of the material. (W/m K) A = Cross sectional area in direction of heat flow. (m2) ∇T = Gradient of temperature (K/m) = ∂T/∂x i + ∂T/∂y j + ∂T/∂z k Note: Since this is a vector equation, it is often convenient to work with one component of the vector. For example, in the x direction: qx = - k Ax dT/dx In circular coordinates it may convenient to work in the radial direction: qr = - k Ar dT/dr

Convection: An energy transfer across a system boundary due to a temperature difference by the combined mechanisms of intermolecular interactions and bulk transport. Convection needs fluid matter.

moving fluid

T∞

Ts>T∞

q
Ts

Newton’s Law of Cooling: q = h As ∆T where: q = heat flow from surface, a scalar, (W) h = heat transfer coefficient (which is not a thermodynamic property of the material, but may depend on geometry of surface, flow characteristics, thermodynamic properties of the fluid, etc. (W/m2 K) As = Surface area from which convection is occurring. (m2) ∆T = TS − T∞ = Temperature Difference between surface and coolant. (K) Free or natural convection (induced by buoyancy forces) Convection Forced convection (induced by external means)

May occur with phase change (boiling, condensation)

20.000 2500 -100. Ts = Absolute temperature of the surface (K) The above equation is derived from Stefan Boltzman law.Table 1.000 • Radiation: Radiation heat transfer involves the transfer of heat by electromagnetic radiation that arises due to the temperature of the body. Ts q”co n v. (K) Tsur = Absolute temperature of surroundings. However.250 liquid: 50 . which describes a gross heat emission rather than heat transfer. Area = A q = ε·σ·A·(Ts4 – Tsur4) where: ε = Surface Emissivity A= Surface Area Ts = Absolute temperature of surface. the rate of radiation heat exchange between a small surface and a large surrounding is given by the following expression: Ts u r q”r a d.25 liquid: 50 – 100 gases: 25 . Typical values of h (W/m2K) Free convection Forced convection Boiling/Condensation gases: 2 . which is a surface property (ε = 1 is black body) σ = Steffan Boltzman constant = 5. and will be dealt with in Module 9.(K) . The expression for the actual radiation heat transfer rate between surfaces having arbitrary orientations can be quite complex. Radiation does not need matter. Emissive power of a surface: where: E=σεTs4 (W/ m2) ε = emissivity.67 x 10-8 W/m2 K4.

. it may be recalled that thermodynamic properties of pure substances are functions of two independent thermodynamic intensive properties.4 0. Thermal conductivity of real gases is largely independent of pressure and may be considered a function of temperature alone.613 0. From the State Postulate given in thermodynamics. Molecules of various materials gain energy through various mechanisms.4 Thermal Conductivity. Lattice vibration may be transferred between molecules as nuclei attract/repel each other. The background for this comes from the introductory Chemistry courses. thermal conductivity is a thermodynamic property of a material. k = k (T) Table 2 gives the values of thermal conductivity for a variety of materials. Thermal Conductivities of Selected Materials at Room Temperature. Kinetic energy transfer between gaseous molecules.0263 It is important that the student gain a basic perspective of the magnitude of thermal conductivity for various materials.1. For solids and liquids. Material Copper Silver Gold Aluminum Steel Limestone Bakelite Water Air Thermal Conductivity. properties are largely independent of pressure and depend on temperature alone. k As noted previously. Gases exhibit energy through the kinetic energy of the molecule.5 2. W/m K 401 429 317 237 60. say temperature and pressure. Energy is gained or lost through collusions of gaseous molecules as they travel through the medium. Table 2.15 1.

being are much more stationary. so that materials which are excellent electrical conductors are excellent thermal conductors. In the case of electrical conductors the electrons are weakly bonded to the molecule and can drift from one molecule to another transporting their energy with them. This is an especially effective transport mechanism. cannot effectively transfer energy through these same mechanisms. . solids may exhibit energy through vibration or rotation of the nucleus. Instead.Solids. Another important mechanism in which materials maintain energy is by shifting electrons into higher orbital rings.

HEAT TRANSFER HEAT TRANSFER PRADIP DUTTA K.SRINIVASAN Department of Mechanical Engg. Indian Institute of Science Bangalore .

What is Heat Transfer? “Energy in transit due to temperature difference.” Thermodynamics tells us: • how much heat is transferred (δQ) • how much work is done (δW) • final state of the system Heat transfer tells us: • how (with what modes) δQ is transferred • at what rate δQ is transferred • temperature distribution inside the body Heat transfer complementary Thermodynamics .

needs matter .without bulk motion of matter Convection .transmission of energy by electromagnetic waves .does not needs matter .molecular phenomenon (diffusion process) .MODES: Conduction .heat carried away by bulk motion of fluid .needs fluid matter Radiation .

Refrigeration and air-conditioning Domestic applications . soldering. stoves.ovens.steam power plant. climate etc) .casting. laser machining Automobiles / aircraft design Nature (weather. toaster Cooling of electronic equipment Manufacturing / materials processing .APPLICATIONS OF HEAT TRANSFER Energy production and conversion . solar energy conversion etc.welding.

. . . T2 . . .. . .. . . T1 . …………. . .. . .. . . . . . .. . . . temperature gradient) (needs medium. .. .Conduction: Conduction: (needs medium. . . . temperature gradient) . .. RATE: q(W) or (J/s) (heat flow per unit time) . . . . T1>T2 . . . .. . . . . . . . . . q’’ . .. .. . .. . . . .. .. . . . . . . .. . solid or stationary fluid . .. .. . . .. . .

x2) q T2 k Heat flux: q” = q / A = .T2) / (x1 . W k = Thermal Conductivity.k A dT/dx.x A T1 Rate equations (1D conduction): • Differential Form q = . m2 T = Temperature. m • Difference Form q = k A (T1 . W/mK A = Cross-sectional Area.k dT/dx (W/m2) (negative sign denotes heat transfer in the direction of decreasing temperature) . K or oC x = Heat flow path.

What is the rate of heat loss through a wall which is 0.7 W/mK.5 m by 3 m on a side ? Solution: To be worked out in class . Measurements made during steady state operation reveal temperatures of 1400 and 1150 K at the inner and outer surfaces. respectively.15 m thick fireclay brick having a thermal conductivity of 1.Example: The wall of an industrial furnace is constructed from 0.

Convection Convection moving fluid Ts>T∞ T∞ q” Ts • Energy transferred by diffusion + bulk motion of fluid) .

. thermodynamics properties etc.nature of flow.Rate equation (convection) Rate equation (convection) U∞ y U∞ u(y) q” y T∞ T(y) Ts Heat transfer rate q = hA( Ts-T ∞ ) W Heat flux q” = h( Ts-T ∞ ) W / m2 h=heat transfer co-efficient (W /m2K) (not a property) depends on geometry .

250 liquid: 50 .25 May occur with phase change (boiling.100 gases: 25 .Free or natural convection (induced by buoyancy forces Convection Convection Forced convection (induced by external means Typical values of h (W/m2K) Free convection Forced convection Boiling/Condensation gases: 2 . condensation) liquid: 50 .20.000 2500 -100.000 .

q” (W/m2) .Radiation: Radiation: q1” q2” T1 no medium T2 necessary RATE: q(W) or (J/s) Flux : Heat flow per unit time.

Emissive power of a surface (energy released per unit area):…………………… E=εσTs4 (W/ m2) ε= emissivity (property)……… σ=Stefan-Boltzmann constant . ) ………… .Rate equation (radiation) Rate equation (radiation) RADIATION: Heat Transfer by electro-magnetic waves or photons( no medium required..

Area = A q”r a d.Rate equations(Contd….) Rate equations(Contd….) Ts u r q”co n v. Ts Radiation exchange between a large surface and surrounding Q”r a d = εσ(Ts4-Tsur4) W/ m2 .

The outside diameter of pipe is 70 mm. what is the rate of heat loss from the surface per unit length of pipe ? Solution: To be worked out in class . If the coefficient associated with free convection heat transfer from the surface to the air is 5 W/m2K. and its surface temperature and emissivity are 200 °C and 0.8.Example: An uninsulated steam pipe passes through a room in which the air and walls are at 25 °C. respectively.

K) which must be applied to the top and side walls to ensure a heat load less than 500 W.k * 45 0 C * 4m 2 L= 500 W L = 0. Thickness of Styrofoam insulation needed to maintain heat load below prescribed Schematic: Assumptions: (1) perfectly insulted bottom. What is the minimum thickness of Styrofoam insulation (k=0.A = k L= 5k∆TW 2 q 5 * 0. a slightly larger heat loss.Module 1: Worked out problems Problem 1: A freezer compartment consists of a cubical cavity that is 2 m on a side.030W/m. (3) steady-state conditions Analysis: Using Fourier’s law. when the inner and outer surfaces are -10 ºC and 350C? Solution: Known: Dimensions of freezer component. Find: value. inner and outer surfaces temperatures. Assume the bottom to be perfectly insulated. for a prescribed value of L.054m = 54mm Comments: The corners will cause local departures from one–dimensional conduction and. the heat rate is given by ∆T A total L Solving for L and recognizing that A total =5*W2 q = q '' . (2) one-dimensional conduction through five walls of areas A=4m2.03W / m. .

as well as with decreasing t. the temperature drop across the chip decreases with increasing k and W. Power dissipated on one surface.P 0.K (0. (5) one-dimensional conduction in chip.k) is of width W=5mm on a side and of thickness t=1mm. Analysis: All of the electrical power dissipated at the back surface of the chip is transferred by conduction through the chip.10 C Comments: for fixed P.001m * 4 W = 2 kW 150W / m. the chip is mounted in a substrate such that its side and back surfaces are insulated. If 4W are being dissipated in circuits mounted to the back surface of the chip.005m 2 ) ∆T = 1. Fourier’s law. . Find: temperature drop across the chip Schematic: Assumptions: (1) steady-state conditions. (4) negligible heat loss from back and sides. while the front surface is exposed to a coolant. what is the steady-state temperature difference between back and front surfaces? Known: Dimensions and thermal conductivity of a chip. Hence.Problem 2: A square silicon chip (k=150W/m. (2) constant properties. (3) uniform dissipation. P = q = kA ∆T = ∆T t t.

.50)m 2 (300 − 40) 0 C q = 8125W Comments: recognize that Newtown’s law of cooling implies a direction for the convection heat transfer rate. Written in the form above. (2) heat transfer coefficient is uniform over plate area Analysis: the heat transfer coefficient rate by convection from the airstreams to the plate can be determined from Newton’s law of cooling written in the form. determine the heat transfer rate from the air to one side of the plate when the plate is maintained at 400C. Substituting numerical values. by 0.Problem 3: Air at 3000C flows over a plate of dimensions 0. if the convection heat transfer coefficient is 250 W/m2.K.25 * 0.50 m. the heat rate is from the air to plate.25 m. q = q '' . q = 250 W / m 2 .K * (0. Known: air flow over a plate with prescribed air and surface temperature and convection heat transfer coefficient.A = hA(T∞ − Ts ) where A is the area of the plate. Find: heat transfer rate from the air to the plate Schematic: Assumptions: (1) temperature is uniform over plate area.

restrictive condition. What is the net transfer rate from the oven walls to the object? Known: spherical object maintained at a prescribed temperature within a oven.67 * 10 −8 W / m 2 . this is an essential. The rate equation is q rad = εA s σ(Tsur 4 − Ts 4 ) Where As=πD2. substituting numerical values.K[(400 + 273) 4 − (80 + 273) 4 ]K 4 q rad = 3.Problem 4 : A water cooled spherical object of diameter 10 mm and emissivity 0. the area of the sphere. q rad = 0. (2) Recognize that the direction of the net heat exchange depends upon the manner in which T sur and Ts are written. Analysis: heat transfer rate will be only due to the radiation mode. Find: heat transfer rate from the oven walls to the object Schematic: Assumptions: (1) oven walls completely surround spherical object. (2) steady-state condition. .04 W Comments: (1) this rate equation is useful for calculating the net heat exchange between a small object and larger surface completely surrounds the smaller one . (3) When performing radiant heat transfer calculations.9 * π(10 * 10 −3 ) 2 m 2 * 5. (3) uniform temperature for areas of sphere and oven walls.9 is maintained at 4000C. it is always necessary to have temperatures in Kelvin (K) units. (4) oven enclosure is evacuated and large compared to sphere.

8(0. Analysis (a) the rate at which radiation is emitted by the surface is emitted q emit = q emit. Known: Find: (a) rate of surface radiation emission.5m2. as well as the heat loss. The surface would. would decrease with increasing time.Problem 5: A surface of area 0. evacuated chamber whose walls are maintained at 25 0C. emissivity 0.K 4 [(423K ) 4 − (298K ) 4 q = 547 W Comments: the foregoing result gives the net heat loss from the surface which occurs at the instant the surface is placed in the chamber.5m 2 )5. cool due to this heat loss and its temperature. (b) net rate of radiation exchange between the surface and chamber walls. Steady-state conditions would eventually be achieved when the temperature of the surface reached that of the surroundings.K 4 [(150 + 273)K ] 4 qemit = 726W (b) The net rate at which radiation is transferred from the surface to the chamber walls is q = εAσ(Ts 4 − Tsurr 4) q = 0.67 * 10 −8 W / m 2 .8 and temperature 1500C is placed in a large.8(0.5m 2 )5. of course. emissivity and temperature of a surface placed in a large. evacuated chamber of prescribed temperature. A = εAσTs 4 qemit = 0. .67 * 10 −8 W / m 2 . Schematic: Assumptions: (1) area of the enclosed surface is much less than that of chamber walls. What is the rate at which radiation is emitted by the surface? What is the net rate at which radiation is exchanged between the surface and the chamber walls? Area.

(2) constant properties. The walls of the chamber are maintained at a lower temperature and a cold gas is circulated through the chamber. Comments: (1) knowing T=Ti at t =0. the foregoing equation could be solved by numerical integration to obtain T (t). (3) chamber walls are large relative to sphere. . . Analysis: applying an energy balance at an instant of time to a control volume about the sphere. Find: equation which could be used to determine the aluminium temperature as a function of time during the cooling process. Temperature and convection coefficient associated with gas flow over the sphere. The validity of the assumption improves with increasing k and decreasing h and D.Problem 6: A solid aluminium sphere of emissivityε is initially at an elevated temperature and is cooled by placing it in chamber. the temperature T of the sphere is uniform. Schematic: Assumptions: (1) at any time t. (2) The validity of assuming a uniform sphere temperature depends upon h. diameter and surface emissivity of a solid aluminium sphere placed in a chamber whose walls are maintained at lower temperature. it follows that E st = − E out . Known: Initial temperature. D. the energy balance has the form d (ρVcT) = −(q conv + q rad ) dt dT A 4 =− [h (T − T∞ ) + εσ(T 4 − Tsurr )] dt ρVc dT 6 4 = [h (T − T∞ ) + εσ(T 4 − Tsurr )] dt ρcD Where A=πD2. Do not attempt to solve. Obtain an equation that could be used to predict the variation of the aluminium temperature with time during the cooling process. and the thermal conductivity of the solid (k). Identifying the heat rates out of the CV due to convection and radiation. V=πD3/6 and A/V=6/D for the sphere.

. The convection coefficient between the plate and the ambient air is 112 w /m2. qs”. (b) For the same ambient air temperature. . qs” and experiences for case (a) Convection process with air at T and for case (b): the same convection process and radiation exchange with surroundings at Tsur Find: temperature of the plate. Schematic: Assumptions: (1) steady state conditions. qconv. calculate the temperature of the plate if its surface emissivity is 0. Known: front surface of insulated plate absorbs solar flux. Considering the plate to have an area As solve for Ts and substitute numerical values to find q '' s . (a) Neglecting radiation exchange with the surroundings.7 0 C = 87 0 C 2 12 W / m . calculate the temperature of the plate under steady-state conditions if the ambient air temperature is 200C.K . the surface energy balance has the form E in − E out =0. and absorbed solar flux. The relevant processes are convection between the plate and the air. (2) no heat loss out backside of plate. . for the two cases. (b) Considering now the radiation exchange between the surface and its surroundings. (3) surroundings large in comparison plate. For an instant of time the conversation requirement is E in − E out =0. identifying the control surface as shown on the schematic.8 and the temperature of the surroundings is also200C. K. Analysis: (a) apply a surface energy balance.Problem 8: The back side of a metallic plate is perfectly insulated while the front side absorbs a solar radiant flux of 800 W/m2.A s − hA s (Ts − T∞ ) = 0 Ts = T∞ + q " / h s Ts = 20 0 C + 800W / m 2 = 20 0 C + 66. . Ts.

Ts decreases as expected.q '' s . Comments: note that by considering radiation exchange.K m .A s − qconv − qrad = 0 4 qs" A s − hA s (Ts − T∞ ) − εA s (Ts 4 − T∞ ) = 0 800 W W W − 12 2 (Ts − [20 + 273]K ) − 0. .3 By trial and error method.8 * 5. find that Ts=338K=650C.K 12Ts + 4.67 * 10 − 8 2 4 (Ts4 − [20 + 273]4K 4 ) = 0 2 m m .536 * 10 − 8Ts 4 = 4650. Note the manner in which qconv is formulated using qconv is formulated using Newton’s law of cooling: since qconv is shown leaving the control surface. the rate equation must be h (Ts − T∞ ) and not h ( T∞ -Ts).

Show that heat flow lines must be normal to isotherms in conduction heat transfer. liquids and solids. How do you respond to this claim? 8.Module 1: Short questions 1. A person claims that heat cannot be transferred in a vacuum. 10. thermal conductivity B. Discuss the mechanism of thermal conduction in gases. When may one expect radiation heat transfer to be important? 6. For which case do you think the energy required will be greater? Why? 7. 9. An ideal gas is heated from 50 ºC to 70 ºC (a) at constant volume and (b) at constant pressure. Will it be true for convection heat transfer? . What is the order of magnitude for the convection heat transfer coefficient in free convection? Forced convection? Boiling? 5. emissivity 3. Which one of the following is not a property of the material ? A. What is the driving force for (a) heat transfer (b) electric current flow and (c) fluid flow? 2. some poor conductors. heat transfer coefficient C. What is the order of magnitude of thermal conductivity for (a) metals (b) solid insulating materials (c) liquids (d) gases? 4. Name some good conductors of heat.

You should also know the physical meaning of each term appearing in the heat equation. Many other geometries exist in nature or in common engineering designs. simple geometry. and you should understand the key thermal properties and how they vary for different substances.Module 2: Learning objectives • The primary purpose of this chapter is to improve your understanding of the conduction rate equation (Fourier’s law) and to familiarize you with heat equation. should be able to develop an appropriate equation to describe systems of arbitrary. by the end of the module. the student should have a fundamental understanding of the conduction process and its mathematical description. SS system with no heat sources for rectangular and cylindrical geometries. using a similar development. Composite thermal resistances for 1-D. • • • • • • . The student should understand to what form does the heat equation reduce for simplified conditions. and what kinds of boundary conditions may be used for its solution? The student should learn to evaluate the heat flow through a 1-D. The student. In short. Steady state heat transfer with no heat sources placed in parallel or in series may be evaluated in a manner similar to electrical resistances placed in parallel or in series. The student should learn to evaluate the heat flow through a 1-D. You should know the origin and implication of Fourier’s law. The student should be comfortable with the use of equivalent thermal circuits and with the expressions for the conduction resistances that pertain to each of the three common geometric. SS system with no heat sources for rectangular and cylindrical geometries.

y. deflection due to temp. E = U = m ⋅ u = m ⋅ c p ⋅ T − Tref = ρ ⋅ V ⋅ c p ⋅ T − Tref ( ) ( ) .Boundary conditions .Initial condition . in a body (i.Design insulation thickness . cp. the latter two terms could often be neglected.Material properties (k.1 Objectives of conduction analysis: The primary objective is to determine the temperature field. t)? .To compute heat flux at any location (using Fourier’s eqn. In case of heat transfer problems. In heat transfer we generalize all such terms as “heat sources”.Heat treatment of metals 2.MODULE 2 ONE DIMENSIONAL STEADY STATE HEAT CONDUCTION 2.Geometry of the body (shape. ½ m 2. ½ mgz.t) depends on: .e.) . size) Why we need T (x. Etc.t). z. expansion.y.Chip temperature calculation .2 General Conduction Equation Recognize that heat transfer involves an energy transfer across a system boundary. The net effect in either case would cause the internal energy of the system to rise. y. . potential energy.z. T(x. & & = Qin + Win system The work in term could describe an electric current flow across the system boundary and through a resistance inside the system. ρ) . or kinetic energy.z.Compute thermal stresses. A logical place to begin studying such process is from Conservation of Energy (1st Law of Thermodynamics) for a closed system: dE dt dE dt & & = Qin − Wout system The sign convention on work is such that negative work out is positive work in. dE dt & & = Qin + Qgen system The energy of the system will in general include internal energy. Alternatively it could describe a shaft turning across the system boundary and overcoming friction within the system. In this case. U. how temperature varies with position within the body) T(x.

the reference temperature. The above equation reduces to: ρ ⋅ c p ⋅ (∆ x ⋅ ∆ y ⋅ ∆ z) ⋅ dT dt system ⎧ ⎡∂ ⎛ ⎫ ∂T ⎞ ⎤ = ⎨ − ⎢ ⎜ − k ⋅ (∆ y ⋅ ∆ z) ⋅ ⎟ ⎥ ⋅ ∆ x ⎬ ∂x ⎠ ⎦ ⎩ ⎣ ∂x ⎝ ⎭ . When we differentiate the above expression with respect to time. disappears: ρ ⋅ cp ⋅ V ⋅ dT dt & & = Qin + Qgen system Consider the differential control element shown below. Heat is assumed to flow through the element in the positive directions as shown by the 6 heat vectors.where Tref is the reference temperature at which the energy of the system is defined as zero. qz+∆z qx qy qy+∆y qx+∆x z y x qz In the equation above we substitute the 6 heat inflows/outflows using the appropriate sign: ρ ⋅ cp ⋅ (∆ x ⋅ ∆ y ⋅ ∆ z) ⋅ ρ ⋅ c p ⋅ (∆x ⋅ ∆y ⋅ ∆z ) ⋅ ∂T ∂t dT dt & = q x − q x + ∆x + q y − q y + ∆y + qz − qz + ∆z + Qgen system Substitute for each of the conduction terms using the Fourier Law: system ⎧ ⎤⎫ ∂T ⎞ ∂T ⎡ ∂T ∂ ⎛ = ⎨− k ⋅ (∆y ⋅ ∆z ) ⋅ − ⎢− k ⋅ (∆y ⋅ ∆z ) ⋅ + ⎜ − k ⋅ (∆y ⋅ ∆z ) ⋅ ⎟ ⋅ ∆x ⎥ ⎬ ∂x ⎣ ∂x ∂x ⎝ ∂x ⎠ ⎦⎭ ⎩ ⎫ ⎧ ⎤⎪ ∂T ⎡ ∂T ∂ ⎛ ∂T ⎞ ⎪ + ⎨ − k ⋅ ( ∆ x ⋅ ∆ z) ⋅ − ⎢ − k ⋅ ( ∆ x ⋅ ∆ z) ⋅ + ⎜ − k ⋅ ( ∆ x ⋅ ∆ z) ⋅ ⎟ ⋅ ∆ y ⎥ ⎬ ∂y ⎣ ∂y ∂y ⎝ ∂y ⎠ ⎪ ⎪ ⎦⎭ ⎩ ⎧ ∂T ⎡ ∂T ∂ ⎛ ∂T ⎞ ⎤⎫ + ⎨ − k ⋅ (∆ x ⋅ ∆ y) ⋅ + ⎢ − k ⋅ (∆ x ⋅ ∆ y) ⋅ + ⎜ − k ⋅ (∆ x ⋅ ∆ y ) ⋅ ⎟ ⋅ ∆ z ⎥ ⎬ ∂z ⎣ ∂z ∂z ⎝ ∂z ⎠ ⎦⎭ ⎩ + &&& ⋅ ( ∆ x ⋅ ∆ y ⋅ ∆ z ) q q where &&& is defined as the internal heat generation per unit volume. being constant.

: system ∂ 2 T ∂ 2 T ∂ 2 T &&& q = 2 + 2 + 2 + ∂x ∂y ∂z k = ∇ 2T + system 1 dT ⋅ α dt &&& q k The vector form of this equation is quite compact and is the most general form. In the case where k is independent of x. the thermal diffusivity: k α ≡ ρ ⋅ cp Then 1 dT ⋅ α dt or. we often find it convenient to expand the del-squared term in specific coordinate systems: Cartesian Coordinates 1 ∂T ∂ 2T ∂ 2T ∂ 2T q ⋅ = + + + a ∂τ ∂x 2 ∂y 2 ∂z 2 k L Circular Coordinates Z R y Θ x .⎧ ⎡∂ ⎛ ⎤⎫ ∂T ⎞ ⎪ ⎪ + ⎨ − ⎢ ⎜ − k ⋅ ( ∆ x ⋅ ∆ z) ⋅ ⎟ ⋅ ∆ y ⎥ ⎬ ∂y ⎠ ⎪ ⎣ ∂y ⎝ ⎦⎪ ⎩ ⎭ ⎧⎡ ∂ ⎛ ∂T ⎞ ⎤⎫ + ⎨ ⎢ ⎜ − k ⋅ ( ∆ x ⋅ ∆ y ) ⋅ ⎟ ⋅ ∆ z ⎥ ⎬ + &&& ⋅ ( ∆ x ⋅ ∆ y ⋅ ∆ z) q ∂z ⎠ ⎦⎭ ⎩ ⎣ ∂z ⎝ Dividing by the volume (∆x⋅∆y⋅∆z). α. ρ ⋅ cp ⋅ dT dt =− system ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞ q ⎜−k⋅ ⎟ − ⎜ − k ⋅ ⎟ − ⎜ − k ⋅ ⎟ + &&& ∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠ which is the general conduction equation in three dimensions. y and z then ρ ⋅ c p dT ⋅ k dt system ∂ 2 T ∂ 2 T ∂ 2 T &&& q = 2 + 2 + 2 + ∂x ∂y ∂z k Define the thermodynamic property. However.

.1 ∂T 1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T ∂ 2T q + + ⋅ = ⎜r ⋅ ⎟+ ⋅ a ∂τ r ∂r ⎝ ∂r ⎠ r 2 ∂θ 2 ∂z 2 k L Spherical Coordinates Z φ r y ϕ x L 1 ∂T 1 ∂ ⎛ 2 ∂T ⎞ 1 ∂ 2T 1 ∂ ⎛ ∂T ⎞ q ⋅ = ⋅ + ⋅ ⎜r ⋅ ⎟+ 2 ⎜ sin θ ⋅ ⎟+ 2 ∂r 2θ 2 2 sin θ a ∂τ r r ∂r ⎠ r ⋅ sin ∂φ ∂θ ⎝ ∂z ⎠ k ⎝ In each equation the dependent variable. partial differential equation. Thus the two partials associated with these directions are equal to zero.τ). One dimensional: If heat is flowing in only one coordinate direction. is a function of 4 independent variables.z. 2. (r. (r. L • Note that the equation is 2nd order in each coordinate direction so that integration will result in 2 constants of integration.y. • • • Steady State: Steady state solutions imply that the system conditions are not changing with time. then it follows that there is no temperature gradient in the other two directions.z. Thus the partial derivative associated with this third direction is equal to zero. Thus ∂T / ∂τ = 0 . No Sources: If there are no heat sources within the system then the term. (x. T.φ. we shall simply look at the simplifications that can be made to the equations to describe specific problems. For the present. q = 0 .θ. The solution of such equations will normally require a numerical solution. Such equations are termed “boundary conditions’. Two dimensional: If heat is flowing in only two coordinate directions. then it follows that there is no temperature gradient in the third direction.τ) and is a 2nd order.3 Boundary and Initial Conditions • The objective of deriving the heat diffusion equation is to determine the temperature distribution within the conducting body. To evaluate these constants two additional equations must be written for each coordinate direction based on the physical conditions of the problem.τ). θ .

2 in y-dir. The internal heat & generation per unit volume is q (W/m3). with T as the dependent variable.z)..• • We have set up a differential equation. qx qx+∆x .4 Heat Diffusion Equation for a One Dimensional System & q T1 q Ax L x T2 y z Consider the system shown above. 2 BC’s needed for each coordinate.Heat equation is first order in time. so that heat can be conducted through the cube only in the x direction.Heat equation is second order in spatial coordinate. The solution will give us T(x. Hence one IC needed.. How many BC’s and IC’s ? . Hence. 2. * 1D problem: 2 BC in x-direction * 2D problem: 2 BC in x-direction.y. Consider the heat flow through an arbitrary differential element of the cube. Solution depends on boundary conditions (BC) and initial conditions (IC). The top. front and back of the cube are insulated. and 2 in z-dir. . bottom. 2 in y-direction * 3D problem: 2 in x-dir.

1) (2.3) q x + ∆x = q x + − kA ∂q x ∆x ∂x (2. If sides were not insulated.7) • • • For T to rise. 3D.5) ∂ ⎛ ∂T ⎞ ∂T & ⎜ k ⎟ + q = ρ c∆ x ∂x ⎝ ∂x ⎠ ∂t Longitudinal conduction Internal heat generation Thermal inertia (2. LHS must be positive (heat input is positive) For a fixed heat input. 2. T rises faster for higher α In this special case. then & ∂ 2T q ρ c ∂ T 1 ∂T = + = 2 ∂x k k ∂t α ∂t (2. heat flow is 1D.5 One Dimensional Steady State Heat Conduction The plane wall: .4) ∂T ∂T ∂ ⎛ ∂T ⎞ ∂T & + kA + A ⎜ k ⎟ ∆ x + A ∆ x q = ρ Ac ∆ x ∂x ∂x ∂x ⎝ ∂x ⎠ ∂t (2.From the 1st Law we write for the element: & & & & ( E in − E out ) + E gen = E st & q x − q x + ∆x + Ax ( ∆x )q = q x = − kA x ∂T ∂x (2.6) If k is a constant. heat flow could be 2D.2) ∂E ∂t (2.

the temperature distribution is obtained: x T ( x ) = (Ts . the above equation may be integrated twice to obtain the general solution: T ( x ) = C1 x + C 2 where C1 and C2 are constants of integration. To obtain the constants of integration.2 − Ts .The differential equation governing heat diffusion is: d ⎛ dT ⎞ ⎜k ⎟=0 dx ⎝ dx ⎠ With constant k.2 ) = Ts .1 − Ts. From Newton’s law of cooling.1 − Ts . 2 Once the constants of integration are substituted into the general equation.1 L The heat flow rate across the wall is given by: dT kA (Ts. we apply the boundary conditions at x = 0 and x = L. in which case T (0) = Ts . q = hA(Ts − T∞ ) the thermal resistance for convection is then T − T∞ 1 = Rt . the equivalent thermal circuit for the plane wall with convection boundary conditions is shown in the figure below .1 and T ( L ) = Ts .conv = s q hA Applying thermal resistance concept to the plane wall.2 qx = − kA = dx L L / kA Thermal resistance (electrical analogy): Physical systems are said to be analogous if that obey the same mathematical equation. The above relations can be put into the form of Ohm’s law: V=IRelec Using this terminology it is common to speak of a thermal resistance: ∆T = qRtherm A thermal resistance may also be associated with heat transfer by convection at a surface.1 ) + Ts .

B and C.1 − Ts . 2 . Since the energy will flow first through block A and then through blocks B and C.The heat transfer rate may be determined from separate consideration of each element in the network. it follows that L 1 1 Rtot = ∑ Rt = + + h1 A kA h2 A Composite walls: Thermal Resistances in Series: Consider three blocks.1 − T∞ . bottom. 2 − T∞ . front and back. we say that these blocks are thermally in a series arrangement.1 Ts . A. as shown. Since qx is constant throughout the network. the heat transfer rate may also be expressed as T −T q x = ∞ . They are insulated on top. and the total thermal resistance Rtot. 2 Ts . it follows that T − Ts . The steady state heat flow rate through the walls is given by: .1 ∞ . 2 q x = ∞ . 2 Rtot Since the resistance are in series.1 = = 1 / h1 A L / kA 1 / h2 A In terms of the overall temperature difference T∞ .

radially.qx = T∞ . Consider the cylinder shown. U is expressed as U= 1 1 L A LB LC 1 + + + + h1 k A k B k C h2 Series-parallel arrangement: The following assumptions are made with regard to the above thermal resistance model: 1) Face between B and C is insulated. L. R2. Assume no heat sources within the wall of the tube. The process will be described by the Fourier Law. from the inside radius.1 − T∞ . 2 = UA∆T L A LB LC 1 1 + + + + h1 A k A k B k C h2 A where U = 1 Rtot A is the overall heat transfer coefficient. 2) Uniform temperature at any face normal to X. The pipe is either insulated on the ends or is of sufficient length. If T1>T2. 2 ∑ Rt = T∞ . 1-D radial conduction through a cylinder: One frequently encountered problem is that of heat flow through the walls of a pipe or through the insulation placed around a pipe. R1. that heat losses through the ends is negligible. to the outside radius. heat will flow outward.1 − T∞ . In the above case. T1 R1 R2 T2 L .

1 − Ts.2 ) = Ts . the thermal resistance in this case can be expressed as: ln rr1 2 2πkL Composite cylindrical walls: Critical Insulation Thickness : Rtot = ln( rr0i ) 2πkL + 1 (2πr0 L)h Insulation thickness : ro-ri .1 − Ts .2 = qx = −kA L / kA dx L Hence.The differential equation governing heat diffusion is: 1 d ⎛ dT ⎞ ⎜r ⎟=0 r dr ⎝ dr ⎠ With constant k. the solution is The heat flow rate across the wall is given by: dT kA (Ts.

we make the second derivative zero: d 2 Rtot = 0 at dro2 r0 = k h = h2 0 2πLk 3 d 2 Rtot −1 1 = + 2 2 2 dro 2πkro L πro hL r = k 0 h Minimum q at ro =(k/h) = rcr (critical radius) 1-D radial conduction in a sphere: . The point of extrema can be found by setting dRtot =0 dr0 or. 1 1 − =0 2πkr0 L 2πhLro2 r0 = k h In order to determine if it is a maxima or a minima. second term decreases. first term increases. or.Objective : decrease q . increase Rtot Vary ro . as ro increases. This is a maximum – minimum problem.

constant k.7 One-Dimensional Steady State Conduction with Internal Heat Generation current carrying conductor.1 − {Ts . 2}⎢ 11−((rr1 //rr )) ⎥ ⎥ ⎢ ⎡ ⎣ ⎤ ⎦ 1 2 → qr = − kA → Rt . steady state.6 Summary of Electrical Analogy System Electrical Cartesian Conduction Cylindrical Conduction Conduction through sphere Convection Current I q q q q Resistance R L kA ln r2 r1 2πkL 1 / r1 − 1 / r2 4π k 1 h ⋅ As Potential Difference ∆V ∆T ∆T ∆T ∆T 2.2 ) = (1 / r1 − 1 / r2 ) dr 1 / r1 − 1 / r2 4π k 2.1 d ⎛ 2 dT ⎞ ⎜ kr ⎟=0 dr ⎠ r 2 dr ⎝ − → T ( r ) = Ts .1 − Ts .cond = dT 4π k (Ts . nuclear reactors.1 − Ts . uniform q Applications: . chemically reacting systems. & E & Energy generated per unit volume is given by q = V & Plane wall with heat source: Assumptions: 1D.

it may be observed that the solution for temperature is no longer linear. . constant k. : r = r0 .1 ⎜1 − 2 ⎟ + s . Hence thermal resistance concept is not correct to use when there is internal heat generation. : x = − L. steady state. Solution : T = − T = Ts . Eliminate Ts. T = Ts dT =0 dr & q 2⎛ r2 ⎞ Solution : T ( r ) = r0 ⎜1 − 2 ⎟ + Ts 4k ⎜ r0 ⎟ ⎠ ⎝ Exercise: Ts may not be known. 2 + ⎜ 2 2 L ⎟ L ⎝ ⎠ dT dx Heat flux : q′′ = −k x Note: From the above expressions.& d 2T q + =0 2 dx k Boundary cond. x = + L. uniform q Start with 1D heat equation in cylindrical co-ordinates & 1 d ⎛ dT ⎞ q ⎜r ⎟+ =0 r dr ⎝ dr ⎠ k Boundary cond.2 + Ts . As an exercise.1 x Ts . Instead.2 & q 2 x + C1 x + C2 2k Use boundary conditions to find C1 and C2 Final solution : T = & qL2 2k ⎛ x 2 ⎞ T − Ts . show that the expression for heat flux is no longer independent of x.1 T = Ts . r = 0. & Cylinder with heat source: Assumptions: 1D. T∞ and h may be specified. using T∞ and h.

. cp.design insulation thickness .y.z.t). how temperature varies with position within the body) T(x. T(x.e. deflection due to temp. size) T(x.y.heat treatment of metals .t) ? .chip temperature calculation . expansion.t) depends on: . etc.y.compute thermal stresses.z) Tb Why we need T(x. ρ …) .initial condition .Objectives of conduction analysis: • to determine the temperature field.material properties (k.z.to compute heat flux at any location (using Fourier’s eqn.boundary conditions . in a body (i.y.geometry of the body (shape.) .z.

insulated on all long sides (1D heat conduction) & q A qx qx+∆x & q = internal heat generation per unit vol. (W/m3) First Law (energy balance) & & & & ( E in − E out ) + E gen = E st q x − q x + ∆x q x = − kA ∂T ∂x ∂E & + A(∆x)q = ∂t ∂q x ∆x ∂x E = (ρ A∆x)u ∂E ∂u ∂T = ρ A∆x = ρA∆xc ∂t ∂t ∂t q x+ ∆x = q x + .Unidirectional heat conduction (1D) Area = A 0 x x x+∆x Solid bar.

. If sides were not insulated. LHS must be positive (heat input is positive) • For a fixed heat input.∂T ∂T ∂ ⎛ ∂T ⎞ ∂T & − kA + kA + A ⎜k ⎟ ∆x + A∆xq = ρ Ac ∆x ∂t ∂x ⎝ ∂x ⎠ ∂x ∂x ∂ ⎛ ∂T ⎞ ∂T & ⎜k ⎟ + q = ρ c∆ x ∂x ⎝ ∂x ⎠ ∂t Longitudinal conduction Internal heat generation Thermal inertia If k is a constant & ∂ 2T q ρ c ∂T 1 ∂T = + = 2 ∂x k k ∂t α ∂t • For T to rise. 3D. heat flow is 1D. heat flow could be 2D. T rises faster for higher α • In this special case.

The solution will give us T(x. • The . * 1D problem: 2 BC in x-direction * 2D problem: 2 BC in x-direction.. 2 BC’s needed for each coordinate.y. and 2 in z-dir. Hence one IC needed. with T as the dependent variable. • We have set up a differential equation.z). Solution depends on boundary conditions (BC) and initial conditions (IC).Boundary and initial conditions: objective of deriving the heat diffusion equation is to determine the temperature distribution within the conducting body.. . 2 in y-dir.Heat equation is second order in spatial coordinate. Hence. • How many BC’s and IC’s ? . 2 in y-direction * 3D problem: 2 in x-dir.Heat equation is first order in time.

.. . ..... ... . .1. .. . 2 ) = q x = − kA = dx L L / kA . . solution is: dT kA T s . . . . . Hot x=0 fluid d dx Ts...... .... .. T∞.1 ....Dimensional Heat Conduction The Plane Wall : ….. . .. .1 .. Ts.. ... . . . ..1 − T s . . . k . .Dimensional Heat Conduction 1..1 − T s ..... 2 (T s .. ... ...... ...2 dT ⎛ ⎜ k dx ⎝ Const. ... … . K...2 x=L ⎞ ⎟ = 0 ⎠ Cold fluid T∞.

Thermal resistance Thermal resistance (electrical analogy) (electrical analogy) OHM’s LAW :Flow of Electricity V=IR elect Voltage Drop = Current flow×Resistance .

Thermal Analogy to Ohm’s Law : ∆ T = qR therm Temp Drop=Heat Flow×Resistance .

x=0 Ts. . .1 qx …. .. . ... ..1 D Heat Conduction through a Plane 1 D Heat Conduction through a Plane Wall Wall T∞...2 T∞. ..1 Ts. . . ...1 Hot fluid T∞...... . ... ....... ...... ... . ... ... ... . k ... .2 Ts.1 Ts. .. . .2 x=L Cold fluid T∞...2 1 h1 A L k A 1 h1 A + L kA + 1 h 2 1 h2 A (Thermal Resistance ) ∑ R t = A .... . . .. . . . . … ... .

.

Resistance expressions Resistance expressions THERMAL RESISTANCES • Conduction Rcond = ∆x/kA • Convection Rconv = (hA) • Fins -1 Rfin = (hηΑ)−1 • Radiation(aprox) 1.5 -1 Rrad = [4AσF(T1T2) ] .

2 = = UA ∆ T L LA LB 1 1 + + + C + h1 A kA kB kC h2 A 1 = Overall heat transfer coefficient where.1 qx LA LB LB kB A LC LC kC A T∞.2 T∞.Composite Walls :: Composite Walls T∞.1 − T ∞ .1 h1 A B C KA KB KC h2 T∞.1 − T ∞ . 2 LA kA A 1 h2 A q x = ∑ Rt T ∞ .2 1 h1 A T ∞ .U = Rtot A .

c = ∆ T q x .Overall Heat transfer Coefficient Overall Heat transfer Coefficient U = 1 R total A = 1 h 1 1 L + Σ k 1 + h 2 Contact Resistance :……… TA TB ∆T A R B t.

U = LA 1 + h1 kA 1 LC LB 1 + + + kB kC h2 Series-Parallel : A T1 KA B KB C Kc LB kB A LC kC A D KD T2 AB+AC=AA=AD LB=LC T1 Assumptions : (1) (2) LA kA A LD kD A T2 Face between B and C is insulated. . Uniform temperature at any face normal to X.

L = 1m Tf = 100°C kII = 10 W/mk AII = 1 m2. L = 1m h = 1000 W/ m2 k Develop an approximate solution for th rate of heat transfer through the wall.Example: Consider a composite plane wall as shown: kI = 20 W/mk qx T1 = 0°C AI = 1 m2. Solution: to be worked out in class .

2 − T∞ .1 D Conduction(Radial conduction in a 1 D Conduction(Radial conduction in a composite cylinder) composite cylinder) h1 r1 r2 k1 T∞.2 1 ( h 2 )( 2π r2 L ) r1 r2 1 ln 2 π Lk ln r2 r3 2 2 π Lk .2 r k 3 2 qr = T∞.1 ∑R t T∞.1 1 ( h 1 )( 2 π r1 L ) T∞ .1 h2 T∞.

second term decreases.Critical Insulation Thickness : T∞ h ri r0 Insulation Thickness : r o-r i Ti R tot 1 = + 2 π kL ( 2 π r0 L ) h ln( r0 ri ) Objective : decrease q . as r0 increases . .first term increases. increases R tot Vary r0 .

Maximum – minimum problem Set dR tot = 0 dr 0 1 1 − 2 π kr 0 L 2 π hLr 2 = 0 0 k r0 = h Max or Min. ? Take : d 2 R tot = 0 dr 2 0 at r0 = k h d 2 R tot 1 −1 = + 2 2 dr 0 2 π kr 0 L π r 2 0 hL h2 = 2 π Lk 0 r0 = k h 3 .

r0 .Minimum q at r0 =(k/h)=r c r (critical radius) R tot good for electrical cables R c r=k/h good for steam pipes etc.

2 Ts. 2 ) = (1 / r1 − 1 / r 2 ) dr − 1 1 2 1 / r1 − 1 / r 2 = 4π k .1 − T s .1 T∞.1 − T s . 1 d ⎛ ⎜ kr 2 r dr ⎝ 2 dT ⎞ ⎟ = 0 dr ⎠ → T ( r ) = T s .2 k Ts.1 D Conduction in Sphere 1 D Conduction in Sphere r2 r1 T∞. cond {T s .1 Inside Solid:……. 2 }⎡⎢⎢⎣ 11−−((rr // rr )) ⎤⎥⎥⎦ dT 4 π k (T s .1 → q r = − kA → R t .

2 Cold fluid 1D.1 T∞. uniform & q x=0 x=+L . steady state.2 T∞. constant k.Conduction with Thermal Energy Generation Conduction with Thermal Energy Generation & E & = Energy generation per unit volume q= V Applications: * current carrying conductors * chemically reacting systems * nuclear reactors The Plane Wall : Assumptions: Ts.1 Hot fluid x= -L k & q Ts.

1 x Ts .) & d 2T q + =0 2 k dx Boundary cond.. 2 + Ts .1 ⎜1 − 2 ⎟ + Final solution : T = + ⎜ L ⎟ 2k ⎝ 2 2 L ⎠ No more linear Heat flux : dT q′′ = −k x dx Derive the expression and show that it is no more independent of x Hence thermal resistance concept is not correct to use when there is internal heat generation . 2 & q 2 Solution : T = − x + C1 x + C2 2k Use boundary conditions to find C1 and C2 & qL2 ⎛ x 2 ⎞ Ts .Conduction with thermal energy generation (cont. : x = − L. T = Ts .1 T = Ts .. x = + L. 2 − Ts .) Conduction with thermal energy generation (cont.

T = Ts Ts may not be known. uniform q & Start with 1D heat equation in cylindrical co-ordinates: & q & 1 d ⎛ dT ⎞ q ⎜r ⎟+ = 0 r dr ⎝ dr ⎠ k Boundary cond. : r = r0 .Cylinder with heat source Cylinder with heat source T∞ h ro r Ts Assumptions: 1D. T∝ and h may be specified. steady state. dT =0 r = 0. dr & q 2⎛ r2 ⎞ Solution : T (r ) = r0 ⎜1 − 2 ⎟ + Ts 4k ⎜ r0 ⎟ ⎝ ⎠ . Instead. constant k. Exercise: Eliminate Ts. using T∝ and h.

and electrical resistivity R = 0. and the heat transfer coefficient is 4W/m2K. The wire is submerged in a liquid at 110°C. Calculate the centre temperature of the wire at steady state condition.99 Ω. The length of the wire is 1m.Cylinder with heat source Cylinder with heat source Example: A current of 200A is passed through a stainless steel wire having a thermal conductivity k=19W/mK. Solution: to be worked out in class . diameter 3mm.

⎤ d ⎡ d d = −k [2bx ] = −2bk ⎢ 2 ⎥ dx ⎣ dx (a + bx ) ⎦ dx q = −2(−2000 0 CC / m 2 ) × 50 W / m. B=-20000c/ m2. (a) What is the heat generation rate in the wall? (b) Determine the heat fluxes at the two wall faces. and x in meters.Problem 1: The steady-state temperature distribution in a one–dimensional wall of thermal conductivity 50W/m.0 × 10 5 W / m 3 (b) The heat fluxes at the wall faces can be evaluated from Fourier’s law. (b) heat fluxes at the wall faces and relation to q. q in the wall. (3) constant properties. (2) one –dimensional heat flow. . one dimensional condition with constant properties is . Schematic: Assumptions: (1) steady-state conditions. where a=2000C. Find: (a) the heat generation rate. Analysis: (a) the appropriate form of heat equation for steady state.K = 2. . In what manner are these heat fluxes related to the heat generation rate? Known: Temperature distribution in a one dimensional wall with prescribed thickness and thermal conductivity.K and thickness 50 mm is observed to be T(0C)= a+bx2. d ⎡ dT ⎤ q = −K dx ⎢ dx ⎥ ⎣ ⎦ q = −k .

050m q '' x (L) = 10.00w / m 2 − 0 x q= x = = 2. ' q 'x (0) − q '' x (L) + q L = 0 . is then x=0 q x ( x ) = −k '' q '' x (0) = 0 ' atX = L. . q " (L) − q " (0) 10. it follows that E in − E out + E g = 0 .000 W / m 2 Comments: from an overall energy balance on the wall.0 × 10 5 W / m 3 L 0.' q 'x ( x ) = − k dT dx x Using the temperature distribution T(x) to evaluate the gradient. . dx The flux at the face.050m . find d [a+bx2]= -2kbx.K (−2000 0 C / m 2 ) × 0. . q 'x (l) = −2kbL = −2 × 50 W / m.

a (1/m).and lower most layers are well mixed and serve to maintain the upper and lower surfaces of the central layer at uniform temperature T1 and T2. Although there is bulk fluid motion in the mixed layers. which is also constant. (b) Determine whether conditions are steady or transient. where T1>T2. A general form for the conduction flux is . (c) Obtain an expression for the rate at which thermal energy is generated in the entire central layer. Consider conditions for which solar radiation absorption in the central layer provides non uniform heat generation of the form q=Ae-ax. per unit surface area. and the temperature distribution in the central layer is A −ax e + bx + c ka 2 The quantities A (W/m3). there is no such motion in the central layer. B (K/m) and C(K) are known constants having the prescribed units. and k is the thermal conductivity. (2) one-dimensional conduction. Find: (a) heat fluxes at lower and upper surfaces of the central layer.Problem 2: A salt gradient solar pond is a shallow body of water that consists of three distinct fluid layers and is used to collect solar energy. (3)constant properties. (b) whether conditions are steady or transient (c) rate of thermal energy generation for the entire central layer. Schematic: Assumptions: (1) central layer is stagnant. (a) Obtain expressions for the rate at which heat is transferred per unit area from the lower mixed layer to the central layer and from central layer to the upper mixed layer. The upper. Analysis (1) the desired fluxes correspond to conduction fluxes in the central layer at the lower and upper surfaces. T( x ) = − Known: Temperature distribution and distribution of heat generation in central layer of a solar pond.

A ⎡ ⎤ q " = ⎢− k e −ax + B⎥ cond ka ⎣ ⎦ Hence A ⎡A ⎤ ⎡ ⎤ q " = q " ( x =L 0 = ⎢− k e −al + B⎥ q " = q " ( x =0) = − k ⎢ + B⎥ l cond u cond ka ⎣ ka ⎦ ⎣ ⎦ (b) Conditions are steady if ∂T/∂t=0. " . Eg = − A −ax e a L 0 =− A −aL A (e − 1) = (1 − e −aL ) a a . necessitating use of minus signs in the above energy balance. ∂ 2 T q 1 ∂T + = ∂ 2 t k α ∂t Hence conditions are steady since . the energy generation is L L E g = ∫ 0 qdx = A ∫ 0 e − axdx . Applying the heat equation. " E g = q1 .q" = (-q”cond(x=0))-(q”cond(x=L)) 2 Eg = k . A A A + B − K e −aL + B = (1 − e −aL ) ka ka a Comments: Conduction is the negative x-direction. from an overall energy balance. . " q " − q1 + E g =0 2 ." Alternatively. − A −ax A −ax 1 ∂Τ = e + e k k α ∂t ∂T =0 ∂t (for all 0<=x<=L) For the central layer.

The heat fluxes at the wall faces can be evaluated from Fourier’s law. d 2T dx 2 Hence. derive expressions for the heat generation rate per unit volume in the wall and heat fluxes at the two wall faces(x=0. '' . L). (3) homogeneous medium. q" (L) = -K [3AL2+2BL+C] x COMMENTS: (1) from an over all energy balance on the wall. T(x)=Ax3+Bx2+CX+d. (2) one-dimensional heat flow. dT q " = −k = − k[3Ax 2 + 2Bx + C] x dx Using the expression for the temperature gradient derived above. '' . d 2T q + =0 dx 2 k Or q = −k . Known: steady-state temperature distribution in one-dimensional wall of thermal conductivity. . L). we can also find L L E g = ∫ 0 q ( x )dx = ∫ − k[6Ax + 2B]dx = − k[3Ax 2 + 2Bx ]0 0 . the heat fluxes are: Surface x=0. '' g = −3AkL2 − 2BkL . Find: expressions for the heat generation rate in the wall and the heat fluxes at the two wall faces(x=0. the generation rate is q=− . Analysis: the appropriate form of the heat diffusion equation for these conditions is . q" (0)=-kC x Surface x=L. . L E . find E in − E out + E g = 0 q " x (0) − q " x (L) = (− kC) − (− K )[3AL2 + 2BL + C] + E g = 0 E g = −3AkL2 − 2BkL . d ⎡ dT ⎤ d 2 ⎢ dx ⎥ = − k dx [3Ax + 2Bx + C + 0] dx ⎣ ⎦ q = − k[6Ax + 2B] which is linear with the coordinate x. From integration of the volumetric heat rate. . .Problem 3: The steady state temperatures distribution in a one-dimensional wall of thermal conductivity and thickness L is of the form T=ax3+bx2+cx+d. Assumptions: (1) steady state conditions. Hence.

te. x=L/2. (b) sketch temperature distributions for following conditions: initial condition (t<=0). (2) no internal heat generation.will a steady-state condition be reached?. Assuming that the insulation is perfect. The electrical heater is suddenly energized providing a uniform heat flux q”o at the surface x=0. sketch the temperature distributions for the initial condition (t<=0) and for several times after the heater is energized. the boundaries at x=L and else where are perfectly insulated. the heater power is switched off. initially at a uniform temperature Ti.] Schematic: Assumptions: (1) one dimensional conduction. (d) After a period of time te has elapsed. Known: one dimensional system. Find: (a) proper form of heat diffusion equation. reached after heater has been switched off the following an elapsed time .cp. L/2.t) at the planes x=0. and the system characteristics M. is suddenly exposed to a uniform heat flux at one boundary while the other boundary is insulated.Ti.te. (d) expression for uniform temperature. (b) On T-x coordinates. with the heater on. Analysis: (a) the appropriate form of the heat equation follows. Derive an expression that can be used to determine Tf a function of the parameters q”o. and A(the heater surface area). identify boundary and initial conditions. Tf. and x=L as a function of time. several times after heater is energized .Problem 4: The one dimensional system of mass M with constant properties and no internal heat generation shown in fig are initially at a uniform temperature Ti. Also the appropriate boundary and initial conditions are: ∂ 2 T 1 ∂T = ∂x 2 α ∂t Initial condition: T(x. (a) Write the differential equation and identify the boundary and initial conditions that could be used to determine the temperature as a function of position and time in the system. 0) =Ti uniform temperature . (c) sketch heat flux for x=0. (3) constant properties. sketch the heat flux q”x(x. L as a function of time. the system will eventually reach final uniform temperature Tf. Will a steady-state temperature distribution ever be reached? (c) On q”x-t coordinates.

E in = E st . ( d) If the heater is energized until t=to and then switched off. . (c) The heat flux as a function of time for positions x=0. the system will eventually reach a uniform temperature . Perform an energy balance on the system. for an interval of time ∆t=te. .Boundary conditions: x=0 x=L q " = − k∂T / ∂x ) 0 t > 0 o ∂T / ∂x ) L = 0 Insulated (b) The temperature distributions are as follows: No steady-state condition will be reached since E in − E out and E in is constant. t E in = Q in = ∫ 0e q " A s dt = q " A s t e 0 o E st = Mc(Tf − Ti ) q" A s t e o Mc It follows that q " A s t e = Mc(Tf − Ti ) o OR Tf = Ti + . . L/2 and L appears as: . Tf.

(2) Constant properties Analysis: for prescribed conditions. Under steady state conditions with no generation. Find: surface convection coefficient Schematic: Assumptions: (1) one-dimensional. and thermal conductivity of a Plate.cond = R " t .K / W h = 30 W / m 2 .02m 2 . Hence.cond L = 0.02 + 1 / h )m .k T1 − T∞ 30 0 C q" = = = 1500W / m 2 2 (L / k ) + (1 / h ) (0. q" = cond T1− T2 15 0 C = = 1500 W / m 2 L / 2 0. steady conduction with no generation.K / W h . Plate midpoint temperature. is constant.K) is well insulated on its sides.K Comments: The contributions of conduction and convection to the thermal resistance are R " t . What is the value of the convection heat transfer coefficient at the bottom surface? Known: length.5m / 50 W / m. while the top surface is at 1000C and the bottom surface is convectively cooled by a fluid at 200C. surfacethermal conditions.Problem 5: A 1–m-long steel plate (k=50W/m.033m 2 . a thermocouple at the midpoint of the plate reveals a temperature of 850C.K / W K 1 = = 0.

Problem 6:

The wall of a building is a composite consisting of a 100-mm layer of common brick, a 100-mm layer of glass fiber(paper faced. 28kg/m2), a 10-mm layer of gypsum plaster (vermiculite), and a 6-mm layer of pine panel. If the inside convection coefficient is 10W/m2.K and the outside convection coefficient is 70W/m2.K, what are the total resistance and the overall coefficient for heat transfer?
Known: Material thickness in a composite wall consisting of brick, glass fiber, and vermiculite and pine panel. Inner and outer convection coefficients. Find: Total thermal resistance and overall heat transfer coefficient. Schematic:
Brick glass Gypsum Pine panel,Kp Kb Kgl hi=10W/m2.K

h0=70W/m2.K

100mm 10mm 6mm
1 h0

Lb Kb

Lg l k gl

L gy k gy

Lp Kp

1 hi

Assumptions: (1) one dimensional conduction, (2) constant properties, (3)

negligible contact resistance. kg1= 0.038W/m.K: gypsum, kgy=0.17W/m.K: pine panel, kp=0.12W/m.K.
Analysis: considering a unit surface Area, the total thermal resistance Properties: Table A-S, T= 300K: Brick, kb=1.3 W/m.K: Glass fiber (28kg/m3),

R" = tot

1 L B L g1 L gy L p 1 + + + + + h 0 K B k g1 k gy K p h i

0.1 0.01 0.006 1 ⎤ m 2 .K ⎡ 1 0.1 + + + + R" = ⎢ + tot ⎣ 70 1.3 0.038 0.17 0.12 10 ⎥ W ⎦ R " = (0.0143 + 0.0769 + 2.6316 + 0.0588 + 0.0500 + 0.1)m 2 .K / W tot R " = 2.93m 2 ..K / W tot

The overall heat transfer coefficient is
U= 1 R tot A = 1 R
" tot 2

= (2.93m 2 .K / W ) −1

U = 0.341W / m .K.

Comments: an anticipated, the dominant contribution to the total resistance is

made by the insulation.

Problem 7:

The composite wall of an oven consists of three materials, two of which are known thermal conductivity, kA=20W/m.K and kC=50W/m.K, and known thickness, LA=0.30m and LC=0.15m. The third material, B, which is sandwiched between materials A and C, is of known thickness, LB=0.15m, but unknown thermal conductivity kB. Under steady-state operating conditions, measurements reveal an outer surface temperature of Ts,0=200C, an inner surface temperature of Ts,i=6000C and an oven air temperature of T =8000C. The inside convection coefficient h is known to be 25W/m2.K. What is the value of kB?
Known: Thickness of three material which form a composite wall and thermal

conductivities of two of the materials. Inner and outer surface temperatures of the composites; also, temperature and convection coefficient associated with adjoining gas.
Find: value of unknown thermal conductivity, kB. Schematic:
Ts,0=200C

Ts,i=600 C T∞=8000C h= 25W/m2.K

0

KA

KB

KC

LA=0.3m LB=LC=0.15m kA=20W/m.K kC=50W/m.K

LA

LB

LC

Assumptions: (1) steady state conditions, (2) one-dimensional conduction, (3)

constant properties, (4) negligible contact resistance, (5) negligible radiation effects.
Analysis: Referring to the thermal circuit, the heat flux may be expressed as

Ts,i −T s,0 (600 − 20) 0 C q = = 0.3m 0.15m 0.15m LA LB LC + + + + 0.018 KB 50 W / m.K KA KB KC
"

=

580 W / m2 0.018 + 0.15 / K B

The heat flux can be obtained from
q " = h (T∞ − Ts,i ) = 25W / m 2 .K (800 − 600) 0 C q '' = 5000W / m 2

Substituting for heat flux,
0.15 580 580 = " − 0.018 = − 0.018 = 0.098 KB 5000 q K B = 1.53W / m.K.

Comments: radiation effects are likely to have a significant influence on the

net heat flux at the inner surface of the oven.

Problem 8:

A steam pipe of 0.12 m outside diameter is insulated with a 20-mm-thick layer of calcium silicate. If the inner and outer surfaces of the insulation are at temperatures of Ts,1=800K and Ts,2=490 K, respectively, what is the heat loss per unit length of the pipe?
Known: Thickness and surface temperature of calcium silicate insulation on

a steam pipe.
Find: heat loss per unit pipe length. Schematic:

Ts,1=800K
D2=0.16m

Steam D1=0.12m Calcium silicate insulation Ts,2=490K

Assumptions: (steady state conditions, (2) one-dimensional conduction, (3) constant properties. Properties:Table, A-3, calcium silicate (T=645K): k=0.089W/m.K Analysis: The heat per unit length is
q 'r = q 'r = q r 2πK (Ts,1 −T s, 2 ) = qL ln(D 2 / D1 ) 2π(0.089 W / m.K )(800 − 490)K ln(0.16m / 0.12m)

q 'r = 603W / m

Comments: heat transferred to the outer surface is dissipated to the surroundings

by convection and radiation.

Problem 9:

A long cylindrical rod of 10 cm consists of a nuclear reacting material (k=0.0W/m.K) generating 24,000W/m3 uniformly throughout its volume. This rod is encapsulated within another cylinder having an outer radius of 20 cm and a thermal conductivity of 4W/m.K. the outer surface is surrounded by a fluid at 1000C, and the convection coefficient between the surface and the fluid is 20W/m2.K. Find the temperatures at the interface between the two cylinders and at the outer surface.
Known: A cylindrical rod with heat generation is cladded with another cylinder

whose outer surface is subjected to a convection process. Find: the temperature at the inner surfaces, T1, and at the outer surface, Tc.
Schematic:

Assumptions: (1) steady-state conditions, (2) one-dimensional radial

conduction, (3), negligible contact resistance between the cylinders.
Analysis: The thermal circuit for the outer cylinder subjected to the

convection process is

1 / 2π × 4 W / m.0276K.1) 2 m 2 = 754. The heat rate equation has the form q = ∆T / R ' .8 0 C TC = 100 0 C + 754. . on the inner cylinder.K × 2π × 0.0 W / m Hence Ti = 100 0 C + 754.m / W = 100 + 30 = 130 0 C Comments: knowledge of inner cylinder thermal conductivity is not needed.m / W . Numerical values: R '2 = 1 / 20W / m 2 . hence ' Ti − T∞ = q ' × (R 1 × R '2 )andq ' = ∆T / R ' ' R 1 = ln 0.m / W = 100 + 50.000 W / m 3 × π × (0.m / W q ' = 24. E out = E g .' R1 = ln ro / r1 2πk 2 1 h 2πro R '2 = Using the energy conservation requirement.K = 0.20m = 0.0398K.0276 + cccc)K. .2 / 0. Find that q ' = q1 × πr12 .0 W / m × (0.8 = 150.0398K.0 W / m × 0.

(c) insulation thickness which provides the lowest value of the maximum insulation temperature. (3) constant properties. with a contact resistance of 0. what is its surface temperature? (b) If a very thin coating of electrical insulation is applied to the cable. what are the insulation and cable surface temperatures? (c) There is some concern about the ability of the insulation to withstand elevated temperatures. diameter and environmental conditions associated with a cable. and the total coefficient associated with convection and radiation between the cable and the environment is approximately 25W/m2.K) will yields the lowest value of the maximum insulation temperature? What is the value of the maximum temperature when the thickness is used? Known: electric current flow. Corresponding value of this temperature. (2) one-dimensional conduction in r. resistance. . Schematic: Ts T∞ q Eg Assumptions: (1) steady-state conditions.e. The cable is in an environment having temperature of 300C.Problem 10: An electrical current of 700 A flows through a stainless steel cable having a diameter of 5mm and an electrical resistance of 6*10-4 /m (i.K. (b) cable surface and insulation temperatures for a thin coating of insulation. Find: (a) surface temperature of bare cable. What thickness of this insulation (k=0. perimeter of cable length).5W/m.02m2K/W. (a) If the cable is bar.

7 0 C (b) With thin coating of insulation.c + ⎟ + T∞ = ⎜ ⎜ π(0.withq ' = I 2 R 'e = (700A) 2 (6 × 10 −4 Ω / m) = 294 W / m. find m 2 .02 ⎟ + 30 0 C + 0. it follows that E g = q or.Analysis: (a) the rate at which heat is transferred to the surroundings is fixed by the rate of heat generation in the cable.c 1 R t .c + h q' = And solving for the surface temperature. The heat transfer rate is determined by heating within the cable. there exists contact and convection resistances to heat transfer from the cable.005m) Ts = 778. q= Ts − T∞ Ts − T∞ = 1 R t . and therefore remains the same. .K )π(0.e Or . however.c + + hπD i L πD i L hπD i L πD i (Ts − T∞ ) 1 R t .K 294W / m ⎛ 1⎞ ⎛ ⎜ 0. I 2 R 'e L = h (πD i L)(Ts − T∞ ).005m) ⎝ W ⎟ W h⎠ ⎝ ⎠ Ts = q' πD i Ts = 11530 C The insulation temperature is then obtained from q= Ts − T∞ R t . Performing an energy balance for a control surface about the cable.04 R t .K ⎞ m 2 . for the bare cable. It follows that Ts = T∞ + 294W / m q' = 30 0 C + hπD i (25W / m 2 .

5 0 C − 294 m 2 .27 + 0.20C.005) + + π(0.005m) (c) The maximum insulation temperature could be reduced by reducing the resistance to heat transfer from the outer surface of the insulation.K = = 0.K Hence.005m. D0 − Di Dcr − Di (0. To minimize the maximum temperature.r /D i ) 2ππ + 1 hππ c.Ti = Ts − qR t .2 C 0 πD i L = 692.005m) 2ππ(0.70C with no insulation or fro 11530C with a thin coating. rcr = k 0.K/W ln(0.c.0175m t= The cable surface temperature may then be obtained from q '` = R "t. 294 Ts − 30 0 C T − 30 0 C W = = s m (1. Use of the critical insulation thickness also reduces the cable surface temperatures to 692.c + Ts − T∞ ln(D c.5 0 C recognizing that .K W × 0. Ti = Ts − qR t.02 W m π(0.04 − 0.5W / m. add insulation in the amount.K Ts = 692.02m h 25W / m 2 . Dcr =0.04m> Di =0.5W/. which exists at the inner surface of the insulation.04m) m .02m 2 .32)m.K/W q = (Ts − Ti )/R t.005)m = = 2 2 2 t = 0. .7 0 C R t .c.02 m W π(0.K/W 2.c = 11530 C − q Ti = 778.66 + 0. ) 1 W 25 2 π(0. Such a reduction is possible Di<Dcr.70C to 318.c πD i hence.25m.005m) Comments: use of the critical insulation in lieu of a thin coating has the effect of reducing the maximum insulation temperature from 778.K × 0.c = 11530 C − πD i L " 294 W m 2 .r = Ts − 30 0 C 0.50C from 778.04/0. = Ts − q Ti = 318. R " t.

in which case " " q 2 = q3 . Known: Temperature distribution in a composite wall. the linear temperatures distributions in A and B indicate no generation. each of constant thermal conductivity. is shown below. and (c) heat fluxes as a function of distance x. (b) relative magnitudes of thermal conductivities. the heat flux in C increases with decreasing. constant k). 2 3 3 (b) Comment on the relative magnitudes of kA and kB and ok kB and kC. (2) one-dimensional conduction. Hence.Problem 11: The steady state temperature distribution in a complete plane wall of three different methods. (c) Plot the heat flux as a function of x. since dT/dx increases with decreasing x. steady state. Schematic: 1 A 2 B 3 C 4 Assumptions: (1) steady-state conditions. (3) constant properties. Analysis: (a) for the prescribed conditions (one-dimensional. " " q3 > q 4 However. Hence. the parabolic temperature distribution in C implies the existence of heat generation. (a) On the relative magnitudes of q " andq " and of q " andq "4 . Find: (a) relative magnitudes of interfacial heat fluxes.

the interface at 4 is adiabatic.A = q "2. q" x O " q2 x2 x3 x Comments: Note that. (d) It follows that the flux distribution appears as shown below.C and dT/dx/B <dT/dx) C. it follows from Fourier’s law that K A > KC Similarly." " (b) Since conservation of energy requires that q3.C=0. B = q3. since q "2. it follows that K A < KB. with dT/dx)4.B and dT/dx) A > dT/dx) B . .

(3) constant properties. Properties: table A-I. If the bar is in ambient air with h=5W/m . K Analysis: the maximum mid plane temperature is qL2 T0 = + Ts 2K . q (I).015W / m 2 . Find: maximum permissible current. a copper bus bar of rectangular cross section (6mm*150mm) experiences uniform heat generation at a rate q = al 2 . 2 its maximum temperature must not exceed that of the air by more than 300C.Problem 12: When passing an electrical current I. whereα = 0. (2) steady-state conditions. (4) negligible radiation effects. K and . Or substituting the energy balance results.A 2 . in a rectangular bus bar. what is the allowable current capacity for the bar? Known: Energy generation. . copper: k 400W/m. Schematic: Assumptions: (1) one-dimensional conduction in x (W>>L).

003 m 1 ⎜ 3 2 + ⎜ 0 . To − T∞ ⎛ I = ⎜ ⎜ 0 . 015 I L ⎜ ⎟. 003 m 800 W / m . 1 ⎞ 1 ⎞ ⎛ L ⎛ L 2 To − T∞ = q L⎜ + + ⎟ = 0 . Ts = T∞ + q L / h. . 015 L ( L / 2 k + 1 / h ) ⎝ ⎞ ⎟ ⎟ ⎠ 1 2 I max ⎛ ⎜ 30 0 C = ⎜ 0 . h ⎠ h ⎠ ⎝ 2k ⎝ 2k hence . A ) 0 .. K ⎝ ⎞2 ⎟ ⎟ ⎟ ⎟ ⎠ 1 I max = 1826 A . K 5W / m 2 . 015 (W / m .

Module 2: Short questions 1. has one maximum. and where would you place the origin? . the maximum occurs at a value of x given by & q x x=0 (A) 0 L (B) 2 & q (C) k (D) L x=L 6. or three-dimensional? Would the heat transfer be steady or transient? Also. How does transient heat transfer differ from steady state heat transfer? 2. What is meant by the term “one-dimensional” when applied to conduction heat transfer? 3. two-. The steady temperature profile in a one-dimensional heat transfer across a plane & slab of thickness L and with uniform heat generation. there is a value for the shell radius for a nonzero shell thickness at which the heat flux is maximized. Consider a cold canned drink left on a dinner table. This value is (A) k/h (B) h/k (C) h/r (D) r/h 5. If the slab is cooled by convection at x = 0 and insulated at x = L. For heat transfer through a single cylindrical shell with convection on the outside. q . Would you model the heat transfer to the drink as one-. What is meant by thermal resistance? Under what assumptions can the concept of thermal resistance be applied in a straightforward manner? 4. which coordinate system would you use to analyse this heat transfer problem.

7. Consider a round potato being baked in an oven? Would you model the heat transfer to the potato as one-. and where would you place the origin? 8. which coordinate system would you use to analyse this heat transfer problem. or three-dimensional? Would the heat transfer be steady or transient? Also. two-. Consider an egg being cooked in boiling water in a pan? Would you model the heat transfer to the egg as one-. two-. or three-dimensional? Would the heat transfer be steady or transient? Also. which coordinate system would you use to analyse this heat transfer problem. and where would you place the origin? .

Learning Objectives: • Students should recognize the fin equation. • Students should be able to write boundary conditions for (a) very long fins. • Students should be able to incorporate fins into an overall electrical network to solve 1-D. (c) convective tip fins and (d) fins with a specified tip temperature. (b) insulated tip fins. • Students should know the 2 general solutions to the fin equation. • Students should be able to apply the concept of fin efficiency to define an equivalent thermal resistance for a fin. • Students should be able to apply the boundary conditions to the fin equation and obtain a temperature profile. SS problems with no sources. . • Students should be able to apply the temperature profile to the Fourier Law to obtain a heat flow through the fin.

electronic equipment (CPUs). the higher the h.2 Extended surface analysis: In this module. so that the extra surface of the protrusions is also in contact with the fluid. 3. portable generators. 1 . automobile radiators. • Many times. consideration will be limited to steady state analysis of rectangular or pin fins of constant cross sectional area. when the first option is not in our control and the second option (i.1 Introduction: Convection: Heat transfer between a solid surface and a moving fluid is governed by the Newton’s cooling law: q = hA(Ts-T∞). Example: a cooling fan. Numerical methods of integration or computer programs can be used to advantage in such cases. Annular fins or fins involving a tapered cross section may be analyzed by similar methods. Example: a heat sink with fins. Fins are protrusions from the base surface into the cooling fluid. where Ts is the surface temperature and T∞ is the fluid temperature.e. we are left with the only alternative of increasing the effective surface area by using fins or extended surfaces. this equation reduces to the form: d 2 T &&& q =0 2 + dx k (2) This is a second order. but will involve solution of more complicated equations which result.). Increase the contact surface area A. to increase the convective heat transfer.MODULE 3 Extended Surface Heat Transfer 3. ordinary differential equation and will require 2 boundary conditions to evaluate the two constants of integration that will arise. increasing h) is already stretched to its limit. One-Dimensional Conduction. one can • • Increase the temperature difference (Ts-T∞) between the surface and the fluid. Increase the convection coefficient h. Most of you have encountered cooling fins on air-cooled engines (motorcycles. This can be accomplished by increasing the fluid flow over the surface since h is a function of the flow velocity and the higher the velocity. air conditioning equipment (condensers) and elsewhere. Therefore. We start with the General Conduction Equation: 1 dT ⋅ system α dτ = ∇ 2T + &&& q k (1) After making the assumptions of Steady State. etc.

Note that as energy is conducted down the length of the fin.Consider the cooling fin shown below: h. by convection. (This is the area through with heat is conducted. T∞ Ts. q ∆x Across this segment the heat loss will be h⋅(P⋅∆x)⋅(T-T∞). from the sides. We further note that the arrows indicating the direction of heat flow point in both the x and y directions. Thus the heat flow varies along the length of the fin. q 2 . which remove an amount of energy equivalent to what would be lost through the sides by convection. h. depending on the geometry of the fin. However. it is convenient to analyse a fin by examining an equivalent one–dimensional system. The fin has a cross sectional area. some portion is lost. quite often. L. where P is the perimeter around the fin. which cools with convective coefficient. Ac.or three-dimensional heat flow. This is an indication that this is truly a two. The equivalent heat sink would be &&& ⋅ ( Ac ⋅ ∆ x ) . h. Consider a differential length of the fin. q L Ac y The fin is situated on the surface of a hot surface at Ts and surrounded by a coolant at temperature T∞. T∞ T0.) and an overall length. The equivalent system will involve the introduction of heat sinks (negative heat sources).

Steady State Conduction with Sources: d 2T h ⋅ P − ⋅ (T − T∞ ) = 0 dx 2 k ⋅ Ac (4) which is the equation for a fin with a constant cross sectional area. Recall that nonhomogeneous differential equations require both a general and a particular solution. we see that h. k and Ac are all independent of x in the defined system (They may not be constant if a more general analysis is desired. 3. First. We can make this equation homogeneous by introducing the temperature relative to the surroundings: θ ≡ T .T∞ Differentiating this equation we find: dθ dT = +0 dx dx (7) (8) Differentiate a second time: d 2θ d 2 T = dx 2 dx 2 (9) Substitute into the Fin Equation: d 2θ − m2 ⋅ θ = 0 dx 2 (10) This equation is a Second Order. a couple of simplifications should be noted.). This is the Second Order Differential Equation that we will solve for each fin analysis. Prior to solving. Let m2 = h⋅ P k ⋅ Ac (5) then: d 2T 2 2 − m ⋅ (T − T ) = 0 ∞ dx (6) Next we notice that the equation is non-homogeneous (due to the T∞ term). Homogeneous Differential Equation. P.3 Solution of the Fin Equation 3 . We replace this ratio with a constant.Equating the heat source to the convective loss: &&& = q − h ⋅ P ⋅ (T − T∞ ) Ac (3) Substitute this value into the General Conduction Equation as simplified for One-Dimension.

The general solution to the above differential equation will be a linear combination of each of the independent solutions. An alternative solution can be obtained as follows: Note that the hyperbolic sin. 4 . the hyperbolic cosine. and hence we need two boundary conditions to determine the two constants of integration. Differentiate this expression twice: dθ = α ⋅ eα ⋅ x dx (11) d 2θ = α 2 ⋅ eα ⋅ x dx 2 Substitute this trial solution into the differential equation: α2⋅eα⋅x – m2⋅eα⋅x = 0 Equation (13) provides the following relation: α=± m (12) (13) (14) We now have two solutions to the equation. Then: θ = A⋅em⋅x + B⋅ e-m⋅x (15) where A and B are arbitrary constants which need to be determined from the boundary conditions. θ = C ⋅ cosh(m ⋅ x ) + D ⋅ sinh(m ⋅ x ) (18) Generally the exponential solution is used for very long fins.We apply a standard technique for solving a second order homogeneous linear differential equation. Try θ = eα⋅x. the hyperbolic solutions for other cases. sinh. are defined as: sinh(m ⋅ x ) = e m⋅ x − e − m⋅x 2 cosh(m ⋅ x ) = e m⋅ x + e − m⋅ x 2 (16) We may write: C ⋅ cosh(m ⋅ x ) + D ⋅ sinh(m ⋅ x ) = C ⋅ e m⋅ x + e − m÷ x e m⋅ x − e − m÷ x C + D m⋅ x C − D − m⋅ x + D⋅ = ⋅e + ⋅ e (17) 2 2 2 2 We see that if (C+D)/2 replaces A and (C-D)/2 replaces B then the two solutions are equivalent. cosh. Note that it is a 2nd order differential equation.

1 h ⋅ P ⋅ k ⋅ Ac ⋅ e − m⋅ x ⋅ θ 0 = Mθ 0 e − mx (26) Often we wish to know the total heat flow through the fin.T∞ ≡ θ0 (19) The second boundary condition depends on the condition imposed at the other end of the fin. i. the end located a long distance from the heat source will approach the temperature of the surroundings. There are various possibilities.e. at the fin base. the heat flow entering at the base (x=0). Now apply the second boundary condition. The first one is obvious. Very long fins: For very long fins. θ(0) = T0 . θ(0) = θ0 = B⋅ e-m⋅0 ⇒ B = θ0 The general temperature profile for a very long fin is then: θ(x) = θ0 ⋅ e-m⋅x If we wish to find the heat flow through the fin. The only way that this equation can be valid is if A = 0. as one end of the fin will be attached to a hot surface and will come into thermal equilibrium with that surface. we may apply Fourier Law: q = − k ⋅ Ac ⋅ dT dθ = − k ⋅ Ac ⋅ dx dx (22) (23) (24) Differentiate the temperature profile: dθ = − θ o ⋅ m ⋅ e − m⋅ x dx (25) So that: ⎡ h ⋅ P ⎤ 2 − m⋅ x q = k ⋅ Ac ⋅ θ 0 ⋅ ⎢ ⎥ ⋅e = ⎣ k ⋅ Ac ⎦ where M = hPkAc . 5 . Hence. θ(∞) = 0 Substitute the second condition into the exponential solution of the fin equation: (20) ∞ 0 (21) θ(∞) = 0 = A⋅em⋅∞ + B⋅ e-m⋅∞ The first exponential term is infinite and the second is equal to zero. as described below. Hence.Boundary Conditions: Since the solution results in 2 constants of integration we require 2 boundary conditions.

dθ = C ⋅ m ⋅ sinh(m ⋅ x ) + D ⋅ m ⋅ cosh(m ⋅ x ) dx (29) (30) Apply the first boundary condition at the base: θ (0) = θ 0 = C sinh( m ⋅ 0) + D cosh(m ⋅ 0) 0 1 (31) So that D = θ0.q= h ⋅ P ⋅ k ⋅ Ac ⋅ θ 0 = Mθ 0 (27) The insulated tip fin Assume that the tip is insulated and hence there is no heat transfer: dθ dx =0 x =L (28) The solution to the fin equation is known to be: θ = C ⋅ cosh(m ⋅ x ) + D ⋅ sinh(m ⋅ x ) Differentiate this expression. Now apply the second boundary condition at the tip to find the value of C: dθ ( L) = 0 = Cm sinh( m ⋅ L) + θ 0 m cosh(m ⋅ L) dx which requires that C = −θ 0 cosh(mL) sinh( mL) (32) (33) This leads to the general temperature profile: θ ( x) = θ0 cosh m( L − x ) cosh(mL) (34) We may find the heat flow at any value of x by differentiating the temperature profile and substituting it into the Fourier Law: q = − k ⋅ Ac ⋅ dT dθ = − k ⋅ Ac ⋅ dx dx (35) 6 .

we see that the primary difference in form is in the hyperbolic tangent term. tanh(mL)→1. which is as the ratio of fin heat transfer and the heat transfer without the fin. Two other possibilities are usually considered for fin analysis: (i) a tip subjected to convective heat transfer. ε f . Distribution cosh m( L − x ) + ( h cosh mL + ( h Fin heat transfer mk mk ) sinh m( L − x ) ) sinh mL M θo sinh mL + ( h ) cosh mL mk cosh mL + ( h ) sinh mL mk cosh m( L − x ) cosh mL (θ L Mθ 0 tanh mL θ b ) sinh m( L − x ) + sinh m( L − x ) sinh mL Mθ 0 (cosh mL − θL θb ) sinh mL D e − mx M θ0 3. Other tip conditions: We have already seen two tip conditions. The expressions for temperature distribution and fin heat transfer for all the four cases are summarized in the table below. for long fins. For an adiabatic fin: εf = qf q = qf hAC (Tb − T∞ ) = hPkAC tanh( mL) = hAC kP tanh( mL) hAC (37) If the fin is long enough. represents the reduced heat loss due to the shortening of the fin. mL>2. That term. Hence.1 Case Tip Condition A Convection heat transfer: hθ(L)=-k(dθ/dx)x=L B C Adiabatic (dθ/dx)x=L=0 Given temperature: θ(L)= θL Infinitely long fin θ(L)=0 Temp.4 Fin Effectiveness How effective a fin can enhance heat transfer is characterized by the fin effectiveness.So that the energy flowing through the base of the fin is: q = hPkAc θ 0 tanh( mL) = Mθ 0 tanh( mL) (36) If we compare this result with that for the very long fin. 7 . Table 3. which always results in a number equal to or less than one.1). and hence it can be considered as infinite fin (case D in Table 3. and (ii) a tip with a prescribed temperature. one being the long fin and the other being the insulated tip.

it is not necessary to enhance heat transfer by adding heat fins. However ε f ≥ 2 is considered unjustifiable because of diminishing returns as fin length increases. The smaller the W.h εf = qf q = qf hAC (Tb − T∞ ) = = Rt . (Why?) P/AC should be as high as possible. To increase ε f . heat fins are more effective if h is low. the fin’s resistance should be lower than the resistance due only to convection.h Rt . f (Tb − T∞ ) / Rt . the fin’s material should have higher thermal conductivity. Use a square fin with a dimension of W by W as an example: P=4W. if h is very high. Well. The effectiveness of a fin can also be characterized by (Tb − T∞ ) / Rt . h. and the higher the ε f . It seems to be counterintuitive that the lower convection coefficient.Conclusion: It is preferred to use thin and closely spaced (to increase the total number) fins. the fin would have no purpose as it would serve as an insulator instead). the higher ε f . An ideal fin is thought to be one made of a perfect or infinite conductor material. P/AC=(4/W). AC=W2.5 Fin Efficiency The fin efficiency is defined as the ratio of the energy transferred through a real fin to that transferred through an ideal fin.εf → kP ⎛k⎞ P = ⎜ ⎟ hAC ⎝ h ⎠ AC (38) In order to enhance heat transfer. A perfect conductor has an infinite thermal conductivity so that the entire fin is at the base material temperature. η= qreal = qideal h ⋅ P ⋅ k ⋅ Ac ⋅ θ L ⋅ tanh(m ⋅ L) h ⋅ ( P ⋅ L) ⋅ θ L (40) 8 . Therefore. 3. Observations: • • If fins are to be used on surfaces separating gas and liquid. ε f should be greater than 1 (In case ε f <1. f (39) It is a ratio of the thermal resistance due to convection to the thermal resistance of a fin. In order to enhance heat transfer. fins are usually placed on the gas side. the higher is the P/AC. k.

h.⎢ ⎥ = (T − T∞ ⎢η.Tb x Real situation Simplifying equation (40): Tb x Ideal situation η= k ⋅ Ac θ L ⋅ tanh(m ⋅ L) tanh(m ⋅ L) = h⋅ P L⋅θL m⋅ L (41) The heat transfer through any fin can now be written as: ⎡ 1 ⎤ q. A f Overall Fin Efficiency: Overall fin efficiency for an array of fins qf qb Define terms: Ab: base area exposed to coolant Af: surface area of a single fin At: total area including base area and total finned surface.h. f = (43) η. At=Ab+NAf N: total number of fins Heat Transfer from a Fin Array: 9 . A f ⎥ ⎣ ⎦ (42) The above equation provides us with the concept of fin thermal resistance (using electrical analogy) as 1 Rt .

to in crease the effective area η O At .O = R t .qt = qb + Nq f = hAb (Tb − T∞ ) + Nη f hAf (Tb − T∞ ) = h[( At − NAf ) + Nη f Af ](Tb − T∞ ) = h[ At − NAf (1 − η f )](Tb − T∞ ) = hAt [1 − NAf At (1 − η f )](Tb − T∞ ) = ηO hAt (Tb − T∞ ) NAf At (1 − η f ) Define overall fin efficiency: ηO = 1 − q t = hAtη O ( T b − T ∞ ) = Tb − T∞ 1 w h ere R t .O = 1 /(hAtηO ) q= T1 − T∞ T1 − T∞ = ∑ R R1 + Rb + Rt .f T o en han ce h eat tran sfer Atη O > > A T h at is.O h Atη O C om p are to h eat tran sfer w ith ou t fin s q = h A ( T b − T ∞ ) = h ( Ab + N Ab .O 10 . f )( T b − T ∞ ) = w h ere Ab . 1 hA is the b ase area (u nex posed) for th e fin Thermal Resistance Concept: L1 t A=Ab+NAb.f T1 T ∞ Rb= t/(kbA) T2 Tb T1 T2 Tb T∞ R1=L1/(k1A) Rt .

Pradeep datta HT/M3/L1/V1/2004 1 . heat transfer problem involving heat generation. • For steady state ∂/ ∂t = 0 • No generation & q=0 • To solve for the full equation. Only one initial condition is needed to account for the transient behavior. it requires a total of six boundary conditions: two for each direction.Multidimensional Heat Transfer H e a t D iffu s io n E q u a tio n ρc p ∂T ∂ 2T ∂ 2T ∂ 2T & & + ) + q = k∇ 2 T + q = k( + 2 2 2 ∂t ∂x ∂y ∂z • This equation governs the Cartesian. temperature distribution for a three-dimensional unsteady.

(refer to handout) Pradeep datta 2 HT/M3/L1/V1/2004 . the heat equation is simplified to ∂2T ∂2T + 2 = 0. it needs two boundary conditions in each direction. (see Table 4. steady state situation. usually will be solved using computers.D. 2 ∂x ∂y There are three approaches to solve this equation: • Numerical Method: Finite difference or finite element schemes. • Graphical Method: Limited use.Two-D.1 for selected configurations) • Analytical Method: The mathematical equation can be solved using techniques like the method of separation of variables. Steady State Case For a 2 . the conduction shape factor concept derived under this concept can be useful for specific configurations. However.

Conduction Shape Factor This approach applied to 2-D conduction involving two isothermal surfaces. with all other surfaces being adiabatic. S=A/L • Common shape factors for selected configurations can be found in Table 4.1 Pradeep datta HT/M3/L1/V1/2004 3 . The heat transfer from one surface (at a temperature T1) to the other surface (at T2) can be expressed as: q=Sk(T1-T2) where k is the thermal conductivity of the solid and S is the conduction shape factor. • The shape factor can be related to the thermal resistance: q=Sk(T1-T2)=(T1-T2)/(1/kS)= (T1-T2)/Rt where Rt = 1/(kS) • 1-D heat transfer can use shape factor also. Ex: heat transfer inside a plane wall of thickness L is q=kA(∆T/L).

T2 From Table 8.5)(3.5W/m.02)(100 + 20) . The horizontal pipe is a thin-walled of outside diameter of 50 cm.5) q = kS(T1 − T2 ) = (0. = 1812(W) heat loss for every meter of pipe T1 Pradeep datta HT/M3/L1/V1/2004 4 . z>3D/2 z=1 m 2πL 2π (1) S= = = 3. L>>D. The pipe is very long and the averaged temperature of the oil is 100°C and the ground soil temperature is at -20 °C (ksoil=0. case 1.K).7. estimate the heat loss per unit length of pipe.02 ln(4z / D) ln(4 / 0.Example An Alaska oil pipe line is buried in the earth at a depth of 1 m.

45°C for every 1 km of pipe length. • The heat transfer can not be considered constant for a long pipe Ground at -20°C Heat transfer to the ground (q) m cpT Pradeep datta HT/M3/L1/V1/2004 m cp (T+ dT) Length dx 5 .K. ∆T = = = 0.045(° C ) & mC P 2000 * 2 Therefore. • Heating might be needed to prevent the oil from freezing up.2 & q = mC P ∆T . q 181.) If the mass flow rate of the oil is 2 kg/s and the specific heat of the oil is 2 kJ/kg.Example (cont. For example. determine the temperature change in 1 m of pipe length. the total temperature variation can be significant if the pipe is very long.

C = 120 & mCP T(x) = -20 + 120e −0. at inlet x = 0.51(T + 20)( dx ) ln(4z / D) & & Energy balance: mC P T − q = mCP (T + dT ) dT dT + 1. = −0. of 100°C • It reaches 0°C at x=4740 m.51(T + 20) = 0.000378 x . reheating is required every 4. 0 1000 2000 x 3000 4000 5000 HT/M3/L1/V1/2004 6 50 Pradeep datta . integrate dx T + 20 T ( x ) = −20 + Ce −0. T(0) = 100° C.7 km.) Heat Transfer at section with a temperature T(x) 2πk(dx) q= (T + 20) = 1.Example (cont.000378 x 100 50 T( x ) 0 • Temperature drops exponentially from the initial temp. therefore.000378dx.

In heat transfer problems. In numerical analysis. especially when the geometry of the object of interest is irregular. numerical solutions obtained using high-speed computer are very useful. two different approaches are commonly used: the finite difference and the finite element methods. or the boundary conditions are nonlinear. The finite difference method involves: Establish nodal networks Derive finite difference approximations for the governing equation at both interior and exterior nodal points Develop a system of simultaneous algebraic nodal equations Solve the system of equations using numerical schemes Pradeep datta HT/M3/L1/V1/2004 7 . analytical solutions usually are not available.Numerical Methods Due to the increasing complexities encountered in the development of modern technology. the finite difference method is used more often and will be discussed here. For these problems.

n+1 Example: m-1. y=n∆y Pradeep datta . Each node now represents a small region where the nodal temperature is a measure of the average temperature of the region.The Nodal Networks The basic idea is to subdivide the area of interest into sub-volumes with the distance between adjacent nodes by Dx and Dy as shown.n-1 intermediate points m+½.n ∆y m-½. If the distance between points is small enough. n x=m∆x.n m+1. the differential equation can be approximated locally by a set of finite difference equations.n m. ∆x m.n HT/M3/L1/V1/2004 8 m.

Finite Difference Approximation & q 1 ∂T Heat Diffusion Equation: ∇ T + = .n = ∆x Tm . approximated the first order differentiation at intermediate points (m+1/2.n = ∆x HT/M3/L1/V1/2004 9 ( m −1/ 2.n ) ∆x Pradeep datta ( m +1/ 2.n ) .n) & (m-1/2.n − Tm .n − Tm −1. k α ∂t k where α = is the thermal diffusivity ρ C PV 2 ∂ & No generation and steady state: q=0 and = 0. ⇒ ∇2T = 0 ∂t First.n ) ∆x ∂T ∆T ≈ ∂x ( m −1/ 2.n ) Tm +1.n) ∂T ∆T ≈ ∂x ( m +1/ 2.

n Pradeep datta HT/M3/L1/V1/2004 10 . the approximation can be applied to the other dimension y ∂ 2T ∂y 2 T m .Finite Difference Approximation (cont. n m . n ≈ (∆x )2 Similarly.) N ext. n + T m −1. n + 1 + T m . n − 2 Tm . n −1 − 2 T m . approximate the second order differentiation at m.n Tm +1.n ∂T / ∂x m +1 / 2 .n ∂ 2T ∂x 2 ∂ 2T ∂x 2 ≈ m .n − ∂T / ∂x ∆x m −1 / 2 . n ≈ (∆y )2 m .

n −1 − 2Tm . no generation heat equation: ∇2T = 0 This approximation can be simplified by specify ∆x=∆y and the nodal equation can be obtained as Tm +1.n To model the steady state.n + Tm −1.lim( ∆y → 0) = ∆x ∂x ∆y ∂y Pradeep datta HT/M3/L1/V1/2004 11 .n ⎛ ∂ 2T ∂ 2T ⎞ + ⎜ ∂x 2 + ∂y 2 ⎟ ≈ 2 ( ∆x ) ( ∆y ) 2 ⎝ ⎠ m . This approximation is improved when the distance between the adjacent nodal points is decreased: ∆T ∂T ∆T ∂T Since lim( ∆x → 0) = .n − 2Tm .n = 0 This equation approximates the nodal temperature distribution based on the heat equation.n +1 + Tm .n +1 + Tm .n Tm .n + Tm .n + Tm −1.n −1 − 4Tm .) Tm +1.Finite Difference Approximation (cont.

Pradeep datta HT/M3/L1/V1/2004 12 .4 =0 T3. the equation is T2. n=4. For each node.5 . there is one such equation.4 + T4.A System of Algebraic Equations • The nodal equations derived previously are valid for all interior points satisfying the steady state. The result is a system of N algebraic equations for a total of N nodal points. no generation heat equation.3 + T3.5) • Nodal relation table for exterior nodes (boundary conditions) can be found in standard heat transfer textbooks (Table 4.4T3.4=(1/4)(T2. For example: for nodal point m=3.3 + T3.2 of our textbook).4 + T3. • Derive one equation for each nodal point (including both interior and exterior points) in the system of interest.4 + T3.4 + T4.

Matrix Form The system of equations: a11T1 + a12T2 + L + a1N TN = C1 a21T1 + a22T2 + L + a2 N TN = C2 M a N 1T1 + a N 2T2 + L + a NN TN = C N A total of N algebraic equations for the N nodal points and the system can be expressed as a matrix formulation: [A][T]=[C] M M M M ⎡ a11 a12 L a1N ⎤ ⎡ T1 ⎤ ⎡ C1 ⎤ ⎢a ⎢T ⎥ ⎢C ⎥ a22 L a2 N ⎥ ⎥ . T = ⎢ 2 ⎥ .C = ⎢ 2 ⎥ where A= ⎢ 21 M M M ⎥ ⎢ M ⎢M ⎥ ⎢ M ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎣ a N 1 a N 2 L a NN ⎦ ⎣TN ⎦ ⎣C N ⎦ Pradeep datta HT/M3/L1/V1/2004 13 .

Pradeep datta HT/M3/L1/V1/2004 14 .com.Numerical Solutions Matrix form: [A][T]=[C]. “Numerical Recipes” by Cambridge University Press or their web source at www.nr. From linear algebra: [A]-1[A][T]=[A]-1[C]. The famous Jacobi & Gauss-Seidel iteration methods will be introduced in the following. For example. iterative methods are usually more efficient. • For high order matrix. [T]=[A]-1[C] where [A]-1 is the inverse of matrix [A]. [T] is the solution vector. • Matrix inversion requires cumbersome numerical computations and is not efficient if the order of the matrix is high (>10) • Gauss elimination method and other matrix solvers are usually available in many numerical solution package.

(Example : a31T1 + a32T2 + a33T3 + L + a1N TN = C1 . (k-1) means the present level and (k) represents the new level.Iteration General algebraic equation for nodal point: ∑a T j =1 ij i −1 j + aiiTi + j =i +1 ∑aT ij N j = Ci . the new values at iteration (k) can be estimated • The iteration will be stopped when max⎜Ti(k)-Ti(k-1)⎟≤ε. i = 3) Replace (k) by (k-1) Rewrite the equation of the form: for the Jacobi iteration i −1 N aij ( k ) aij ( k −1) Ci (k ) Ti = − ∑ Tj − ∑ Tj aii j =1 aii j =i +1 aii • (k) . • An initial guess (k=0) is needed to start the iteration. where ε specifies a predetermined value of acceptable error Pradeep datta HT/M3/L1/V1/2004 15 . • By substituting iterated values at (k-1) into the equation.specify the level of the iteration.

11 1 1 .Y 2 X + Y + 4 Z = 16 2 4 0=Y0=Z0=1.Yk-Yk-1.(1/2) X0 .Example Solve the following system of equations using (a) the Jacobi methos. (b) the Gauss Seidel iteration method.Zk-Zk-1⎥ ≤ 0. Y = + X+0*Z 2 2 − X + 2Y + 0 * Z = 3. (a) Jacobi method: use initial guess X stop when max⎜Xk-Xk-1. 1 1 Z = 4.(1/2)Y0 .Z 4 2 4 3 1 4 X + 2Y + Z = 11.(1/4)Z0 = 2 Y1= (3/2) + (1/2)X0 = 2 Z1= 4 .(1/4)Y0 = 13/4 X= Pradeep datta HT/M3/L1/V1/2004 16 ⎡ 4 2 1 ⎤ ⎡ X ⎤ ⎡11⎤ ⎢ −1 2 0 ⎥ ⎢ Y ⎥ = ⎢ 3 ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ 2 1 4 ⎥ ⎢ Z ⎥ ⎢16⎥ ⎣ ⎦⎣ ⎦ ⎣ ⎦ Reorganize into new form: .1 First iteration: X1= (11/4) .X.Y.

3] Pradeep datta HT/M3/L1/V1/2004 17 . 128 ⎥ ⎣ ⎦ Final solution [1. ⎥ . Z1=13/4 X2 = (11/4) .1 ⎡ 133 31 393 ⎤ ⎢ 128 . 2. . Y1=2.(1/4)Z1 = 15/16 Y2 = (3/2) + (1/2)X1 = 5/2 Z2 = 4 .) Second iteration: use the iterated values X1=2.(1/4)Y1 = 5/2 Converging Process: 13 ⎤ ⎡ 15 5 5 ⎤ ⎡ 7 63 93 ⎤ ⎡ [1. Y 5 − Y 4 .2. 4 ⎦ ⎣ 16 2 2 ⎦ ⎣ 8 32 32 ⎦ ⎣ ⎡ 519 517 767 ⎤ ⎢ 512 .(1/2) X1 . 2. 16 . Stop the iteration when ⎣ ⎦ max X 5 − X 4 .1].Example (cont. ⎥ .996] Exact solution [1.(1/2)Y1 . ⎢ . ⎥ .02. Z 5 − Z 4 ≤ 0.014. 256 . . ⎢ .1. 2. 256 ⎥ . ⎢ 2.

Z = 4 − X − Y 4 2 4 2 2 2 4 11 1 1 Immediate substitution First iteration: X1 = − (Y 0 ) − ( Z 0 ) = 2 4 2 4 3 1 3 1 5 Y 1 = + X 1 = + (2) = 2 2 2 2 2 1 1 1 1 ⎛ 5 ⎞ 19 Z 1 = 4 − X 1 − Y 1 = 4 − (2) − ⎜ ⎟ = 2 4 2 4⎝2⎠ 8 X= ⎡ 5 19 ⎤ ⎡ 29 125 783 ⎤ ⎡ 1033 4095 24541 ⎤ Converging process: [1. . Y = + X .9995. ⎢ 2.996] and it converges faster Pradeep datta HT/M3/L1/V1/2004 18 .009.Example (cont. 2048 . 2.) (b) Gauss-Seidel iteration: Substitute the iterated values into the iterative process immediately after they are computed.1. ⎢ .1]. ⎥ . ⎢ 1024 . 1. Use initial guess X 0 = Y 0 = Z 0 = 1 11 1 1 3 1 1 1 − Y − Z. . 8192 ⎥ ⎣ 2 8 ⎦ ⎣ 32 64 256 ⎦ ⎣ ⎦ The iterated solution [1. ⎥ .

n + Tm −1.n +1 + Tm .n Pradeep datta m +1. A In this case.n −1 − 4Tm . n m.n = 0 HT/M3/L1/V1/2004 19 .n+1 q1 m-1.n +1 + Tm .n q2 q3 m.Numerical Method (Special Cases) A For all the special cases discussed in the following.n + Tm .n q4 m+1.n Therefore the standard nodal equation can be written as Tm +1.n −1 − 4Tm . Tm-1. the derivation will be based on the standard nodal point configuration as shown to the right.n=Tm+1.n axis A-A = 2T + Tm . m.n-1 Symmetric case: symmetrical relative to the A-A axis.

n ⎛ ∆x ⎞ Tm . Write the energy balance equation (q2=0): Insulated surface A m.n q4 q3 m+1.n k⎜ ⎟ +k⎜ ⎟ + k ∆y =0 ∆y ∆y ∆x ⎝ 2 ⎠ ⎝ 2 ⎠ 2Tm +1. n q1 + q3 + q4 = 0 m.n − Tm .n q2 m.n = 0 This equation is identical to the symmetrical case discussed previously.n −1 − Tm . the surface area for q1 and q3 is only half of their original value.n +1 + Tm .n +1 − Tm .n-1 A Tm +1. Also.) Insulated surface case: If the axis A-A is an insulated wall.n+1 q1 m-1. therefore there is no heat transfer across A-A. Pradeep datta HT/M3/L1/V1/2004 20 .n ⎛ ∆x ⎞ Tm .Special cases (cont.n + Tm .n −1 − 4Tm .

Based on the energy balance concept: q1 + q2 + q3 + q4 + G q1 + q2 + q3 + q4 + g ( ∆x )( ∆y )(1) = 0 Use 1 to represent the dimension along the z-direction.n −1 − 4Tm .n +1 + Tm .n −1 − 4Tm .n +1 + Tm .n + =0 k Pradeep datta HT/M3/L1/V1/2004 21 .n + Tm .Special cases (cont.n + Tm −1. k (Tm +1.n + Tm −1.n ) + g ( ∆x ) = 0 2 g ( ∆x ) 2 Tm +1.) With internal generation G=gV where g is the power generated per unit volume (W/m3).n + Tm .

n-1 m+1. qrad Tsurr m.n − Tm .n − Tm .n ⎛ ∆y ⎞ Tm +1.n − 2εσ ( ∆x ) 4 εσ ( ∆x ) 4 Tm .n = −2 Tsurr k k Non-linear term.n −1 − 4Tm .n From energy balance concept: q2+q3+q4=qrad + + = T −T ⎛ ∆y ⎞ Tm −1.n + 2Tm .n + Tm +1. no generation to simplify the derivation).n + k ⎜ = εσ ( ∆x ) Tm .n − Tsurr ) Tm −1. surrounding temperature Tsurr.n −1 − 4Tm .n 4 4 k⎜ + k ( ∆x ) m . Given surface emissivity ε.n m-1.n ) = 2εσ ( ∆x ) ( Tm .n −1 m .n + 2Tm .Special cases (cont.n − Tsurr ⎟ ⎟ ∆x ∆y ∆x ⎝ 2 ⎠ ⎝ 2 ⎠ ( ) 4 4 k ( Tm −1.n + Tm +1. can solve using the iteration method Pradeep datta HT/M3/L1/V1/2004 22 .) Radiation heat exchange with respect to the surrounding (assume no convection.n q2 q3 q4 m.

if the diameter is tripled. q f (3D ) q f ( D) 3 = 3 2 = 5. Table A-1. aluminum (pure): k=240W/m. copper (pure): k=400W/m. (2) one-dimensional conduction. K Analysis: (a) for an infinitely long fin. (a) If the diameter of the rod is triples. by how much would the rate of heat removal change? Known: long. Schematic: T∞.Problem 1: A long. the fin rate is qr = M = (hpkAc ) 2 θ b q f = (hπDkπD 2 / 4) 2 θ b = 1 1 π 2 (hk ) 2 D 2 θ b 1 3 Where P=πD and Ac=πD2/4 for the circular cross-section. by how much would the rate of heat removal change? (b) If a copper rod of the same diameter is used in place of aluminium. K. Find: (a) increase in heat transfer if diameter is tripled and (b) increase in heat transfer if copper is used in place of aluminum. circular aluminium rod attached at one end to the heated wall and transfers heat through convection to a cold fluid.h Aluminum Or copper q D Assumptions: (1) steady-state conditions.2 And there is a 520 % increase in heat transfer. Note that qf≈D3/2. . (3) constant properties. (5)rod is infinitely long. (4) uniform convection coefficient. aluminum cylinder acts as an extended surface. Properties: Table A-1. Hence.

since qf ≈ k1/2.(b) in changing from aluminum to copper. it follows that q f (Cu ) ⎛k = ⎜ Cu q f ( Al ) ⎜ k Al ⎝ ⎞ 2 ⎛ 400 ⎞ 2 ⎟ =⎜ ⎟ = 1. Comments: (1) because fin effectiveness is enhanced by maximum P/Ac = 4/D. (2) From the standpoint of cost and weight. . aluminum is preferred over copper. the use of a larger number of small diameter fins is preferred to a single large diameter fin.29 ⎟ ⎝ 240 ⎠ ⎠ 1 1 And there is a 29 % increase in the heat transfer rate.

(5) negligible radiation exchange with surroundings. Properties: table A-1: copper T = (650+25)0C ≈ 600K: k=379 W/m. with solder having melting point of 6500C. Schematic: Copper ∞ ∞ D=1cm Junction Tb=6500C Air T∞ = 250C H=10W/m2. h and (7) infinitely long rods.The rods are in the air at 250C with a convection coefficient of 10W/m2.K Analysis: the junction must be maintained at 6500C while energy is transferred by conduction from the junction (along both rods). (6) uniform.Problem 2: Two long copper rods of diameter D=1 cm are soldered together end to end. (3) constant properties. Find: Minimum power needed to solder the rods. K Assumptions: (1) steady-state conditions. (2) one-dimensional conduction along the rods. The minimum power is twice the fin heat rate for an infinitely long fin. K. What is the minimum power input needed to effect the soldering? Known: Melting point of solder used to join two long copper rods. (4) no internal heat generation. .

01m)⎜ 379 ⎟ (0. 1 ⎛ ⎞2 W W ⎞π ⎛ 2 0 = 2⎜10 q (π × 0.K ⎠ 4 ⎝ ⎝ m ⎠ therefore. .K m. ⎜ ⎟ min 2 .9W min Comments: radiation losses from the rod are significant.1 = 2q = 2(hpkA ) 2 (T − T ) q ∞ min f c b substituting numerical values. thereby requiring a larger power input to maintain the junction at 6500C.01m) ⎟ (650 − 25) C. q = 120. particularly near the junction.

K.25 × 10 −6 m 2 ⎟ ⎝ ⎠ 1/ 2 L3c/ 2 (h w / KA P )1 / 2 = 0. Known: Dimensions and number of rectangular aluminum fins. The fins are 50mm long.5mm thick.80 it follows that. Schematic: L=50mm N=250m-1 W=width hw = 30W/m2.K × 2 × 0.16 W / m / K ( wθ b ) . K.0505m) 3/ 2 ⎛ ⎞ 30 W / m 2 .Problem 3: Determine the percentage increase in heat transfer associated with attaching aluminium fins of rectangular profile to a plane wall.05m × ( wθ b ) = 2.0505m × 0. K Analysis: evaluate the fin parameters t = 0.72 × 30 W / m 2 .K × 25. and are equally spaced at a distance of 4mm (250fins/m).0505m 2 A p = L c t = 0. while that resulting from attachment of the fins is 30W/m2.25 × 10 −6 m 2 Lc = L + L3c/ 2 (h w / KA P ) 1/ 2 (0.K(with fins) hwo=40W/m2. pure: k≈240W/m. (6) uniform convection coefficient.K ⎜ ⎟ ⎜ 240 W / m. (4) negligible fin contact resistance. The convection coefficient affected associated with the bare wall is 40W/m2.72h w 2 wLθ b q f = 0.72. ηf = 0. Convection coefficient with and without fins. (2) one-dimensional conduction. hence q f = ηf q max = 0. Aluminum. Properties: table A-1.K(without fins) Aluminium Assumptions: (1) steady-state conditions. (3) constant properties. Find: percentage increase in heat transfer resulting from use of fins. 0.5 × 10 −3 m = 25.

68w θ b .16 = 1416% q wo 40 wθ b (hPkAc) Comments: If the finite fin approximation is made.K Without the fins.K ( wθ b ) m. 1/2 1/2 θ b = (30*2*240*5*10 ) -4 1/2 = w θ b =2.63) ( wθ b ) = 566wθ b m. the heat transfer from the walls is q w = Nq f + (1 − Nt ) wh w θ b W ( wθ b ) + (1m − 250 × 5 × 10 −4 m) × 30 W / m 2 .16 Hence the percentage increases in heat transfer is qw 566 wθ b = = 14.K W q w = (540 + 26.With the fins. Hence qf is . q w = 250 × 2. q wo = h wo 1m × wθ b = 40wθ b . it follows that qf θ b = [hw2wkwt] overestimated.

11. except that the diameter of one of them is twice the diameter of the other? For which fin will the a) efficiency and (b) effectiveness be higher? Explain. 9. 2. Would you recommend adding fins to the inside or outside the tubes? Why? When would you recommend adding fins both inside and outside the tubes? 8. The fins attached to a surface are determined to have an effectiveness of 0. What is the difference between fin effectiveness and fin efficiency? 4. Is the heat flow through a fin truly one-dimensional? Justify the one-dimensional assumption made in the analysis of fins. The heat loss from the tip is usually insignificant compared to the rest of the fin. For which case do you think the fins will provide greater enhancement in heat transfer? Explain. Two plate fins of constant rectangular cross section are identical. Insulated tip condition is often used in fin analysis because (choose one answer) A. whereas they are attached to the second surface afterwards by welding or tight fitting. Under what circumstances the addition of fins may actually decrease heat transfer? 6. Would you recommend adding fins to the inside or outside the tubes? Why? 7. Hot water is to be cooled as it flows through the tubes exposed to atmospheric air. Do you think the rate of heat transfer from the surface has increased or decreased as a result of addition of fins? 5. There is no heat loss from the tip C. Two pin fins are identical. Does the (a) efficiency and (b) effectiveness of a fin increase or decrease as the fin length is increased? 10. Fins are to be attached in order to enhance heat transfer. Fins are normally meant to enhance heat transfer.Module 3: Short questions 1. except that the thickness of one of them is twice the thickness of the other? For which fin will the a) efficiency and (b) effectiveness be higher? Explain. Fins are usually deliberately insulated at their tips B. Hot air is to be cooled as it is forced through the tubes exposed to atmospheric air. . Consider two finned surface which are identical except that the fins on the fist surface are formed by casting or extrusion. and the insulated tip condition makes the problem mathematically simple 3. Fins are to be attached in order to enhance heat transfer.9.

. except that the convection heat transfer coefficient of one of them is twice that of the other? For which fin will the a) efficiency and (b) effectiveness be higher? Explain.12. Two finned surface are identical.

and know how to formulate the finite difference equations for the discrete points of a nodal network. The student should be able to determine whether the shape factor is known for the system. such as the finite difference method. The student should understand what a conduction shape factor is..Module 4: Learning objectives • • • • • The primary objective of this module is to develop an appreciation for the nature of two. use to solve the heat transfer problem.or multi-dimensional conduction problems and the methods that are available for its solutions. For a multi-dimensional problem. the student should be able to determine whether an exact solution is known. the student should be able to use a numerical solution. The student should appreciate the inherent nature of the discretization process. However. Although one may find it convenient to solve these equations using hand calculations for a coarse mesh. This may be done by examining one or more of the many excellent references in which exact solutions to the heat equation are obtained. . and if available. one should be able to treat fine meshes using standard computer algorithms involving direct or iterative techniques. and link it to the concept of thermal resistances in 2D problems. if conditions are such that the use of a shape factor or an exact solution is not possible.

MODULE 4

MULTI-DIMENSIONAL STEADY STATE HEAT CONDUCTION
4.1 Introduction
We have, to this point, considered only One Dimensional, Steady State problems. The reason for this is that such problems lead to ordinary differential equations and can be solved with relatively ordinary mathematical techniques. In general the properties of any physical system may depend on both location (x, y, z) and time (τ). The inclusion of two or more independent variables results in a partial differential equation. The multidimensional heat diffusion equation in a Cartesian coordinate system can be written as:
1 ∂T ∂ 2 T ∂ 2 T ∂ 2 T q ⋅ = + + + a ∂τ ∂x 2 ∂y 2 ∂z 2 k
L

(1)

The above equation governs the Cartesian, temperature distribution for a three-dimensional unsteady, heat transfer problem involving heat generation. To solve for the full equation, it requires a total of six boundary conditions: two for each direction. Only one initial condition is needed to account for the transient behavior. For 2D, steady state (∂/ ∂t = 0) and without heat generation, the above equation reduces to:

∂ 2T ∂ 2T + =0 ∂x 2 ∂y 2

(2)

Equation (2) needs 2 boundary conditions in each direction. There are three approaches to solve this equation: • •
Analytical Method: The mathematical equation can be solved using techniques like the method of separation of variables. Graphical Method: Limited use. However, the conduction shape factor concept derived under this concept can be useful for specific configurations. (see Table 4.1 for selected configurations) Numerical Method: Finite difference or finite volume schemes, usually will be solved using computers.

Analytical solutions are possible only for a limited number of cases (such as linear problems with simple geometry). Standard analytical techniques such as separation of variables can be found in basic textbooks on engineering mathematics, and will not be reproduced here. The student is encouraged to refer to textbooks on basic mathematics for an overview of the analytical solutions to heat diffusion problems. In the present lecture material, we will cover the graphical and numerical techniques, which are used quite conveniently by engineers for solving multi-dimensional heat conduction problems.

4.2 Graphical Method: Conduction Shape Factor
This approach applied to 2-D conduction involving two isothermal surfaces, with all other surfaces being adiabatic. The heat transfer from one surface (at a temperature T1) to the other surface (at T2) can be expressed as: q=Sk(T1-T2) where k is the thermal conductivity of the solid and S is the conduction shape factor. The shape factor can be related to the thermal resistance: q=S.k.(T1-T2)=(T1-T2)/(1/kS)= (T1-T2)/Rt where Rt = 1/(kS) is the thermal resistance in 2D. Note that 1-D heat transfer can also use the concept of shape factor. For example, heat transfer inside a plane wall of thickness L is q=kA(∆T/L), where the shape factor S=A/L. Common shape factors for selected configurations can be found in Table 4.1 Example: A 10 cm OD uninsulated pipe carries steam from the power plant across campus. Find the heat loss if the pipe is buried 1 m in the ground is the ground surface temperature is 50 ºC. Assume a thermal conductivity of the sandy soil as k = 0.52 w/m K. Solution: T2

Z=1 m

T1

The shape factor for long cylinders is found in Table 4.1 as Case 2, with L >> D: S = 2⋅π⋅L/ln(4⋅z/D) Where z = depth at which pipe is buried. S = 2⋅π⋅1⋅m/ln(40) = 1.7 m Then q' = (1.7⋅m)(0.52 W/m⋅K)(100 oC - 50 oC) q' = 44.2 W

Table 4.1 Conduction shape factors for selected two-dimensional systems [q = Sk(T1-T2)] System Schematic
T2 z Isothermal sphere buried in as finite medium T1 T2 Horizontal isothermal cylinder of length L buried in a semi finite medium z T1 D T2 Vertical cylinder in a semi finite medium T1 D
2πL 2 ⎛ 4 w 2 − D12 − D 2 cosh ⎜ 2 D1 D 2 ⎝
−1 ⎜

Restrictions

Shape Factor
2πD 1 − D / 4z

z>D/2

D

L>>D L L>>D z>3D/2

2πL cosh −1 (2 z / D)

2πL ln(4 z / D)

L

L>>D

2πL ln(4 L / D)

Conduction between two cylinders of length L in infinite medium

D1 T1 w

D2 T2

L>>D1,D2 L>>w

⎞ ⎟ ⎟ ⎠

Horizontal circular cylinder of length L midway between parallel planes of equal length and infinite width

∞ z z

T2

∞ z>>D/2 L>>2

2πL ln(8 z / πD)

T1 T2

D ∞

Circular cylinder of length L centered in a square solid of equal length

T2 w T1 D W>D L>>w
2πL ln(1.08w / D)

Eccentric circular cylinder of length L in a cylinder of equal length

D

d T1

T2 z

D>d L>>D

2πL ⎛ D 2 + d 2 − 4z 2 cosh ⎜ 2 Dd ⎝
−1 ⎜

⎞ ⎟ ⎟ ⎠

4.3 Numerical Methods
Due to the increasing complexities encountered in the development of modern technology, analytical solutions usually are not available. For these problems, numerical solutions obtained using high-speed computer are very useful, especially when the geometry of the object of interest is irregular, or the boundary conditions are nonlinear. In numerical analysis, three different approaches are commonly used: the finite difference, the finite volume and the finite element methods. Brief descriptions of the three methods are as follows: The Finite Difference Method (FDM) This is the oldest method for numerical solution of PDEs, introduced by Euler in the 18th century. It's also the easiest method to use for simple geometries. The starting point is the conservation equation in differential form. The solution domain is covered by grid. At each grid point, the differential equation is approximated by replacing the partial derivatives by approximations in terms of the nodal values of the functions. The result is one algebraic equation per grid node, in which the variable value at that and a certain number of neighbor nodes appear as unknowns. In principle, the FD method can be applied to any grid type. However, in all applications of the FD method known, it has been applied to structured grids. Taylor series expansion or polynomial fitting is used to obtain approximations to the first and second derivatives of the variables with respect to the coordinates. When necessary, these methods are also used to obtain variable values at locations other than grid nodes (interpolation). On structured grids, the FD method is very simple and effective. It is especially easy to obtain higher-order schemes on regular grids. The disadvantage of FD methods is that the conservation is not enforced unless special care is taken. Also, the restriction to simple geometries is a significant disadvantage.

Finite Volume Method (FVM) In this dissertation finite volume method is used. The FV method uses the integral form of the conservation equations as its starting point. The solution domain is subdivided into a finite number of contiguous control volumes (CVs), and the conservation equations are applied to each CV. At the centroid of each CV lies a computational node at which the variable values are to be calculated. Interpolation is used to express variable values at the CV surface in terms of the nodal (CV-center) values. As a result, one obtains an algebraic equation for each CV, in which a number of neighbor nodal values appear. The FVM method can accommodate any type of grid when compared to FDM, which is applied to only structured grids. The FVM approach is perhaps the simplest to understand and to program. All terms that need be approximated have physical meaning, which is why it is popular. The disadvantage of FV methods compared to FD schemes is that methods of order higher than second are more difficult to develop in 3D. This is due to the fact that the FV approach requires two levels of approximation: interpolation and integration. Finite Element Method (FEM) The FE method is similar to the FV method in many ways. The domain is broken into a set of discrete volumes or finite elements that are generally unstructured; in 2D, they are usually triangles or quadrilaterals, while in 3D tetrahedra or hexahedra are most often used. The distinguishing feature of FE methods is that the equations are multiplied by a weight function before they are integrated over the entire domain. In the simplest FE methods, the solution is approximated by a linear shape function within each element in a way that guarantees continuity of the solution across element boundaries. Such a function can be constructed from its values at the corners of the elements. The weight function is usually of the same form. This approximation is then substituted into the weighted integral of the conservation law and the equations to be solved are derived by requiring the derivative of the integral with respect to each nodal value to be zero; this corresponds to selecting the best solution within the set of allowed functions (the one with minimum residual). The result is a set of non-linear algebraic equations. An important advantage of finite element methods is the ability to deal with arbitrary geometries. Finite element methods are relatively easy to analyze mathematically and can be shown to have optimality properties for certain types of equations. The principal drawback, which is shared by any method that uses unstructured grids, is that the matrices of the linearized equations are not as well structured as those for regular grids making it more difficult to find efficient solution methods.

4.4 The Finite Difference Method Applied to Heat Transfer Problems:
In heat transfer problems, the finite difference method is used more often and will be discussed here in more detail. The finite difference method involves: Establish nodal networks Derive finite difference approximations for the governing equation at both interior and exterior nodal points Develop a system of simultaneous algebraic nodal equations Solve the system of equations using numerical schemes The Nodal Networks:

The basic idea is to subdivide the area of interest into sub-volumes with the distance between adjacent nodes by ∆x and ∆y as shown. If the distance between points is small enough, the differential equation can be approximated locally by a set of finite difference equations. Each node now represents a small region where the nodal temperature is a measure of the average temperature of the region. Example:

∆x

m,n+1

m-1,n ∆y

m,

m+1, n

x=m∆x, y=n∆y

m,n-1 m-½,n intermediate points

m+½,n

Finite Difference Approximation:
Heat Diffusion Equation: ∇2T + where α = & q 1 ∂T , = k α ∂t

k is the thermal diffusivity ρ C PV

∂ = 0, ⇒ ∇ 2T = 0 ∂t First, approximated the first order differentiation & No generation and steady state: q=0 and at intermediate points (m+1/2,n) & (m-1/2,n) ∂T ∆T ≈ ∂x ( m+1/ 2,n ) ∆x ∂T ∆T ≈ ∂x ( m−1/ 2,n ) ∆x =
( m +1/ 2,n )

Tm +1,n − Tm ,n ∆x Tm ,n − Tm −1,n ∆x

=
( m −1/ 2,n )

Next, approximate the second order differentiation at m,n ∂ 2T ∂x 2 ∂ 2T ∂x 2 ≈
m ,n

∂T / ∂x m +1/ 2,n − ∂T / ∂x m −1/ 2,n ∆x Tm+1,n + Tm −1,n − 2Tm ,n ( ∆x ) 2

m ,n

Similarly, the approximation can be applied to the other dimension y ∂ 2T ∂y 2 ≈
m ,n

Tm ,n +1 + Tm ,n −1 − 2Tm ,n ( ∆y ) 2

5) • Nodal relation table for exterior nodes (boundary conditions) can be found in standard heat transfer textbooks.4 + T4.n + Tm −1.n To model the steady state.3 + T3.n = 0 This equation approximates the nodal temperature distribution based on the heat equation. n=4.n ⎛ ∂ 2T ∂ 2T ⎞ + ⎜ ∂x 2 + ∂y 2 ⎟ ≈ ( ∆x ) 2 ( ∆y ) 2 ⎝ ⎠ m . • Derive one equation for each nodal point (including both interior and exterior points) in the system of interest.n + Tm −1. no generation heat equation: ∇2T = 0 This approximation can be simplified by specify ∆x=∆y and the nodal equation can be obtained as Tm+1. This approximation is improved when the distance between the adjacent nodal points is decreased: ∆T ∂T ∆T ∂T Since lim( ∆x → 0) = . For each node.C = ⎢ 2 ⎥ where A= M M M ⎥ ⎢ M ⎢M ⎥ ⎢ M ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎣ a N 1 a N 2 L a NN ⎦ ⎣TN ⎦ ⎣C N ⎦ .n Tm .4 + T3.5 . there is one such equation. Matrix Form The system of equations: a11T1 + a12T2 + L + a1N TN = C1 a21T1 + a22T2 + L + a2 N TN = C2 M M M M M a N 1T1 + a N 2T2 + L + a NN TN = CN A total of N algebraic equations for the N nodal points and the system can be expressed as a matrix formulation: [A][T]=[C] .n +1 + Tm .4T3. For example: for nodal point m=3.4 =0 T3.lim( ∆y → 0) = ∆x ∂x ∆y ∂y Table 4. The result is a system of N algebraic equations for a total of N nodal points.4 + T4. T = ⎢ 2 ⎥ .4 + T3.n −1 − 4Tm . no generation heat equation. ⎡ a11 a12 L a1N ⎤ ⎡ T1 ⎤ ⎡ C1 ⎤ ⎢a ⎥ ⎢T ⎥ ⎢C ⎥ a22 L a2 N ⎢ 21 ⎥ .4=(1/4)(T2.n + Tm . the equation is T2.n −1 − 2Tm .Tm +1.2 provides a list of nodal finite difference equation for various configurations.3 + T3.n +1 + Tm . A System of Algebraic Equations • The nodal equations derived previously are valid for all interior points satisfying the steady state.n − 2Tm .

n + Tm.Tm. n −1 + Tm +1.n = 0 k k ⎠ ⎝ 2(Tm −1. n = 0 T∞ − 2⎜ k ⎝ k ⎠ .n −1 ) + 2 h∆x ⎛ h∆x ⎞ + 2 ⎟Tm.n +1 + Tm .n +1 + Tm. n + Tm . n = 0 2(Tm −1. n +1 ) + (Tm +1.n −1 ) + 2 h∆x h∆x ⎞ ⎛ T∞ − 2⎜ 3 + ⎟Tm. n − 4Tm.n + Tm −1.n + Tm.

n −1 ) + 2 h∆x ⎛ h∆x ⎞ T∞ − 2⎜ + 1⎟Tm.n = 0 a +1 b +1 a(a + 1) b)b + 1) ⎝a b⎠ Case 5.nr. Node at an external corner with convection h. Replace (k) by (k-1) for the Jacobi iteration (Example : a31T1 + a32T2 + a33T3 + L + a1N TN = C1 .n a x x m. “Numerical Recipes” by Cambridge University Press or their web source at www.n b y m+1. iterative methods are usually more efficient.∞ 2(Tm −1. n +1 + Tm. Node near a curved surface maintained at a non uniform temperature Numerical Solutions Matrix form: [A][T]=[C]. For example.n m. [T]=[A]-1[C] where [A]-1 is the inverse of matrix [A].com. [T] is the solution vector. The famous Jacobi & Gauss-Seidel iteration methods will be introduced in the following. i = 3) Rewrite the equation of the form: N a C i −1 a Ti ( k ) = i − ∑ ij T j( k ) − ∑ ij T j( k −1) aii j =1 aii j =i +1 aii .n + Tm.Table 4. Iteration General algebraic equation for nodal point: ∑a T ij j =1 i −1 j + aiiTi + j =i +1 ∑aT ij N j = Ci .n = 0 k ⎝ k ⎠ m.n + Tm. • Matrix inversion requires cumbersome numerical computations and is not efficient if the order of the matrix is high (>10) • Gauss elimination method and other matrix solvers are usually available in many numerical solution package. • For high order matrix.n-1 Case 4. From linear algebra: [A]-1[A][T]=[A]-1[C]. n −1 + T1 + T2 − ⎜ + ⎟Tm.n y x m.n+1 y T1 m-1.n-1 T2 m.n 2 2 2 2 ⎛2 2⎞ Tm +1.2 Summary of nodal finite-difference methods m-1.

(k-1) means the present level and (k) represents the new level.• • • (k) . An initial guess (k=0) is needed to start the iteration. the new values at iteration (k) can be estimated The iteration will be stopped when max⎜Ti(k)-Ti(k-1) ⎟ε≤ .specify the level of the iteration. • By substituting iterated values at (k-1) into the equation. where ε specifies a predetermined value of acceptable error .

in metallurgy. quenching (rapid cool) can harden the strain boundary and increase strength. For example. On the other hand. Annealing (slow cool) can soften metals and improve ductility. the heat treating process can be controlled to directly affect the characteristics of the processed materials. or =∇ T ρc ∂t α ∂t k where α = is the thermal diffusivity ρc .UNSTEADY HEAT TRANSFER Many heat transfer problems require the understanding of the complete time history of the temperature variation. the full unsteady equation is needed: ∂T 1 ∂T 2 2 = k ∇ T . In order to characterize this transient behavior.

we are going to assume the temperature inside the solid is constant and equals to the surface temperature. These particles will eventually melt and impinge on the processing surface and solidify to form a layer of protecting coating. In other words. Let us look at a practical example about a plasma spray process involving the injection of tiny particles into a plasma jet at a very high temperature.Lumped Capacitance Method (LCM) The simplest situation in an unsteady heat transfer process is to neglect the temperature distribution inside the solid and only deals with the heat transfer between the solid and the ambient fluids. .

000 W/m2.K and specific heat cp=1560 J/kg. (b) determine the time for the particle to melt completely after it reaches the melting temperature. The melting temperature of the particle is 2318 K and the latent heat of fusion is 3577 kJ/kg.Example: (Plasma Spray) Assume spherical alumina particles are used in the plasma jet. (Special notes: why the particles follow the plasma jet? Does the particles travel at the same velocity as the local jet velocity? Does the jet has a uniform velocity? Energy balance: energy in = energy storage in solid h. T∞ T dE d ( mcPT ) = dt dt dT hAS (T∞ − T ) = ρ cPV dt wher AS is the surface area of the sphere hAS (T∞ − T ) = .000 K and has a convection coefficient of h=30. (diameter D=50 mm. thermal conductivity k=10. density ρ=3970 kg/m3.5 W/m.K. and an initial temperature of 300 K) The plasma is maintained at a temperature of 10. (a) Determine the time required to heat a particle to its melting point.

3t ) 3 −9700 ⎣ ⎝ 3970 * (1/ 6)π D *1560 ⎠ ⎦ T (t ) = 10000 − 9700exp( −581.) Define a new variable θ =T-T∞ dθ dθ hAS ρVcP dt = − hASθ .3t ) .Example (cont. =− dt θ ρVcP Integrate with respect to time and apply the initial condition T(t=0)=Ti =300(K) or θ (t=0)=θ i =Ti -T∞ =300-10000=-9700(K) ⎡ ⎛ hAS ⎞ ⎤ −hAS ln θ = t + C1 . θ (t ) = C2 exp ⎢ − ⎜ ⎟ t⎥ ρVcP ⎣ ⎝ ρVcP ⎠ ⎦ Substitute the initial condition: θ (0)=θ i = C2 The general solution for lumped capacitance method ⎡ ⎛ hAS ⎞ ⎤ θ (t) T (t ) − T∞ = = exp ⎢ − ⎜ ⎟ t⎥ Ti − T∞ θi ⎣ ⎝ ρVcP ⎠ ⎦ ⎡ ⎛ ⎞ ⎤ T (t ) − 10000 30000 * π D 2 = exp ⎢ − ⎜ ⎟ t ⎥ = exp( −581.

01 sec.Example (cont.0004 sec. (why?) 8000 T( t ) 6000 TT ( t ) 4000 2000 300 0 0 0 ? 0.0004 sec.01 4 • Temperature of the particle increases exponentially from 300 K to 10000K in a very short time (<0.006 0. the true temperature variation is described by the blue curve.) Temperature distribution as a function of time: T (t ) = 10000 − 9700exp( −581.) • It only takes 0. To reach the melting temperature • Therefore.01 t 0.004 0.01 0.002 0.008 0.3t ) 1 10 9971. 0.0009 sec. .

to melt the particle The total time to completely melt the particle will be 9 × 10-4 s.Example (cont. the heat input will not increase the temperature of the particle anymore. the heat will be absorbed by the solid particle as latent heat of melting in order for it to melt. t2 ∫q t1 conv dt ==∆Esf = mhsf hAS (T∞ − Tmelt )(t2 − t1 ) = ρVhsf 3970(50 × 10−6 )(3577000) t2 − t1 = = = 5 × 10−4 ( s ) hAS (T∞ − Tmelt ) 6(30000)(10000 − 2318) It will take an additional 5 × 10-4 s. ρVhsf .) After the particle reaches its melting temperature. Rather.

• Heat transfer from the exterior of the solid to the interior of the solid is through the conductive heat transfer: 4π k (TS .SPATIAL EFFECTS QUESTION: When can we neglect the temperature variation inside a solid? We can do that if the heat transfer inside the solid is much more effective than that occurs between the solid and the ambient fluid.2 ) q= (1/ r1 ) − (1/ r2 ) The thermal resistance is R sphere = 1 ⎛1 1⎞ ⎜ − ⎟ 4π k ⎝ r1 r2 ⎠ . If can be demonstrated as the following: • Heat transfer from the fluid to the solid through convection: q=hA(TS-T∞) and the thermal resistance of this process is Rconv=1/(hA).1 − TS .

LC = rO /2 (half-radius) when the solid is a cylinder. . defined as the Biot number) 1 >> ⎜ k ⎝ r1 r2 ⎠ k where LC is a characteristic length scale: relate to the size of the solid involved in the problem. LC = rO / 3 (one-third radius) when the solid is a sphere. the thermal resistance between the solid and the fluid should be much greater compared to the thermal resistance inside the solid. the temperature variation inside the solid should be much smaller than the temperature difference between the surface and the fluid: TS-T∞ << TS.1-TS.2. In other words. 1 1 1 1 ⎛1 1⎞ 1 = − ⎟ (of the order (of the order ) >> ) ⎜ 2 2 hA h(4π r2 ) 4π k ⎝ r1 r2 ⎠ kLC hLC 2 hr2 ⎛ 1 1 ⎞ hLC − ⎟ (of the order = BI . LC = L (half thickness) when the solid is a plane wall with a 2L thickness. For example.Biot Number In order to be able to use the lumped capacitance assumption. Rconv>>Rsolid.

Use temperature variation of the Alumina particles in a plasma jet process as an example: ⎡ ⎛ hAS ⎞ ⎤ ⎡ ⎛ hAS T (t ) − T∞ = exp ⎢ − ⎜ ⎟ t ⎥ = exp ⎢ − ⎜ Ti − T∞ ⎣ ⎝ ρVcP ⎠ ⎦ ⎣ ⎝ Vk 2 ⎞ ⎛ kt ⎞ ⎤ ⎥ ⎟⎜ ρ cP ⎟ ⎦ ⎠⎝ ⎠ AS 4π rO 3 1 = = = For a sphere: 4 3 rO LC V π rO 3 ⎡ ⎛ h ⎞ ⎛ kt ⎞ ⎤ ⎡ ⎛ hLC T (t ) − T∞ = exp ⎢ − ⎜ ⎟⎜ ⎟ ⎥ = exp ⎢ − ⎜ Ti − T∞ ⎣ ⎝ kLC ⎠ ⎝ ρ cP ⎠ ⎦ ⎣ ⎝ k ⎡ ⎛ α t ⎞⎤ = exp ⎢ − ( Bi ) ⎜ 2 ⎟ ⎥ = exp( − Bi ⋅ τ ) ⎝ LC ⎠ ⎦ ⎣ Define Fourier number (Fo): τ = ⎞ ⎛ kt ⎟⎜ 2 ⎠ ⎝ ρ cP LC ⎞⎤ ⎟⎥ ⎠⎦ αt L 2 C as dimensionless time .Biot and Fourier Numbers Define Biot number Bi=hLC/k In general. Bi<0.1 for the lumped capacitance assumption to be valid.

temperature is either a max.Spatial Effects Re-examine the plasma jet example: h=30. we need to solve the unsteady heat equation: 1 ∂ ⎛ 2 ∂T ⎞ 1 ∂T ⎟= ⎜ kr 2 r ∂r ⎝ ∂r ⎠ α ∂t with initial condition of T (r. if the diameter of the particle is increased to 1 mm. Therefore.5 W/m.476>0.K.1. then we have to consider the spatial effect since Bi=0.0238<0.1. or a min. ∂r ∂T = h(T (rO . Bi=(h/k)(rO/3)=0.000 W/m2. -k ∂r .K. and a diameter of 50 µm. t ) − T∞ ) at r = rO : conduction = convection at the surface. t = 0) = Ti and boundary conditions of the form: ∂T = 0 at r = 0: symmetric. To determine the temperature distribution inside the spherical particle. the use of the LCM is valid in the previous example. However. The surface temperature can not be considered as the same as the temperature inside the particle. k=10. at the center.

m . . Equations 9-10.12 show the one-term a pproximate solutions of the exact solutions for a plane wall. This approximation can also be represented in a graphical form as shown in Appendix D. a cylinder and a sphere respectively. Q where Q is the amount of heat transfer into (or out of) the solid Qmax Qmax is the maximum amount of energy transfer when t → ∞ Qmax = ρ c pV (Ti − T∞ ) or the energy to increase (or decrease) the overall temperature of the solid fro Ti to T∝ . (Sometimes they are called the Heisler charts) Terminology: θ 0 T0 − T∞ θ = = where T0 is the centre temperature.Example The solution can be determined using themethod of separation of variables. θ i Ti − T∞ * 0 T∝ is the ambient temperature.11. Ti is the initial temperature.

695×10-6(m2/s). or = exp( −0. Tmelt=2318 .K.49. Ti=300 K. 3 . τ = αt L 2 C =0. (a) Determine the time for the center temperature to reach the melting point.K. α=k/(ρcP)=1. hsf=3577 kJ/kg. we will use it anyway to compare the difference. T (t ) − T∞ T-10000 = exp( − Bi ⋅ τ ). First. However. First.008(s) rO since LC = for a spherical particle.5 W/m.476>0. ρ=3970 kg/m3. (b) Total time to melt the entire particle. k=10.Example (cont.476 ⋅ τ ) 300-10000 Ti − T∞ To reach T=2318 K. or t=0. (h=30000 W/m2. cP=1560 J/kg. T∞=10000 K. rO=0. Bi=(h/k)(rO/3)=0.5 mm.1 should not use LCM.) Recalculate the plasma jet example by assuming the diameter of the particles is 1 mm. use the lumped capacitance method and then compare the results to that determined using Heisler charts.

to melt the particle The total time to completely melt the particle will be 0. we need to use the Heisler chart: First.) After the particle reaches the melting point.792.01( s ) 6(30000)(10000 − 2318) It will take an additional 0. the heat transfer will be used to supply to the solid for phase transition t2 ∫q t1 conv dt ==∆Esf = mhsf .01 s.018 s.7 (Note : this definition is different from hrO the previous Biot number used to validate the lumped capacitance method) * θO = TO − T∞ 2318 − 10000 = = 0.001)(3577000) = 0. To consider the spatial effects. Bi-1 = k = 0. hAS (T∞ − Tmelt )(t2 − t1 ) = ρVhsf t2 − t1 = ρVhsf hAS (T∞ − Tmelt ) = 3970(0. Now look it up in figure 9-15 300 − 10000 Ti − T∞ .Example (cont.

792 Note: this is the magnified upper left corner of Fig. D-1 TO − T∞ Ti − T∞ t*=0.16=αt/rO t=(rO2/α)(0.7 t*=τ .023(s) much longer than the value calculated using LCM (t=0.16)=0.008 s) 2 Bi-1=0.Heisler Chart θO*=0.

173 2 λ1 ⎝ A1 ⎠ (1.768) ⎝ 1.Approximate Solution We can also use the one-term approximation solution in equation 9-12 to calculate. A1 = 1.768.43 through interpolation. Determine the constants in Table 9.1: for Bi=1.16) estimated from using the chart −1 .792 ⎞ τ = 2 ln ⎜ ln ⎜ = ⎟ ⎟ = 0.362. r* = ( r / rO ) λ1r * 2 1 θ O * = A1 exp( −λ12τ ) at the center ⎛ θO * ⎞ −1 ⎛ 0. 1 θ * = A1 exp( −λ τ ) sin(λ1r*).362 ⎠ It is close to the value (0. λ1 = 1.

173) = 0. Bi 2τ = (1.48 Bi=1.001)2 (300 − 10000) = 31.428) 2 (0.353 0. 438( J ) Bi = 1.Heat Transfer How much heat has been transfered into the particle during this period of time? Determine Q max = ρ cPV (Ti − T∞ ) = (3970)(1560)(1/ 6)π (0.428 2345 0.428.353 68 .

• To make matter even more complicated.) Total heat transfer during this period: Q=Qmax(Q/Qmax)=31438(0.48)=15090(J) • The actual process is more complicated than our analysis. the temperature will not increase once the solid reaches the melting temperature unless the solid melts into liquid form. First. Therefore. • No analytical solutions or numerical analysis is necessary. . the interface between the melt alumina and the solid is continuously moving inward. • If the outer layer melts then we have double convective conditions: convection from the plasma gas to the liquid alumina and then from there to the inner solid.Heat Transfer (cont. the actual heat transfer process will probably slower than the estimation.

As we discuss earlier. molten materials will attach to the substrate and cool down rapidly. thermal spray process deposits thin layer of coating materials on surface for protection and thermal resistant purposes. The heated.Unsteady Heat Transfer in Semi-infinite Solids Solidification process of the coating layer during a thermal spray operation is an unsteady heat transfer problem. as shown. α . latent heat of fusion: hsf S(t) solid δ Substrate. k. The cooling process is important to prevent the accumulation of residual thermal stresses in the coating layer. liquid Coating with density ρ.

the cooling process can now be modeled as heat loss through a semi-infinite solid. h sf to the substrate dt (q=q”A) where S is solidified thickness ρ dS = q" dt .Example As described in the previous slide. (Since the substrate is significantly thicker than the coating layer) The molten material is at the fusion temperature Tf and the substrate is maintained at a constant temperature Ti. From energy balance: Heat transfer from hsf ∆m (solidified mass during ∆t) = ∆Q = q" A∆t (energy input) the molten material dm = q" A. The heat loss to the substrate is solely supplied by the release of the latent heat of fusion. Derive an expression for the total time that is required to solidify the coating layer of thickness δ. • Assume the molten layer stays at a constant temperature Tf throughout the process. where m = ρV = ρAS.

However. it can be modelled as constant surface temperature case by setting h=∞. If the surface temperature is Ts and the initial temperature of the bolck is Ti . Ts=T∞). the analytical solution of the problem can be found: The temperature distribution and the heat transfer into the block are: ⎛ x ⎞ ⎜ ⎟ .) Identify that the previous situation corresponds to the case discussed in chapter 9-3 in the text as a semi-infinite transient heat transfer problem with a constant surface temperature boundary condition (note: the case in the textbook corresponds to an external convection case. where erf( ) is the Gaussian error function. ⎝ 2 αt ⎠ 2 w − v2 It is defined as erf(w)= ∫ e dv T(x.Example (cont.t)-Ts = erf Ti − Ts π 0 k(Ts − Ti ) qs"(t)= παt . therefore.

and hsf=3. Tf=2318 K. Cooling time t = 2 ⎜ δ (t ) = 4k ⎜ T f − Ti ρ hsf πα ⎝ Use the following values to calculate: k=120 W/m. Ti=300K.K.) From the previous equation k(Tf − Ti ) k(Tf − Ti ) dt dS =q"= . and δ=2 mm ⎞ ⎟ ⎟ ⎠ 2 . δ ∝ t. therefore. α=4×10-5 m2/s. and ∫ dS = ρ hsf dt πα t ρ hsf πα ∫ t 0 0 t δ 2k(Tf − Ti ) πα ⎛ δρ hsf t .577 ×106 J/kg.Example (cont. ρ=3970 kg/m3.

.00304 t δ (t ) = ρhsf πα 0. it takes 0.8 1 • δ(t) ∝ t1/2 • Therefore.43 seconds.003 δ( t )0.) 2k(Tf − Ti ) t = 0. the layer solidifies very fast initially and then slows down as shown in the figure • Note: we neglect contact resistance between the coating and the substrate and assume temperature of the coating material stays the same even after it solidifies.2 0.6 0.Example (cont.004 0.001 0 0 0.002 0. • To solidify 2 mm thickness.4 t 0.

Example (cont.)
What will be the substrate temperature as it varies in time? The temperature distribution is:
T ( x, t ) − TS = erf Ti − TS ⎛ x ⎞ ⎜ ⎟, ⎝ 2 αt ⎠ ⎞ ⎟ = 2318 − 2018erf ⎠ x ⎞ ⎛ 79.06 ⎟ ⎜ t⎠ ⎝

⎛ x T ( x, t ) = 2318 + (300 − 2318)erf ⎜ ⎝ 2 αt

Example (cont.)
• For a fixed distance away from the surface, we can examine the variation of the temperature as a function of time. Example, 1 cm deep into the substrate the temperature should behave as:

x 0.79 = 2318 − 2018erf T ( x = 0.01, t ) = 2318 − 2018erf 79.06 t t
2000 1600 Temperature T1( t ) 1200 T2( t ) 800 T3( t ) 400 0 0 2 4 t Time 6 8 10

x=1 cm x=2 cm x=3 cm

• At x=1 cm, the temperature rises almost instantaneously at a very fast rate. A short time later, the rate of temp. increase slows down significantly since the energy has to distribute to a very large mass. • At deeper depth (x=2 & 3 cm), the temperature will not respond to the surface condition until much later.

Example (cont.)
We can also examine the spatial temperature distribution at any given time, say at t=1 second. x = 2318 − 2018erf 79.06 x T ( x, t = 1) = 2318 − 2018erf 79.06 t
3000 Temperature (K)

T1( x ) T2( x )

2000

T3( x ) 1000

0 0 0.01 0.02 0.03 0.04 0.05 x distance (m)

t=1 s. t=5 s. t=10 s.

• Heat penetrates into the substrate as shown for different time instants. • It takes more than 5 seconds for the energy to transfer to a depth of 5 cm into the substrate • The slopes of the temperature profiles indicate the amount of conduction heat transfer at that instant.

Numerical Methods for Unsteady Heat Transfer
Unsteady heat transfer equation, no generation, constant k, twodimensional in Cartesian coordinate:

1 ∂T ∂ 2T ∂ 2T = 2 + 2 α ∂t ∂x ∂y We have learned how to discretize the Laplacian operator into system of finite difference equations using nodal network. For the unsteady problem, the temperature variation with time needs to be discretized too. To be consistent with the notation from the book, we choose to analyze the time variation in small time increment ∆t, such that the real time t=p∆t. The time differentiation can be approximated as:
∂T ∂t
+ TmP,n1 − TmP,n , while m & n correspond to nodal location ≈ ∆t

m ,n

such that x=m∆x, and y=n∆y as introduced earlier.

Finite Difference Equations
m,n+1 m-1,n m,n m+1, n
From the nodal network to the left, the heat equation can be written in finite difference form:

m,n-1 P +1 P P P P P P P 1 Tm ,n − Tm ,n Tm +1,n + Tm −1,n − 2Tm ,n Tm ,n +1 + Tm ,n −1 − 2Tm ,n = + 2 ∆t α ( ∆x ) ( ∆y ) 2 α∆t Assume ∆x=∆y and the discretized Fourier number Fo= 2 ( ∆x )
+ TmP,n1 = Fo (TmP+1,n + TmP−1,n + TmP,n +1 + TmP,n −1 ) + (1 − 4 Fo)TmP,n

This is the explicit, finite difference equation for a 2-D, unsteady heat transfer equation. The temperature at time p+1 is explicitly expressed as a function of neighboring temperatures at an earlier time p

Nodal Equations
Some common nodal configurations are listed in table for your reference. On the third column of the table, there is a stability criterion for each nodal configuration. This criterion has to be satisfied for the finite difference solution to be stable. Otherwise, the solution may be diverging and never reach the final solution. For example, Fo≤1/4. That is, α∆t/(∆x)2 ≤1/4 and ∆t≤(1/4α)(∆x)2. Therefore, the time increment has to be small enough in order to maintain stability of the solution. This criterion can also be interpreted as that we should require the coefficient for TPm,n in the finite difference equation be greater than or equal to zero. Question: Why this can be a problem? Can we just make time increment as small as possible to avoid it?

by specifying initial conditions for all points inside the nodal network. p+3. • Important: check stability criterion for each points. •From the explicit equation. The following transient response can then be determined by marching out in time p+2. and so on. . we can determine all temperature at the next time level p+1=0+1=1. That is to specify values for all temperature at time level p=0.Finite Difference Solution Question: How do we solve the finite difference equation derived? • First.

it satisfies the case 1 configuration in table.1(T1P + T3P ) + 0. Fo=(α∆t)/(∆x)2=0.K. is suddenly immersed into a cold temperature bath of 0 deg. ∆x=0.Example Example: A flat plate at an initial temperature of 100 deg.01 m x Bi=(h∆x)/k=1. h=1000 W/m2. T2P +1 = Fo(T1P + T3P + T2P + T2P ) + (1 − 4 Fo)T2P = Fo(T1P + T3P ) + (1 − 2 Fo)T2P = 0. T∞=0°C.1 1 2 3 There are three nodal points: 1 interior and two exterior points: For node 2. k=10 W/m. Ti=100°C. Use the unsteady finite difference equation to determine the transient response of the temperature of the plate.it is satisfied 2 . α=10×10-6 m2/s.1 ≤ .02 m.8T2P 1 Stability criterion: 1-2Fo ≥ 0 or Fo=0.K. L(thickness)=0.

6T3P 1 ≥ Fo(1 + Bi ) = 0.1(T11 + T31 ) + 0. T2 = 100 T12 = 0.1(T10 + T30 ) + 0.1(T1P + T3P ) + 0.8T20 = 100 T31 = 0.2T2P + 0.6T30 = 80 1 1 Marching in time. System of equations T1P +1 = 0.2T2P + 0.1(80 + 80) + 0.2T21 + 0.2T21 + 0.6T31 = 0. T11 = T3 = 80.2(100) + 0. T10 = T20 = T30 = 100.8(100) = 96 T32 = 0.2T2P + 0.2T20 + 0.6T10 = 80 T21 = 0. Node 1: T1P +1 = Fo(2T2P + T1P + T1P + 2 BiT∞ ) + (1 − 4 Fo − 2 BiFo)T1P = Fo(2T2P + 2 BiT∞ ) + (1 − 2 Fo − 2 BiFo)T1P = 0.6T3P Stability criterion: (1-2Fo-2BiFo) ≥ 0.Example For nodes 1 & 3. and so on .2T20 + 0.8T2P T3P +1 = 0.6(80) = 68.6T1P T2P +1 = 0.6T1P Node 3: T3P +1 = 0.6T11 = 0.2T2P + 0.2 and it is satisfied 2 Use initial condition. they are consistent with the case 3 in table. T11 = 0.8T21 = 0.6(80) = 68 T22 = 0.2(100) + 0.

0.5) = T − T1 2 = T2 − T1 π ∑ n =1 ∞ nπ sinh( nπ / L) ( −1) n +1 + 1 sin . the corresponding term is zero. 0.3.5) and recognizing x/L =1/2. n L sinh( nπ W / L) Considering now the point (x. y ) = T − T1 2 = T2 − T1 π ∑ n =1 ∞ nπ x sinh( nπ y / L) ( −1) n +1 + 1 sin . the temperature distribution is θ ( x. hence we need only consider n=1.4.. y/L=1/4 and W/L=1/2. the distribution has the form θ (1.5) by considering the first five nonzero terms of the infinite series that must be evaluated. calculate the temperature at the midpoint (1. Analysis: From analytical solution. .0. y) = (1. Schematic: Assumptions: (1) Two-dimensional.6…. 0. Assess the error from using only the first three terms of the infinite series. (2) constant properties. steady-state conduction. Find: temperature at the midpoint using the exact solution considering the first five nonzero terms: Assess the error from using only the first three terms.).Module 4: Worked out problems Problem 1: A two dimensional rectangular plate is subjected to the uniform temperature boundary conditions shown. Using the results of the analytical solution for the heat equation. Known: Two-dimensional rectangular plate to prescribed uniform temperature boundary conditions. n L sinh( nπ / L) When n is even (2.7 and 9 as the first five non-zero terms.5.

5°C If only the first term of the series. 0.5) = θ(1.5) = ⎨ ⎬ 7π 9π ⎪ π⎪ sinh sinh 2 ⎛ 7π ⎞ ⎪ 4 + 2 sin⎛ 9π ⎞ 4 ⎪ sin⎜ ⎟ ⎜ ⎟ ⎪ 7 ⎝ 2 ⎠ sinh 7π 9 ⎝ 2 ⎠ sinh 9π ⎪ ⎪ ⎪ 2 2 ⎭ ⎩ θ (1.000] = 0. . it follows that T (1.008 − 0.446 that is.001 + 0. 0.445(150-50) +50=94.2% effect.5) =0.445 Since θ = (T-T1)/ (T2-T1).3π 5π ⎫ π ⎧ sinh sinh sinh ⎪ ⎛π ⎞ 4 + 2 sin⎛ 3π ⎞ 4 + 2 sin⎛ 5π ⎞ 4 +⎪ ⎜ ⎟ ⎜ ⎟ ⎪2 sin⎜ ⎟ 3π 5 ⎝ 2 ⎠ 5π ⎪ π 3 ⎝ 2 ⎠ ⎝ 2 ⎠ sinh ⎪ ⎪ sinh sinh 2⎪ 2 2 2 ⎪ θ (1.755 + 0. 0.0. there is less than a 0. the result will be θ(1. Eq (2) is considered.0.5) (T2-T1) +T1=0.5) = 2 π [0.063 + 0.

K) is at 300K. and to render the cable superconducting (essentially zero power dissipation). (2) constant properties.Problem 2: A long power transmission cable is buried at a depth (ground to cable centerline distance) of 2m.005W/m.1 m diameter.K) of 0. Analysis: The heat rate per unit length is q' = q' = T g − Tn ' R g + R I' T g − Tn [k g ( 2π / ln(4 z / Do ))] −1 + ln( Do / Di ) / 2π k i where table 4. (4) one-dimensional conduction in insulation. Find: required cooling load. (3) two-dimensional conduction in soil.2W/m. Known: Operating conditions of a buried superconducting cable.05-m thickness and the surface of the earth (kg=1. The cable is encased in a thin walled pipe of 0.1 have been used to evaluate the ground resistance. . Schematic: Assumptions: (1) steady-state conditions. the space between the cable and pipe is filled with liquid nitrogen at 77 K. Hence. what is the cooling load in W/m which must be maintained by a cryogenic refrigerator per unit pipe length. If the pipe is covered a super insulator(ki=0.

2))] −1 + ln(2) / 2π × 0.005 W / m.K 223K q' = (0.9W / m q' = (300 − 77) K Comments: the heat gain is small and the dominant contribution to the thermal resistance is made by the insulation.2W / m.K / W q ' = 9.064)m.[(1.K ) (2π / ln(8 / 0. .489 + 22.

075m) 2 S = 2π / cosh −1 = 2π / cosh −1 (65.075m S = 2π / 4.29 × 0. respectively.1m) 2 − (0. (4) pipe lines are buried very deeply approximating burial in an infinite medium.5W/m. l hence.5W / m. Schematic: Assumptions: (1) steady state conditions. Known: Surfaces temperatures of two parallel pipe lines buried in soil. the pipes have outer-diameters of 100 and 75 mm with surface temperatures of 175°C and 5°C.1 as S= 2πL 2 ⎛ 4w 2 − D12 − D 2 cosh −1 ⎜ ⎜ 2 D1 D 2 ⎝ ⎞ ⎟ ⎟ ⎠ Substituting numerical values 4 × (0.K (175 − 5)°C = 110W / m .K. (3) constant properties.29. Find: heat transfer per unit length between the pipe lines. Estimate the heat transfer rate per unit length between the two pipe lines. (5) pipe length>> D1 or D2 and w>D1 or D2 Analysis: the heat transfer rate per length from the hot pipe to the cool pipe is q' = q S = k (T1 − T2 ) L L The shape factor S for this configuration is given in table 4.5m) 2 − (0. the heat rate per unit length is q ' = 1.88 = 1.Problem 3: Two parallel pipelines spaced 0.5 m apart are buried in soil having a thermal conductivity of 0.63) L 2 × 0.1m × 0. (2) two-dimensional conduction.

How would you estimate the heat gain if the soil were at 25°C? .Comments: The heat gain to the cooler pipe line will be larger than 110W/m if the soil temperature is greater than 5°C.

1W / m. Properties: From table of properties. Schematic: Assumptions: (1) steady-state conditions.1.14m S C = 0.25m L 0.15L = 0. q(W). Known: Cubical furnace. with external dimensions of 0. with 50mm thick walls.52 × 0.25 × 0.25m 2 = = 1.05m − S E = 0. calculate the heat loss from the furnace. 350mm external dimensions. Plane walls (6) Edges (12) Corners (8) SW = A 0. Find: The heat loss. (3) constant properties.Problem 4: A furnace of cubical shape. and the outer surface temperature is 75°C.K Analysis: using relations for the shape factor from table 4. If the wall thickness is 50mm. is constructed from a refractory brick (fireclay). fireclay brick ( T = (T1 + T2 ) / 2 = 610 K ) : k ≈ 1. (2) two-dimensional conduction.05m = 0. the inner surface temperature is 600°C.25m = 0.15 × 0.008m The heat rate in terms of the shape factor is .54 D = 0.35m.

30kW Comments: Be sure to note that the restriction for SE and SC has been met.25m + 12 × 0.K q = 5.q = kS (T1 − T2 ) = k (6S W + 12 S E + 8S C )(T1 − T2 ) q = 1.15 × 0.14m + 0.1 W (6 × 1. .008m)(600 − 75)°C m.

1⎟ (T∞ − Tm. h) (b) Both boundaries of external corner are perfectly insulated. (2) two-dimensional conduction. Known: External corner of a two-dimensional system whose boundaries are subjected to prescribed conditions. (4) no internal generation.2. With the upper boundary insulated and the side boundary and the side boundary subjected to a convection process. The heat transfer processes at the surface of the CV are identified as q1.1⎟ ∆y ∆x ⎝ 2 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠ . Schematic: Assumptions: (1) steady-state conditions. q2…. Derive the finite-difference equation under steady-state conditions for the following situations. .perform an energy balance wherein the processes are expressed using the appropriate rate equations. the energy balance has the form E in − E out = 0 q1 + q 2 + q 3 + q 4 = 0 Tm.n ⎛ ⎛ ∆y ⎞ ⎛ ∆x ⎞ k⎜ .n ) + 0 = 0 + k ⎜ .1. (3) constant properties. (a) The upper boundary of the external corner is perfectly insulated and the side boundary is subjected to the convection process (T∞. table 4. (∆x/2)( ∆y/2). (b) both boundaries are perfectly insulated.n ∆y ⎞ Tm −1. the control volume about the nodes shaded area above of unit thickness normal to the page has dimensions.1⎟ + h⎜ . Find: finite-difference equations for these situations: (a) upper boundary is perfectly insulated and side boundary is subjected to a convection process. n −1 − Tm. Analysis: Consider the nodal point configuration shown in schematic and also as case 4.Problem 5: Consider nodal configuration 4 of table 4. n − Tm.2.

The result is Tm. n − 2Tm. the energy balance of Eq(2) would have q3=q4=0. the same result would be obtained by letting h=0 in the finite difference equation. and regrouping.n = 0 Comments: Note the convenience resulting formulating the energy balance by assuming that all the heat flow is into the node.n = 0 k ⎝2 k ⎠ with both boundaries insulated. . find Tm.Letting ∆x=∆y.n −1 + Tm −1.n −1 + Tm −1. Eq(3).n + h∆x ⎛ 1 h∆x ⎞ T∞ − 2⎜ + 1⎟Tm.

will these two methods result in the same or a different set of equations? 6. What is the basis of energy balance method in numerical analysis? How does it differ from the formal finite difference method using Taylor series approximation? For a specified nodal network. two-. Consider a medium in which the finite difference formulation of a general interior node is given in its simplest form as & Tm −1 − 2Tm + Tm +1 g m + =0 2 ∆x k (a) Is heat transfer in this medium steady or transient? (b) Is heat transfer one-. When performing numerical calculations of heat diffusion on a structured Cartesian grid in two dimensions. For what kind of problems in multidimensional heat transfer are analytical solutions possible? Name some common analytical methods in steady state multidimensional heat transfer? 2. two-. What is an irregular boundary? What is a practical way of handling irregular boundary surfaces with the finite difference method? 9. Consider a medium in which the finite difference formulation of a general interior node is given in its simplest form as & g l2 Tleft + Ttop + Tright + Tbottom − 4Tnode + m = 0 k (a) Is heat transfer in this medium steady or transient? (b) Is heat transfer one-. why are analytical solutions useful? 3. How do numerical solution methods differ from analytical methods? What are the advantages and disadvantages of numerical and analytical methods? 5. or three-dimensional? (c) Is there heat generation in the medium? (d) Is the nodal spacing constant or variable? (e) Is thermal conductivity of the medium constant or variable? 7. What are the limitations of analytical methods? In spite of their limitations. a simplified form of the equations states that the . What is meant by “shape factor” in two-dimensional heat transfer analysis? What is the advantage of using such a method? Is there a shape factor in 1D heat transfer? 4.Module 4: Short questions 1. or three-dimensional? (c) Is there heat generation in the medium? (d) Is the nodal spacing constant or variable? (e) Is thermal conductivity of the medium constant or variable? 8.

What assumption is NOT required to allow this simplified form a) must have no heat generation b) must not be at a domain boundary c) must have uniform cell dimensions in both directions d) must be a solid medium .temperature at a node is simply the average of its four adjacent neighbours.

he/she may use the lumped capacitance method to obtain accurate results with minimum computational requirements. recourse must be made to an approximate numerical technique. The simplest case is the lumped capacity condition. spatial effects must be considered and some other method must be used. • • • . if the Biot number is not much less than unity. and the semi-infinite solid. If this number is much less than unity. such as the finite difference method or the finite volume method. and the student should understand under what conditions one can apply this assumption. The student should learn to recognize several classes of problems in unsteady conduction. the finite cylinder.Module 5: Learning objectives • The primary objective of this chapter is to learn various methods of treating transient conduction which occurs in numerous engineering applications. the sphere. However. The student should know when and how to use these results. and should be able to apply appropriate simplicity conditions before attempting to solve the problems. Analytical results are available in convenient graphical and equation form for the plane wall. If geometrical complexities and /or the form of the boundary conditions preclude the use of analytical solutions. The first thing a student should do is calculate the Biot number.

the heat treating process can be controlled to directly affect the characteristics of the processed materials. A more meaningful approach is to consider the general problem of transient cooling of an object. also be internal – e. Boundary conditions. 5.MODULE 5 UNSTEADY STATE HEAT CONDUCTION 5. Annealing (slow cool) can soften metals and improve ductility. we have considered conductive heat transfer problems in which the temperatures are independent of time. are frequently specified along the physical boundary of an object. the internal temperature gradients in the body may be quite small and insignificant. in metallurgy.1) k is the thermal diffusivity.1 Introduction To this point. we need two boundary conditions in x-direction and one initial condition. In many applications. Ts > T∞ Fig. 5.2 Biot and Fourier numbers In some transient problems. quenching (rapid cool) can harden the strain boundary and increase strength.1) we can note that such could be the case for large thermal diffusivity α . Yet the temperature at a given location. however. or the average temperature of the object. the full unsteady equation is needed: 1 ∂T ∂ 2 T ∂ 2 T ∂ 2 T q ⋅ = + + + a ∂τ ∂x 2 ∂y 2 ∂z 2 k where α = L (5.1. Without any heat generation and considering spatial ρc variation of temperature only in x-direction. On the other hand. For example. From eq. the above equation reduces to: 1 ∂T ∂ 2T ⋅ = a ∂τ ∂x 2 (5. such as the hollow cylinder shown in figure 5. In order to characterize this transient behavior.2) For the solution of equation (5. they can. the temperatures are varying with time. (5.1 . may be changing quite rapidly with time. however.g.2). and we require the understanding of the complete time history of the temperature variation. as the name implies. a known temperature gradient at an internal line of symmetry.

for example a plate.3) and (5. The rate of heat transfer away from the outer surface by convection is q = h ( 2πro l )(Ts − T∞ ) (5. a characteristic length scale may be obtained by dividing the volume of the solid by its surface area: V (5.5) Bi = resistance to internal heat flow resistance to external heat flow Whenever the Biot number is small.For very large ri.7) . lumped capacity assumption is applied. The lumped capacity assumption implies that the object for analysis is considered to have a single massaveraged temperature. the internal temperature gradients are also small and a transient problem can be treated by the “lumped thermal capacity” approach. In general. if the Biot number turns out to be less than 0. Equating (5. In the derivation shown above. In this context. the corresponding Biot number may be evaluated for objects of any shape.3) q ≈ −k ( 2πro l )⎜ s ⎜ r − r ⎟ = k ( 2πro l )⎜ L ⎟ ⎟ ⎝ ⎠ ⎝ o i ⎠ where l is the length of the cylinder and L is the material thickness.1. L = ro − ri .4) where h is the average heat transfer coefficient for convection from the entire surface. the significant object dimension was the conduction path length. can be obtained by multiplying the dimensional time by the thermal diffusivity and dividing by the square of the characteristic length: dimensionless time = αt L2 = Fo (5. a dimensionless time. a cylinder. the heat transfer rate by conduction through the cylinder wall is approximately ⎛ T − Ti ⎞ ⎛ Ti − Ts ⎞ (5. As a thumb rule. known as the Fourier number. and it can be thought of as the ratio (5.6) L= As Using this method to determine the characteristic length scale.4) gives Ti − Ts h L = Biot number = Ts − T∞ k The Biot number is dimensionless. or a sphere.

3 Lumped thermal capacity analysis The simplest situation in an unsteady heat transfer process is to use the lumped capacity assumption. T (t ) − T∞ = e −bt Ti − T∞ where h As b= (1/s) ρcV ln (5. c.9) Integrating and applying the initial condition T (0) = Ti . q = h As (T − T∞ ) Solid T(t) ρ.2 The solid object shown in figure 5.12) . Thermal energy is leaving the object from all elements of the surface. 5. V T∞ h Fig. we are assuming that the temperature inside the solid is constant and is equal to the surface temperature.11) (5. hA T (t ) − T∞ =− s t ρcV Ti − T∞ Taking the exponents of both sides and rearranging.8) or. In other words.2 is a metal piece which is being cooled in air after hot forming.10) (5. wherein we neglect the temperature distribution inside the solid and only deal with the heat transfer between the solid and the ambient fluids. (5. this equation can be written as h As [T (t ) − T∞ ]dt = − ρcVdT h As dT =− dt (T − T∞ ) ρcV (5. The first law of thermodynamics applied to this problem is ⎛ heat out of object ⎞ ⎛ decrease of internal thermal ⎞ ⎜ ⎟=⎜ ⎜ during time dt ⎟ ⎜ energy of object during time dt ⎟ ⎟ ⎝ ⎠ ⎝ ⎠ Now. and this is shown for simplicity by a single arrow.5. if Biot number is small and temperature of the object can be considered to be uniform.

variations in temperature during the transient process. As an exercise. Question: What is the significance of b? Answer: According to eq. the temperature changes rapidly in the beginning. plot T vs. and then slowly.Note: In eq. the heat equation and initial/boundary conditions are: 1 ∂T ∂ 2T ⋅ = a ∂ τ ∂x 2 T ( x. t ) − T∞ ] x =l . which has the dimension of time. In other words. Rate of convection heat transfer at any given time t: & Q (t ) = hAs [T (t ) − T∞ ] Total amount of heat transfer between the body and the surrounding from t=0 to t: Q = mc [T (t ) − Ti ] Maximum heat transfer (limit reached when body temperature equals that of the surrounding): Q = mc [T∞ − Ti ] 5.12. A larger value of b indicates that the body will approach the surrounding temperature in a shorter time. You can visualize this if you note the variables in the numerator and denominator of the expression for b. 5.11. The Plane Wall: Solution to the Heat Equation for a Plane Wall withSymmetrical Convection Conditions • For a plane wall with symmetrical convection conditions and constant properties.4 Spatial Effects and the Role of Analytical Solutions If the lumped capacitance approximation can not be made. the temperature of a body approaches the ambient temperature T∞ exponentially. 0) = Ti ∂T ∂x −k =0 x =0 ∂T ∂x = h[T ( L. 5. The reciprocal of b is usually called time constant. b is a positive quantity having dimension (time)-1. consideration must be given to spatial. t for various values of b and note the behaviour. as well as temporal.

there is existence of seven different independent variables.Note: Once spatial variability of temperature is included. Cn = . α ) How may the functional dependence be simplified? • The answer is Non-dimensionalisation. t . k . h. We first need to understand the physics behind the phenomenon. θ T − T∞ Dimensionless temperature difference: θ * = = θ i Ti − T∞ x Dimensionless coordinate: x* = L αt * Dimensionless time: t = 2 = Fo L hL Bi = The Biot Number: k solid The solution for temperature will now be a function of the other non-dimensional quantities θ * = f ( x * . identify parameters governing the process. Bi ) Exact Solution: θ * = ∑ C n exp(− ζ n2 Fo )cos(ζ n x * ) ∞ n =1 4 sin ζ n ζ n tan ζ n = Bi 2ζ n + sin(2ζ n ) The roots (eigenvalues) of the equation can be obtained from tables given in standard textbooks. T∞ . Ti . T = T ( x. Fo. Non-dimensionalisation of Heat Equation and Initial/Boundary Conditions: The following dimensionless quantities are defined. and group them into meaningful non-dimensional numbers.

one can obtain C1 and ζ 1 as a function of Bi.The One-Term Approximation Fo > 0. Variation of temperature with location ( x * ) and time ( Fo ): θ * = θ 0* = cos(ζ 1 x * ) Change in thermal energy storage with time: ∆E st = −Q ⎛ sin ζ 1 ⎞ * ⎟θ Q = Q0 ⎜1 − ⎜ ζ1 ⎟ 0 ⎝ ⎠ Q0 = ρcV (Ti − T∞ ) Can the foregoing results be used for a plane wall that is well insulated on one side and convectively heated or cooled on the other? Can the foregoing results be used if an isothermal condition (Ts ≠ Ti ) is instantaneously imposed on both surfaces of a plane wall or on one surface of a wall whose other surface is well insulated? Graphical Representation of the One-Term Approximation: The Heisler Charts Midplane Temperature: .2 * Variation of mid-plane ( x = 0) temperature with time ( Fo ) T − T∞ θ 0* = ≈ C1 exp(− ζ 12 Fo ) Ti − T∞ From tables given in standard textbooks.

Temperature Distribution Change in Thermal Energy Storage § § Assumptions in using Heisler charts: l Constant Ti and thermal properties over the body l Constant boundary fluid T∞ by step change l Simple geometry: slab.2) . cylinder or sphere Limitations: l Far from edges l No heat generation (Q=0) l Relatively long after initial times (Fo >0.

5.Radial Systems Long Rods or Spheres Heated or Cooled by Convection Bi = hr0 / k Fo = αt / r02 Similar Heisler charts are available for radial systems in standard text books. Important tips: Pay attention to the length scale used in those charts.5 Numerical methods in transient heat transfer: The Finite Volume Method Considering the steady convection-diffusion equation: ∂ ( ρφ ) + div( ρφ u) = div (Γ grad φ ) + Sφ ∂t The time and control volume integrations give: t + ∆t t + ∆t t + ∆t ⎞ ⎛ ⎞ ⎛ t + ∆t ∂ ⎞ ⎛ ⎞ ⎛ ⎜ ∫ (ρφ )dt ⎟dV + ∫ ⎜ ∫ n ⋅ (ρφ u )dA ⎟dt = ∫ ⎜ ∫ n ⋅ (Γ grad φ )dA ⎟dt + ∫ ⎜ ∫ Sφ dV ⎟dt ∫ ⎜ t ∂t ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ CV ⎝ t ⎝A t ⎝A t ⎝ CV ⎠ ⎠ ⎠ ⎠ • Unsteady one-dimensional heat conduction: ∂T ∂ ⎛ ∂T ⎞ = ⎜k ρc ⎟+S ∂t ∂x ⎝ ∂x ⎠ t W (δx)WP w P ∆x (δx)PE e E x . and calculate your Biot number accordingly.

Consider the one-dimensional control volume. TE and Tw with time. Integration over the control volume and over a time interval gives: t + ∆t t + ∆t t + ∆t ⎛ ∂ ⎛ ∂T ⎞ ⎞ ⎛ ⎞ ⎛ ⎛ ∂T ⎞ ⎞ ⎜ ∫ ⎜ ρc ⎟dV ⎟dt + ∫ ⎜ ∫ SdV ⎟dt ⎟dV ⎟dt = ∫ ⎜ ∫ ⎜ k ∫ ⎜ CV ⎝ ∂t ⎠ ⎟ ⎜ ∂x ⎝ ∂x ⎠ ⎟ ⎜ ⎟ t ⎝ t ⎝ cv t ⎝ CV ⎠ ⎠ ⎠ Re-written e t + ∆t t + ∆t t + ∆t ⎛ ⎛ ∂T ⎞ ∂T ∂T ⎞ ⎜ ∫ ρc ⎟dV = ∫ ⎜ ⎛ kA ⎞ − ⎛ kA ⎞ ⎟dt + ∫ (S ∆V )dt ∫ ⎜ t ∂t dt ⎟ ⎜ ⎜ ∂x ⎟ ⎜ ∂x ⎟ ⎟ ⎠e ⎝ ⎠w ⎠ w⎝ t ⎝⎝ t ⎠ If the temperature at a node is assumed to prevail over the whole control volume. we have: t + ∆t t + ∆t ⎛⎛ T − TW ⎞ ⎞ T − TP ⎞ ⎛ ⎟ ⎟dt + ∫ (S ∆V )dt ⎟ − ⎜ kw A P ρc(TP − TP0 )∆V = ∫ ⎜ ⎜ k e A E ⎜⎜ δx PE ⎟ ⎜ δxWP ⎟ ⎟ ⎠ ⎝ ⎠⎠ t ⎝⎝ t An assumption about the variation of TP. By generalizing the approach by means of a weighting parameter θ between 0 and 1: t + ∆t IT = ∫ T dt = [θT P t P + (1 − θ )TP0 ∆t ⎞ ⎛ TP − TW ⎟ − ⎜ kw ⎟ ⎜ δxWP ⎠ ⎝ ⎞⎤ ⎟⎥ ⎟ ⎠⎦ ⎞⎤ ⎟⎥ + S ∆x ⎟ ⎠⎦ ] Therefore. the resulting scheme is “fully implicit”. ⎡⎛ T − T ⎛T −T 0 ⎞ ρc⎜ P P ⎟∆x = θ ⎢⎜ k e E P ⎜ ∆t ⎟ ⎜ δx PE ⎝ ⎠ ⎣⎝ 0 ⎡⎛ T 0 − TP0 ⎞ ⎛ TP0 − TW ⎟ − ⎜ kw + (1 − θ ) ⎢⎜ k e E ⎜ δx PE ⎟ ⎜ δxWP ⎠ ⎝ ⎣⎝ Re-arranging: ⎡ ∆x ⎛ k k ⎞⎤ k k 0 + θ ⎜ e + w ⎟⎥TP = e θTE + (1 − θ )TE0 + w θTW + (1 − θ )TW ⎢ ρc ⎟ ⎜ δx δxWP ⎠⎦ δx PE δxWP ∆t ⎝ PE ⎣ [ ] [ ] ⎡ ∆x k k ⎤ − (1 − θ ) e − (1 − θ ) w ⎥TP0 + S ∆x + ⎢ ρc δx PE δxWP ⎦ ⎣ ∆t Compared with standard form: 0 0 a PTP = aW θTW + (1 − θ )TW + a E θTE + (1 − θ )TE0 + a P − (1 − θ )aW − (1 − θ )a E TP0 + b where a P = θ (aW + a E ) + a ∆x 0 a P = ρc ∆t kw aW = δxWP k aE = e δx PE [ ] [ ] [ ] 0 P b = S ∆x When θ = 0. the resulting scheme is “explicit”. When 0 < θ ≤ 1. When θ = 1. applying the central differencing scheme. . the resulting scheme is “implicit”.

.When θ = 1/2. the resulting scheme is “the Crank-Nicolson”.

All coefficient must be positive in the discretised equation: 0 a P − (aW + a E − S P ) > 0 or k k ∆x ρc − ( w + e ) > 0 ∆t δxWP δx PE or ∆x 2k > ρc ∆t ∆x or (∆x) 2 ∆t < ρc 2k It becomes very expensive to improve spatial accuracy. This method is not recommended for general transient problems. [ ] . the explicit scheme described above is efficient for simple conduction calculations. provided that the time step size is chosen with care. Nevertheless.Explicit scheme 0 0 a PTP = aW θTW + (1 − θ )TW + a E θTE + (1 − θ )TE0 + a P − (1 − θ )aW − (1 − θ )a E TP0 + b [ ] [ ] [ 0 p ] The source term is linearised as b = S u + S pT and set θ = 0 The explicit discretisation: 0 0 a PTP = aW TW + a E TE0 + a P − (aW + a E ) TP0 + S u where 0 aP = aP ∆x 0 a P = ρc ∆t kw aW = δxWP k aE = e δx PE The scheme is based on backward differencing and its Taylor series trunction error accuracy is first-order with respect to times.

[ ] [ ] [ ] . it is normally used in conjunction with spatial central differencing. The Crank-Nicolson method is based on central differencing and hence it is second-order accurate in time. So. All coefficient must be positive in the discretised equation: a + aW 0 aP > E 2 or (∆x) 2 ∆t < ρc k This is only slightly less restrictive than the explicit method.Crank-Nicolson scheme 0 0 a PTP = aW θTW + (1 − θ )TW + a E θTE + (1 − θ )TE0 + a P − (1 − θ )aW − (1 − θ )a E TP0 + b Set θ = 1/2 0 ⎛ T + TW ⎞ ⎡ 0 a E aW ⎤ 0 ⎛ T + TE0 ⎞ ⎟ + aP − ⎟ + aW ⎜ W a PTP = a E ⎜ E − TP + b ⎜ 2 ⎟ ⎜ 2 ⎟ ⎢ 2 2 ⎥ ⎦ ⎝ ⎠ ⎠ ⎣ ⎝ where 1 1 0 a P = (a E + aW ) + a P − S P 2 2 ∆x 0 a P = ρc ∆t k aW = w δxWP k aE = e δx PE 1 b = S u + S pT p0 2 The method is implicit and simultaneous equations for all node points need to be solved at each time step.

The time marching procedure starts with a given initial field of temperature T0. The system is solved after selecting time step ∆t. [ ] [ ] [ ] .The fully implicit scheme 0 0 a PTP = aW θTW + (1 − θ )TW + a E θTE + (1 − θ )TE0 + a P − (1 − θ )aW − (1 − θ )a E TP0 + b Set θ = 1 0 a PTP = a E TE + aW TW + a PTP0 where 0 a P = a P + a E + aW − S P ∆x 0 a P = ρc ∆t kw aW = δxWP k aE = e δx PE A system of algebraic equations must be solved at each time level. All coefficients are positive. which makes the implicit scheme unconditionally stable for any size of time step. The implicit method is recommended for general purpose transient calculations because of its robustness and unconditional stability. The accuracy of the scheme is first-order in time.

Analysis: Microwave T( C) 100 Radiant t2 t3 T( C) 100 t3 t2 50 t1 -L 0 t0 +L -L 50 t0 t1 . (3) uniform surface heating for radiant oven. Find: Sketch temperature distributions at specific times during heating and cooling.Module 5: Worked out problems Problem 1: A microwave oven operates on the principle that application of a high frequency field causes electrically polarized molecules in food to oscillate. Schematic: Assumptions: (1) one-dimensional conduction in x. The net effect is a uniform generation of thermal energy within the food. (5) symmetry about mid plane. which enables it to be heated from refrigeration temperatures to 90º in as short a time as 30 s. (4) heat loss from surface of meat to surroundings is negligible during the heat process. (2) uniform internal heat generation for microwave. Known: Microwave and radiant heating conditions for a slab of beef.

the maximum temperature is at the mid plane.Comments: (1) With uniform generation and negligible surface heat loss. The situation is reversed shortly after cooling begins. the temperature distribution remains nearly uniform during microwave heating. (2) The interior of the meat is heated by conduction from the hotter surfaces during radiant heating. and the maximum temperature is at the mid plane. . During the subsequent surface cooling. and the lowest temperature is at the mid plane.

k=389W/m. and then justify.K Analysis: the time temperature history is given by Rt = 1 hAs ⎛ θ (t ) t = exp⎜ − ⎜ RC θi t t ⎝ ⎞ ⎟ ⎟ ⎠ Where C t = ρVc p As = πD 2 V= θ = T − T∞ πD 3 6 Recognize that when t = 69 s ⎛ t θ (t ) (55 − 27 )° C = = 0. that the sphere behaves as a space-wise isothermal object and calculate the heat transfer coefficient. pure copper (333K): =8933 kg/m3. . Assume. (3) constant properties. (2) negligible radiation exchange.718 = exp⎜ − ⎜ τ ° θi (66 − 27 ) C ⎝ t ⎞ ⎛ 69 s ⎞ ⎟ = exp⎜ − ⎟ ⎟ ⎜ τ ⎟ t ⎠ ⎠ ⎝ And noting that τ t = Rt C t find τ t = 208s Hence. cp=389 J/kg.time history of a sphere fabricated from pure copper. 69 s after the sphere is inserted into an air stream.K. Known: The temperature-time history of a pure copper sphere in air stream. The sphere which is 12. Properties: From table of properties. A thermocouple on the outer surface of the sphere indicates 55ºC.Problem 2: The heat transfer coefficient for air flowing over a sphere is to be determined by observing the temperature.7 mm in diameter is at 66º C before it is inserted into an air stream having a temperature of 27ºC. Find: The heat transfer coefficient between and the air stream Schematic: Assumptions: (1) temperature of sphere is spatially uniform.

K = 1.K Comments: Note that with Lc =D0 / 6 Bi = hLc 0.88 × 10 − 4 k 6 Hence Bi<0.0127 3 m 3 / 6)389 J / kg.3 W / m 2 .0127 2 m 2 × 208s h = 35.K π 0.h= ρVc p As τ t = 8933kg / m 3 (π 0.1 and the spatially isothermal assumption is reasonable.0127 = 35.3W / m 2 .K × m / 398 W / m. .

c= 1033 J/kg.K. Temperature of storage medium at this time. Schematic: Assumptions: (1) one-dimensional conduction. the lumped capacitance method may be used. Find: Time required achieving 75% of maximum possible energy storage. which is well insulated on its outer surface and enclosed alternating layers of the storage material and the flow passage.011 k 231 W / m. (3) negligible heat exchange with surroundings.05m which is at an initial temperatures of 25ºC. Analysis: recognizing the characteristic length is the half thickness. Q = ( ρVc)θ i [1 − exp(−t / τ i )] = − ∆E st − ∆E st . (2) constant properties. Each layer of the storage material is aluminium slab of width=0.K. pure (T 600K=327C): k=231W/m. (1) and (2).Problem 3: A thermal energy storage unit consists of a large rectangular channel. consider the conditions for which the storage unit is charged by passing a hot gas through the passages. find hL 100 W / m 2 . with the gas temperature and convection coefficient assumed to have constant values of T=600ºC and h=100W/m2.K throughout the channel how long will it take to achieve 75% of the maximum possible energy storage? What is the temperature of the aluminium at this time? Known: Configuration.025m = = 0.K × 0. the condition sought is for . Properties: From any table of properties: Aluminum.max = ( ρVc)θ i Dividing eq.K Bi = Hence. =2702kg/m3. initial temperature and charging conditions of a thermal energy storage unit.

025m × 1033 J / kg. the property temperatures dependence is significant and some error is incurred by assuming constant properties. However. selecting at 600K was reasonable for this estimate.25 T − T∞ = exp(−t / τ th ) Ti − T∞ T = T∞ + (Ti − T∞ ) exp(−t / τ th ) = 600°C − (575°C ) exp(−968 / 698) or t=968s.75 Solving for τ th and substituting numerical values. T=456°C Comments: for the prescribed temperatures.K = 698s Hence.max = 1 − exp(−t / τ th ) = 0.K 100W / m 2 . the time required is -exp (-t/698s) =-0. find τ th = ρVc hAs = ρLc h = 2702kg / m 3 × 0. .∆E st / ∆E st .

h T(L.05 m = = 0. is suddenly immersed in an oil bath and subjected to a convection cooling process. Schematic: K=50W/m. find . ρ=7835 kg/m3.1m x Ti=T(x.50 = 2 and for X/L=1.qmin) 2L=0.05 m ) 2 And Bi-1=1/0. Calculate the surface temperature of the wall 9 min after immersion.K T =30 C h=500W/m2.3 θi Ti − T∞ We know that Bi-1=1/0.k =7835kg/m3 c=465J/kg. The properties of the wall are k=50 W/m. Fo = αt L2 = (k / ρ c) t L2 = 50W / m.K Since Bi>0.0)=250 C Assumptions: The Biot number for the plane wall is Bi = hLc 500 W / m 2 .Problem 4: A one-dimensional plane wall with a thickness of 0.K.50 = 2.K. assuming the convection heat transfer coefficient for the wall in the bath is 500 W/m2.1 m initially at a uniform temperature of 250C is suddenly immersed in an oil bath at 30°C.K. and c=465 J/kg.K T .K × 0. initially at a uniform temperature. Known: plane wall. t ) − T∞ = ≈ 0.96 ( 0 .1. lumped capacitance analysis is not appropriate. Find: Surface temperature of the wall nine minutes after immersion.50 k 50 W / m. find θ 0 T (0. 9 min). T (L.K × (9 × 60 ) s = 2 .K / 7835 kg / m 3 × 465 J / kg .

Fig applies to the centerline temperature which is a function of the Biot number and the Fourier number.3(250 − 30)°C = 66°C (2) Since F0>=0. t ) = T∞ + 0.8(0. t ) =0. it follows that T ( L.24 (Ti − T∞ ) = 30°C + 0.8( θ 0 ) = 0.8 θ0 By combining equation. the approximate analytical solution for θ* is valid.50. The centerline temperature at t=9min follows from equation with T (0. t ) − T∞ and θ i = Ti − T∞ .240 From this value.0701.3 θ i ) =0.6533 rad and C1=1. Substituting numerical values into equations θ*=0.3(Ti − T∞ ) = 0. 9 min) =83°C which is identical to graphical result. θ (1.θ (1.2. . find T (L. t ) ≈ 0. From table with Bi=0.303 and θ*(1. t ) − T∞ = 0.24 θ i Recalling that θ = T ( L. and ζ1 =0.24 ( 250 − 30)°C = 83°C Comments: (1) note that figure provides a relationship between the temperature at any x/L and the centerline temperature as a function of only the Biot number. FO) =0.

Problem 5: A long cylinder of 30mm diameter. while the convection coefficient is 50W/m2. constant-temperature oil bath at 350K. c=1600 J/kg. find Fo = − 1 ⎡ θ* ⎤ ln ⎢ ⎥ 2 ⎢ C1 J o (ς 1 ) ⎥ ⎣ ⎦ T (ro . Known: A long cylinder.015m / 2) = = = 0.1019.7W/m.7W / m. find ς 1 =0.231 (1000 − 350) K ς 12 where θ * (1. θ * (r * .K Since BI c >0. Using the approximate series solutions for the infinite cylinder. is suddenly quenched in a large.441. BI c = hLc h(r0 / 2) 50W / m 2 .1. . initially at a uniform temperature of 1000K.8882 rad and C1=1.K. F0 ) = From table. Calculate the time required for the surface cylinder to reach 500K. Fo) = C1 exp( −ς 12 Fo) × J o (ς 12 r * ) Solving for Fo and letting =1. Bi=0. From table find Jo ( ς 12 ) =0. t o ) − T∞ Ti − T∞ = (500 − 350) K = 0.K. method is not suited. is suddenly quenched in large oil bath.221 k k 1. and ρ=400 kg/m3.K.K (0. Schematic: Assumptions: (1) one dimensional radial conduction. Substituting numerical values into equation. (2) constant properties Analysis: check whether lumped capacitance methods are applicable. The cylinder properties are k=1. Find: time required for the surface to reach 500K. initially at a uniform temperature.8121.

K = 145 s Comments: (1) Note that Fo>=0.27.29 and Bi-1 =2. Fo = αt ro 2 = Fo . find Fo as follows. so approximate series solution is appropriate.Fo = − 1 (0. t ) − T∞ ] = 537 K 0. find from fir r/ro=1 that θ (ro .8882) 2 ln[0.K / 1.72 (0. t ) T (ro . t ) − T∞ = ≈ 0.7 and eventually obtain t ≈ 144s. With Bi-1=2.015m) 2 × 400 kg / m 3 × 1600 J / kg . t ) = T∞ + 1 [T (ro .1019 × 0.8 or θo T (0. θi .ro 2 ρc k t = 1. t )0 − T∞ hence T (0.8121] = 1.8 θo (537 − 350) K = = 0.2.27. (2) Using the Heisler chart. find Fo ≈ 1.231 / 1.29 θ i (1000 − 350) K From fig.7W / m. with θo =0.72 From the definition of the Fourier number.

the problem is to find the time when the location r=9mm reaches 1000K. ρ=7800 kg/m3. k=50 W/m. Estimate the time required to harden the outer millimeter of a ball of diameter 20 mm if its initial temperature is 300 K. Then a 1mm outer layer is hardened. Schematic: Assumptions: (1) one-dimensional radial conduction. Assume that any location within the ball whose temperature exceeds 1000 K will be hardened.K) it is desirable to increase the surface temperature for a short time without significantly warming the interior of the ball.00 k 50 W / m. Analysis: since any location within the ball whose temperature exceeds 1000K will be hardened.Problem 6: In heat treating to harden steel ball bearings (c=500 J/kg. find Fo = − 1 ς 12 ⎡ ⎤ 1 ln ⎢θ * / C1 sin(ς 1 r * ) ⎥ * ς 1r ⎢ ⎥ ⎣ ⎦ . (2) constant properties.2. begin by finding the Biot number.K. (3) Fo≥0.K Using the appropriate solution form for a sphere solved for Fo . Find: time required to harden outer layer of 1mm. Known: A ball bearing is suddenly immersed in a molten salt bath. Bi = hro 5000 W / m 2 .020m / 2) = = 1.K. This type of heating can be accomplished by sudden immersion of the ball in a molten salt bath with T∞=1300 K and h= 5000 W/m2. heat treatment to harden occurs at locations with T>1000K. Using the approximate series solution.K (0.

5708 rad and C1 =1.03). .30 = = 0. Fo = − 1 (1.0 and r/r0=0.30 θi since θ θ θo = . At this time it can be easily shown the center temperature is T(0. From fig.r 2 ρc α kg J ⎛ 0.2) s.4s)=871K.5 ( ± 0.From table.2732.5708 × 0.02) θ i 0. t = Fo r 2o = Fo . (2) The Heisler charts can also be used.69(±0. read θ/θo =0. substitute numerical values.9rad )⎥ = 0. since θ = T − T∞ = 1000 − 1300 = −300 K θ i = Ti − T∞ = −1000 K It follows that θ = 0. with r* =r/ro= (9mm/10mm)=0.0.5708) 2 ⎡ (1000 − 1300) K ⎤ 1 sin(1. θi θi And then θ o 0.441× ⎜ / 50W / m.K = 3.43( ±0.5708 × 0.3) and t=3.43.69 o .K ⎝ 2 ⎠ m 2 Comments: (1) note the very short time required to harden the ball.2732 ln ⎢ 1.020 ⎞ = 0. read FO =0. θi θo θi then θ θ = 0. Bi-1=1.4 s ⎟ 7800 3 × 500 k kg.00. θi From fig at Note the use of tolerances assigned as acceptable numbers dependent upon reading the charts to ±5%. with Bi=1.45( ± 0.9.441 / 1. with Bi-1=1.69 θo =0. for the sphere find ς 1 =1.9.3.9 (300 − 1300) K ⎣ ⎦ From the definition of the Fourier number with α=k/ρc.

K K=15W/m.Problem 7: The convection coefficient for flow over a solid sphere may be determined by submerging the sphere. hence. for Bi=1.1m T(ro.K and 105 2 m /s. Surface temperatures after immersion in a fluid of prescribed temperatures and convection coefficient. which is at 75°C and measuring its surface temperature at some time during the transient heating process. at what time will a surface temperature of 60°C be recorded? Known: Initial temperatures and properties of solid sphere. 60 − 75 sin 90° = 0.5708 =0. ς 1 =1. respectively.333 k 15W / m.2732.t)=60°C T8 =75 C h=300W/m2. Find: The process time Schematic: D=0.370 . from table .2732 exp (−1.K (0.5708 2 Fo ) +Exp (-2. If the convection coefficient is 300W/m2. (2) constant properties. Analysis: the Biot number is Bi = h(r0 / 3) 300W / m 2 .30 = 1. The sphere has a diameter of 0. into the flow. which is initially at 25°C.0.K.467F0) 25 − 75 1. radial conduction.05m / 3) = = 0. r * =1.K Hence the lumped capacitance methods should be used.5708 rad and C1=1. From equation T − T∞ Ti − T∞ = C1 exp(−ς 12 Fo ) sin(ς 1 r * ) ς 1r * At the surface.1m. and its thermal conductivity and thermal diffusivity are 15 W/m.K a=10-5m2/s Ti=25 C Assumptions: (1) one-dimensional.

403 (0. .Fo = αt r02 = 0.05m) 2 10 −5 m 2 / s t=100s Comments: Use of this technique to determine h from measurement of T (ro) at a prescribed t requires an iterative solution of the governing equations.

Consider a sphere and a cylinder of equal volume and made of copper. and when is it not? . During the second minute. What is lumped capacity analysis? When is it applicable? What is the physical significance of the Biot number? 2. For which case is the lumped capacity assumption more applicable? 4. when is it proper to treat an actual cylinder as an infinitely long one. The Biot number is used when considering a solid body subject to convection in a surrounding fluid. Both are heated to the same temperature and then kept in air for cooling. while the second one is cooled by natural convection in air. the temperature drops by 10ºC. In transient heat transfer analysis. The product of the block’s resistance to convection and its lumped thermal capacitance is a) Bi b) Nu c) Fo d) τ e) None of the above 8. less than or same as that during the first minute? 5. Consider heat transfer between two identical hot solid bodies and the surrounding air. A block of metal is cooled in a water bath. Which one is likely to cool faster? 7.Module 5: Short questions 1. Consider a hot boiled potato kept on a plate and cooled by natural convection in air. In what medium will the lumped capacity assumption more likely to be valid: in air or in water? 6. will the temperature drop be more than. Its unsteady temperature is considered uniform and is thus modelled using a lumped capacitance method. During the first minute. The first solid is cooled by a fan. It is a comparison of a) Convection to conduction in the surrounding fluid b) Conduction in the surrounding fluid to conduction in the solid c) Convection at the solid surface to conduction within the solid d) The thermal diffusivity in the solid to the kinematic viscosity in the fluid e) None of the above 3.

The appropriate length scale for this is therefore 0. What is a semi-infinite medium? Give examples of solid bodies that can be treated as semi-infinite mediums for the purpose of transient heat transfer studies? Under what conditions can a plane wall be treated as a semi-infinite medium? 12.5 (c) (ht ) (d) ρct (e) none of the above .5 (a) (αt ) (b) αt 0.9. there is no geometric length scale with which to construct a Biot number. Why are the transient temperature charts prepared using non-dimensionalised quantities such as the Biot and the Fourier numbers and not the actual variables such as thermal conductivity and time? 10. What is the physical significance of the Fourier number? Will the Fourier number of a specific transient heat transfer problem double if the time is doubled? 11. When modelling the unsteady conduction in a semi-infinite slab with convection at the surface.

Introduction to Heat Exchangers Contents: What are exchangers for? Main heat exchanger types Heat exchanger analysis LMTD method Effectiveness method Pradeep datta HT/M6/L1/V1/2004 1 .

What are heat exchangers for? Heat exchangers are practical devices used to transfer energy from one fluid to another To get fluid streams to the right temperature for the next process reactions often require feeds at high temp. To condense vapours To evaporate liquids To recover heat to use elsewhere To reject low-grade heat To drive a power cycle Pradeep datta HT/M6/L1/V1/2004 2 .

Application: Power cycle Steam turbine Boiler Feedwater heater Condenser Pradeep datta HT/M6/L1/V1/2004 3 .

Main categories of exchanger Heat exchangers Recuperators Wall separating streams Wall separating streams Regenerators Direct contact Most heat exchangers have two streams. but some have more than two Pradeep datta HT/M6/L1/V1/2004 4 . hot and cold.

Recuperators/Regenerators Recuperative Has separate flow paths for each fluid which flow simultaneously through the exchanger transferring heat between the streams Regenerative Has a single flow path which the hot and cold fluids alternately pass through. Pradeep datta HT/M6/L1/V1/2004 5 .

Compactness Can be measured by the heat-transfer area per unit volume or by channel size Conventional exchangers (shell and tube) have channel size of 10 to 30 mm giving about 100m2/m3 Plate-type exchangers have typically 5mm channel size with more than 200m2/m3 More compact types available Pradeep datta HT/M6/L1/V1/2004 6 .

can have very many tubes thus becoming a shell-and-tube Pradeep datta HT/M6/L1/V1/2004 7 .inner tube may have longitudinal fins on the outside However. most have a number of tubes in the outer tube .Double Pipe Simplest type has one tube inside another .

Shell and Tube Typical shell and tube exchanger as used in the process industry Pradeep datta HT/M6/L1/V1/2004 8 .

Shell-side flow Pradeep datta HT/M6/L1/V1/2004 9 .

Plate-fin exchanger Made up of flat plates (parting sheets) and corrugated sheets which form fins Brazed by heating in vacuum furnace Pradeep datta HT/M6/L1/V1/2004 10 .

Configurations Pradeep datta HT/M6/L1/V1/2004 11 .

h 1 1 1 = + + Rw + + (hηo A)h (ηo A)h UA (hηo A)c (ηo A)c • A is wall total surface area on hot or cold side • R”f is fouling factor (m2K/W) • ηo is overall surface efficiency (if finned) Pradeep datta HT/M6/L1/V1/2004 12 .c R′f′.Heat Transfer considerations: • Overall heat transfer coefficient Internal and external thermal resistances in series 1 1 1 = = UA (UA)c (UA)h R′f′.

Such deposits are termed fouling and may significantly affect heat exchanger performance. MgSiO3. Many metals. and Li2CO3. CaSiO3. form adherent oxide coatings which serve to passivate the surface and prevent further corrosion. Ca(OH)2. LiSO4. Ca3(PO4)2. Examples of such salts are CaCO3. Mg(OH)2. CaSO4.Heat Transfer considerations: • Fouling factor Material deposits on the surfaces of the heat exchanger tube may add further resistance to heat transfer in addition to those listed above. copper and aluminum being specific examples. Na2SO4. Pradeep datta HT/M6/L1/V1/2004 13 . • Scaling is the most common form of fouling and is associated with inverse solubility salts. • Corrosion fouling is classified as a chemical reaction which involves the heat exchanger tubes.

• Freezing fouling is said to occur when a portion of the hot stream is cooled to near the freezing point for one of its components. When food products are involved this may be termed scorching but a wide range of organic materials are subject to similar problems.0001 2 14 .• Chemical reaction fouling involves chemical reactions in the process stream which results in deposition of material on the heat exchanger tubes. m K/watt 0.0001 0.0002 0. This is most notable in refineries where paraffin frequently solidifies from petroleum products at various stages in the refining process. • Biological fouling is common where untreated water is used as a coolant stream.0002 0.001 0. Problems range from algae or other microbes to barnacles.0009 0.0002-0. Fluid Seawater and treated boiler feedwater (below 50oC) Seawater and treated boiler feedwater (above 50oC) River water (below 50oC) Fuel Oil Regrigerating liquids Steam (non-oil bearing) Pradeep datta HT/M6/L1/V1/2004 R”. obstructing both flow and heat transfer.

Basic flow arrangement in tube in tube flow t1 t2 t2 t1 T1 T1 Temperature Parallel Flow T2 T 1 T 1 Temperature C ounter Flow T 2 T2 t1 t2 T 2 t2 t1 Position Position Pradeep datta HT/M6/L1/V1/2004 15 .

Heat Exchanger Analysis Log mean temperature difference (LMTD) method & Want a relation Q = UA∆Tm where ∆Tm is some mean ∆T between hot and cold fluid Pradeep datta HT/M6/L1/V1/2004 16 .

Pradeep datta HT/M6/L1/V1/2004 17 .

Pradeep datta HT/M6/L1/V1/2004 18 .

Pradeep datta HT/M6/L1/V1/2004 19 .

concentric tube heat exchanger is used to cool the lubricating oil for a large industrial gas turbine engine. What should be the length of the tube if if the outlet temperature of the oil is to be 60 °C ? Soultion: To be worked out in class.2 kg/s. respectively. while the flow rate of oil through the outer annulus (Do = 45 mm) is 0.1 kg/s.Example:A counterflow. Pradeep datta HT/M6/L1/V1/2004 20 . The flow rate of cooling water through the inner tube (Di = 25 mm) is 0. The oil and water enter at temperatures of 100 and 30 °C.

it is common to design for multiple tubes within a single heat exchanger.Multipass HX Flow Arrangements •In order to increase the surface area for convection relative to the fluid volume. indicating that the shell side fluid passes through the unit once. 1-2 pass heat exchanger. •With multiple tubes it is possible to arrange to flow so that one region will be in parallel and another portion in counter flow. Pradeep datta HT/M6/L1/V1/2004 21 . By convention the number of shell side passes is always listed first. the tube side twice.

.t = Shell / tube side. FT. 6. can be found theoretically and presented in analytical form. The equation given below has been shown to be accurate for any arrangement having 2. FT. 2 = inlet / outlet HT/M6/L1/V1/2004 .... 4.The LMTD formulas developed earlier are no longer adequate for multipass heat exchangers. LMTDcf. 1. and to apply a correction factor. . ⎡ 1− P ⎤ R 2 + 1 ln ⎢ ⎥ ⎣1 − R ⋅ P ⎦ FT = ⎤ ⎡ 2 − P⎛ R + 1 − R 2 + 1 ⎞ ⎥ ⎜ ⎟ ⎢ ⎝ ⎠ (R − 1)ln ⎢ ⎥ ⎛ R +1+ R2 +1⎞ ⎥ ⎢ 2 − P⎜ ⎟ ⎢ ⎝ ⎠⎥ ⎦ ⎣ Effectiveness : P = 1 − X 1/ N shell R− X 1 / N shell . such that ∆θ eff = FT ⋅ LMTDCF The correction factors. for R ≠ 1 P= N shell Po ..2n tube passes per shell pass to within 2%. for R = 1 − Po ⋅ (N shell − 1) X = Po ⋅ R − 1 Po − 1 22 Pradeep datta T1 − T2 Capacity ratio R = t 2 − t1 t 2 − t1 Po = T1 − t1 T. Normal practice is to calculate the LMTD for counter flow.

1 0.1.0 R=0.0 Pradeep datta HT/M6/L1/V1/2004 23 .5 0.0 FT R=10.0 P 1.

max One fluid ∆T → ∆Tmax = Th.in − Tc. perform for given inlet conditions ? & Qactual Define effectiveness : ε = & Qmax & where Q is for an infinitely long H.Ex.Ex.Effectiveness-NTU method How will existing H.in & & & and since Q = (mc A )∆T A = (mc B )∆TB = C A ∆T A = C B ∆TB then only the fluid with lesser of C A . C B heat capacity rate can have ∆Tmax Pradeep datta HT/M6/L1/V1/2004 24 .

in − Tc. Qmax = C min ∆Tmax and ε = & or..) UA NTU = C min 25 .in ) & Q Want expression for ε which does not contain outlet T' s & Substitute back into Q = UA( LMTD ) ..exp ⎢ ⎜ C ⎟ ⎢ C min ⎝ max ⎠ ⎥ ⎣ ⎦ ε= ⎡ ... min ∴ε = ε ⎜ C max ⎝ Pradeep datta ⎞⎤ ⎟⎥ ⎟ ⎠⎥ ⎦ ⎞ ⎟ ⎟ ⎠ HT/M6/L1/V1/2004 and No. ⎡ ..UA ⎛ C min C min ⎜1 − 1− exp ⎢ ⎜ C max ⎢ C min ⎝ C max ⎣ ⎛ C ⎜ NTU .& i. of transfer units (size of HEx.UA ⎛ C min ⎞⎤ ⎜1 − ⎟⎥ 1 .e. Q = εC min (Th..in − Tc.in ) C min (Th...

→ ε from chart Pradeep datta HT/M6/L1/V1/2004 26 & Q = ε C min (Th. Cmin/Cmax Get UA/Cmin..CHARTS FOR EACH CONFIGURATION .in ) ..in − Tc. Procedure: Determine Cmax.

Determine area of the HEX using the effectiveness-NTU method.09 kJ/kg-°C) enter at 200°C and leave at 93°C. Solution: (to be worked out in class) Pradeep datta HT/M6/L1/V1/2004 27 . The gases (cpg=1.5 kg/s of water from 35°C to 85°C.Example: Hot exhaust gases are used in a finned-tube cross-flow HEX to heat 2. The overall heat-transfer coefficient is 180 W/m2°C.

Pradeep datta HT/M6/L1/V1/2004 28 .

we have been considering convection heat transfer in homogeneous single-phase (HSP) systems • Boiling and condensation. provide much higher heat transfer rates than those possible with the HSP systems . however.Condensation and Boiling • Until now.

Condensation • Condensation occurs when the temperature of a vapour is reduced below its saturation temperature • Condensation heat transfer Film condensation Dropwise condensation • Heat transfer rates in dropwise condensation may be as much as 10 times higher than in film condensation .

Laminar film condensation on a vertical wall (VW) µl ∂u ⎞ ⎟ A ∂y ⎟ y ⎠ ∂u ⎞ µl ⎟ A ⎟ ∂y ⎠ y + ∆y (ρ l − ρ v )gA∆y .

Laminar film condensation on a vertical wall (cont.943⎢ ⎢ L(Tsat − Tw )ν l ⎥ ⎣ ⎦ q = hL A(Tsat − Tw ) & m= h A(Tsat − Tw ) q = L h fg h fg 1/ 4 . where L is the plate length. Total heat transfer rate : Condensation rate : ⎡ h fg g ( ρ l − ρ v )k l3 ⎤ ⎥ hL = 0..) ⎡ 4 xkl (Tsat − Tw )ν l ⎤ δ ( x) = ⎢ ⎥ ⎢ h fg g ( ρ l − ρ v ) ⎥ ⎦ ⎣ ⎡ h fg g ( ρ l − h( x ) = ⎢ ⎢ 4 x(Tsat − Tw )ν l ⎣ 1/ 4 ρ v )k l3 ⎤ ⎥ ⎥ ⎦ 1/ 4 Average coeff.

50 cm high. (ii) the total condensation rate. (ii) total condensation rate. Laminar Solution: . Effect of liquid subcooling negligible 3. determine: (i) the average heat transfer coefficient. (iii) and film thickness 1. and (iii) the film thickness at the bottom of the tube. If the saturation temperature of the steam is 302 K. Given: Film condensation of saturated steam Required: (i) Average heat transfer coefficient. and cooling water maintains the wall temperature at 299 K.Example Laminar film condensation of steam Saturated steam condenses on the outside of a 5 cm-diameter vertical tube. Effect of tube curvature negligible 2.

Example (cont.) .

Example (cont.) .

Example (cont.) .

g.). heat transfer coefficient also depends on those factors. thermophysical properties of fluid (e.. liquid density. surface tension. Hence. . etc. nature of surface. vapour density. which grow and subsequently detach from the surface • Bubble growth and dynamics depend on several factors such as excess temp.Boiling • Boiling occurs when the surface temperature Tw exceeds the saturation temperature Tsat corresponding to the liquid pressure Heat transfer rate : q′′ = h(Tw − Tsat ) = h∆Te s where ∆Te = Tw − Tsat (excess temperature) • Boiling process is characterised by formation of vapour bubbles.

Pool Boiling Curve q′′ s (∆Te ) = .

Modes of Pool Boiling ∆Te ≈ 5 o C • Free convection boiling • Nucleate boiling • Transition boiling • Film boiling 5o C ≤ ∆Te ≤ 30 o C 30 C ≤ ∆Te ≤ 120 C ∆Te ≥ 120 C o o o .

Radiation requires no medium for its propagation. . atoms and electrons. The rate at which radiation energy is emitted is usually quantified by the modified Stefan-Bolzmann law: where the emissivity.67x10-8 W/(m2K4)). can take place also in vacuum. Tb is the absolute temperature of the surface (in Kelvin).Radiation Heat Transfer Thermal energy emitted by matter as a result of vibrational and rotational movements of molecules. The energy is transported by electromagnetic waves (or photons). is a property of the surface characterizing how effectively the surface radiates compared to a "blackbody" (0<ε<1 ). E=q/A (W/m2) is the surface emissive power. therefore. σ is the Stefan-Boltzmann constant (σ=5. ε . All matters emit radiation as long as they have a finite (greater than absolute zero) temperature.

Gα=αG Absorbed irradiation .

.Electromagnetic radiation spectrum Thermal radiation spectrum range: 0.1 to 100 mm It includes some ultraviolet (UV) radiation and all visible (0.76 mm) and infrared radiation (IR).4-0.

.

The Planck distribution is shown in the following figure as a function of wavelength for different body temperatures. .742x108 (W.K) are two constants.µm4/m2) and C2=1.439x104 (µm.The Planck Distribution The Planck law describes theoretical spectral distribution for the emissive power of a black body. It can be written as where C1=3.

Spectral blackbody emissive power .

the emissive power increases with increasing temperature • As the temperature increases. • Sun can be approximated as a blackbody at 5800 K .more emissive energy appear at shorter wavelengths • For low temperature (< 800 K). can glow. all radiant energy falls in the infrared region and is not visible to the human eyes.Planck Distribution • At given wavelength. such as molten iron. That is why only very high temperature objects.

Ω. which like the standard angle has no dimensions.Angles and Arc Length We are well accustomed to thinking of an angle as a two dimensional object. when multiplied by the radius squared will have dimensions of length squared. It may be used to find an arc length: α L L = r·α Solid Angle We generalize the idea of an angle and an arc length to three dimensions and define a solid angle. or area. and will have the magnitude of the encompassed area. The solid angle. A = r2·dΩ r .

dq I= cos θ ⋅ dA1 ⋅ dΩ . Aprojected = cos θ·Anormal θ dA1 dA1·cos θ Intensity The ideal intensity. per unit solid angle.Projected Area The area. as seen from the prospective of a viewer. per unit projected area. as cos θ·dA1. This smaller area is termed the projected area. Ib. may now be defined as the energy emitted from an ideal body. dA1. situated at an angle θ from the normal to the surface. per unit time. will appear somewhat smaller.

azimuthal and zenith directions. and θ. The three basic coordinates shown are R. φ. dA2. Note that the area proscribed on the hemisphere. the spherical coordinate system provides a convenient tool for analysis. R·sin θ dA2 dA1 θ R ∆φ In general dA1 will correspond to the emitting φ surface or the source.Spherical Geometry Since any surface will emit radiation outward in all directions above the surface. dA = R2·dΩ we find that: dΩ = R2·sin θ·dθ·dφ . may be written as: dA = [( R ⋅ sin θ ) ⋅ dϕ ] ⋅ [ R ⋅ dθ ] 2 dA2 = R 2 ⋅ sin θ ⋅ dϕ ⋅ dθ ] Recalling the definition of the solid angle. representing the radial. The surface dA2 will correspond to the receiving surface or the target.

which are somewhat less than that obtained theoretically by Boltzman. G Transmitted Radiation . reflection or Reflected Radiation transmission. i. those that emit energy according to the Stefan-Boltzman law: Eb = σ·Tabs4 Real surfaces have emissive powers. the fraction of incident radiation absorbed. •Absorptivity.Real Surfaces Thus far we have spoken of ideal surfaces. • Transmissivity. E. E ε≡ Eb Emissive power from any real surface is given by: E = ε·σ·Tabs4 Receiving Properties Targets receive radiation in one of three ways. they absorption. Incident Radiation. τ. the fraction of incident radiation transmitted. the fraction of incident Radiation radiation reflected. ρ. α. ε. To account for this reduction. we introduce the emissivity.e. Absorbed •Reflectivity.

0. TA Surface B. both emitting radiation toward one another. from Conservation of Energy. and Emissivity. so that: α+ρ=1 Relationship Between Absorptivity.ε Consider two flat.e. TB Now suppose that the two surfaces are at exactly the same temperature. τ = 0.We see. If follows then that: αA = εA . εB = 1.α.αA·σ·TB4 Surface A. surface A and surface B. we will deal with only opaque surfaces. that: α+ρ+τ =1 In this course. i. Surface B is assumed to be an ideal emitter. Surface A will emit radiation according to the Stefan-Boltzman law as: EA = εA·σ·TA4 and will receive radiation as: GA = αA·σ·TB4 The net heat flow from surface A will be: q΄΄ = εA·σ·TA4 . infinite planes. The heat flow must be zero according to the 2nd law.

and on the temperature of surface B.e. aluminum. ε. which absorbs all visible light. The terms ideal surface and black surface are used interchangeably. brass. we speak of surfaces with α = 1 as also being “black” or “thermally black”. of surface A will depend on the material of which surface A is composed. λ. It follows that for such a surface. etc. α. any material. α and ε may depend on temperature. appears as black. brass. . Extending this concept to the much broader thermal band. The wave length of radiation will. etc. and on the temperature of surface A. i. Black Surfaces Within the visual band of radiation. in turn.e.Thermodynamic properties of the material. steel. The emissivity. The absorptivity. i. of surface A will depend on the material of which surface A is composed. ε = 1 and the surface will behave as an ideal emitter. steel. depend on the temperature of the source of radiation. aluminum. this will be the case as radiative properties will depend on wavelength. In general.

or incident radiation.reflected . Grey Surface: A surface for which the spectral absorptivity and the emissivity are independent of wavelength over the spectral regions of surface irradiation and emission. .Diffuse Surface: Refers to directional independence of the intensity associated with emitted.

Relationship Between Emissive Power and Intensity By definition of the two terms. Eb = hemisphere ∫I b ⋅ cos θ ⋅ dΩ Replacing the solid angle by its equivalent in spherical angles: Eb = ∫ ∫ 0 2⋅π π 2 0 I b ⋅ cos θ ⋅ sin θ ⋅ dθ ⋅ dϕ Integrate once. emissive power for an ideal surface. Eb. (Note that the derivative of sin θ is cos θ·dθ.) Eb = 2 ⋅ π ⋅ I b ⋅ ∫ π 2 0 2 cos θ ⋅ sin θ ⋅ dθ π 2 sin θ Eb = 2 ⋅ π ⋅ I b ⋅ 2 = π ⋅ Ib 0 Eb = π·Ib . holding Ib constant: Integrate a second time. and intensity for an ideal surface. Ib.

First find the projected area of surface dA2. Ω. dA2. we integrate over both surfaces: q1→ 2 = ∫ ∫ I ⋅ cos θ1 ⋅ dA1 ⋅ cos θ 2 dA2 R2 A2 A1 .) Then find the solid angle. (θ2 is the angle between the normal to surface 2 and the position vector. which encompasses this area. R. dA1. dA2·cos θ2.I= dq cos θ ⋅ dA1 ⋅ dΩ dq = I ⋅ cos θ ⋅ dA1 ⋅ dΩ dA2 dA2·cos θ2 Next we will project the receiving surface onto the hemisphere surrounding the source. R To obtain the entire heat transferred from a finite area. to a finite area.

I ⋅ cos θ1 ⋅ dA1 ⋅ cos θ 2 dA2 q1→ 2 A2 A1 R2 F1→ 2 = = π ⋅ I ⋅ A1 qemitted ∫ ∫ dr F1→ 2 1 = ⋅ A1 ∫ ∫ cos θ1 ⋅ cosθ 2 ⋅ dA1 ⋅ dA2 A2 A1 π ⋅ R2 D Example Consider a diffuse circular disk of diameter D and area Aj and a plane diffuse surface of area Ai << Aj. and Ai is located at a distance L from the center of Aj. which directly strikes surface 2. F1-2. Obtain an expression for the view factor Fij.Total energy emitted from surface 1: qemitted = E1·A1 = π·I1·A1 View Factors-Integral Method Define the view factor. The surfaces are parallel. as the fraction of energy emitted from surface 1. dAj Aj θj R θi dAi dAi R L .

F1→ 2 = ∫ L2 ⋅ 2 ⋅ ρ ⋅ dρ A2 ρ4 D F1→2 = − 2 ⋅ L ⋅ 2 ρ −2 A2 2 ⎡ 1 ⎤ 2 = −L ⋅ ⎢ 2 2⎥ ⎢ L + ρ ⎥0 ⎦ ⎣ 2 D F1→ 2 4 1⎤ 2 D2 ⎡ = −L ⋅ ⎢ − 2⎥ = 2 2 L ⎦0 4 ⋅ L2 + D 2 ⎣4⋅ L + D 2 . Then 2·ρ·dρ = 2·r·dr.F1→ 2 1 = ⋅ A1 ∫ ∫ cos θ1 ⋅ cosθ 2 ⋅ dA1 ⋅ dA2 A2 A1 Since dA1 is a differential area F1→ 2 F1→ 2 = ∫ ( R ) ⋅ 2π ⋅ r ⋅ dr L 2 π ⋅ R2 cos θ1 ⋅ cos θ 2 ⋅ dA2 = A2 π ⋅ R2 ∫ A2 π ⋅ R2 F1→ 2 = ∫ L2 ⋅ 2 ⋅ r ⋅ dr R4 A2 Let ρ2 ≡ L2 + r2 = R2.

1 = A1 ⋅ F1. 1 2 A2 ⋅ F2. part will return to surface 2. we can then see that: ∑ j =1 N Fi . We imagine that the surrounding surfaces act as an enclosure about the heat source which receive all emitted energy. Find F2.Enclosures In order that we might apply conservation of energy to the radiation process. Part of the radiant energy leaving the inside surface of object 2 will strike surface 1. For an N surfaced enclosure. 2 A2 A1 ⎡ D1 ⎤ = =⎢ ⎥ A2 ⎣ D2 ⎦ 2 . All radiation leaving the outside of surface 1 will strike surface 2.1 . This relationship is known as “Reciprocity”.1 = A1 ⋅ F1. Reciprocity Ai ⋅ Fi → j = A j ⋅ F j →i Example: Consider two concentric spheres shown to the right. j = 1 This relationship is known as the Conservation Rule”. we must account for all energy leaving a surface. Apply reciprocity. 2 ⇒ F2.

ε·Eb ρ·G G J ≡ ε·Eb + ρ·G . Radiosity Radiosity. is defined as the total energy leaving a surface per unit area and per unit time. the fraction of energy leaving surface i and striking the combined surface j+k will equal the fraction of energy emitted from i and striking j plus the fraction leaving surface i and striking k. from conservation of energy. J.Associative Rule Consider the set of surfaces shown to the right: Clearly. i j k Fi ⇒ ( j + k ) = Fi ⇒ j + Fi ⇒ k This relationship is known as the “Associative Rule”.

Net Exchange Between Surfaces Consider the two surfaces shown. Radiation will travel from surface i to surface j and will also travel from j to i.Jj·Aj· Fj→j From reciprocity we note that F1→2·A1 = F2→1·A2 so that qj→i (net) = Ji·Ai· Fi→j . qi→j = Ji·Ai· Fi→j likewise.Jj· Ai· Fi→j = Ai· Fi→j·(Ji – Jj) Jj Ji . qj→i = Jj·Aj· Fj→j The net heat transfer is then: qj→i (net) = Ji·Ai· Fi→j .

Net Energy Leaving a Surface The net energy leaving a surface will be the difference between the energy leaving a surface and the energy received by a surface: q1→ = [ε·Eb – α·G]·A1 Combine this relationship with the definition of Radiosity to eliminate G.ε·Eb]/(1 – α)}·A1 Put the equation over a common denominator: q1→ ⎡ (1 − α ) ⋅ ε ⋅ Eb − α ⋅ J + α ⋅ ε ⋅ Eb ⎤ ⎡ ε ⋅ Eb − α ⋅ J ⎤ =⎢ ⎥ ⋅ A1 = ⎢ ⎥ ⋅ A1 1−α 1−α ⎣ ⎦ ⎣ ⎦ assume that α = ε q1→ ⎡ ε ⋅ Eb − ε ⋅ J ⎤ ⎡ ε ⋅ A1 ⎤ =⎢ ⎥ ⋅ A1 = ⎢ 1 − ε ⎥ ⋅ (Eb − J ) 1− ε ⎦ ⎣ ⎣ ⎦ . q1→ = {ε·Eb – α·[J . J ≡ ε·Eb + ρ·G G = [J .ε·Eb]/ρ}·A1 Assume opaque surfaces so that α + ρ = 1 ρ = 1 – α. and substitute for ρ.ε·Eb]/ρ q1→ = {ε·Eb – α·[J .

Electrical Analogy for Radiation We may develop an electrical analogy for radiation. resistance and current flows. The two analogies should not be mixed: they have different dimensions on the potential differences.J Net Exchange Between Surfaces qi? j J1 – J2 . similar to that produced for conduction. Equivalent Current Equivalent Resistance Potential Difference Ohms Law I R ∆V Net Energy Leaving Surface q1? ⎡1 − ε ⎤ ⎢ε ⋅ A⎥ ⎣ ⎦ 1 A1 ⋅ F1→2 Eb .

its temperature will rise. moves across the grate as it burns and radiates to the walls and top of the furnace.Example: Consider a grate fed boiler. SH Coal Chute section Surface 3 Superheat Tubes Surface 2 Water Walls Surface 1 Burning Coal . It is common practice to subdivide the super-heater tubes into sections. it is circulated through the superheat tubes at the top of the boiler. each having nearly uniform temperature. Saturated water is introduced at the bottom of the walls and leaves at the top at a quality of about 70%. In our case we will use only one superheat section using an average temperature for the entire region. Coal is fed at the bottom. The walls are cooled by flowing water through tubes placed inside of the walls. After the vapour is separated from the water. Since the steam is undergoing a sensible heat addition.

.

.

. Surface 1.Energy will leave the coal bed. That part of the circuit is represented by a potential difference between Radiosity: 1 A2 ⋅ F2→3 J2 J3 R12 J1 R1 Eb1 J1 R1 Eb1 ⎡ 1− ε ⎤ ⎢ ⎥ ⎣ ε ⋅ A1 ⎦ σ·T14 J2 R12 1 A1 ⋅ F1→ 2 J3 R13 1 J1 A1 ⋅ F1→3 R1 Eb1 1 A1 ⋅ F1→ 2 R13 1 A1 ⋅ F1→3 It should be noted that surfaces 2 and 3 will also radiate to one another. We draw the equivalent electrical network as seen to the right: The heat leaving from the surface of the coal may proceed to either the water walls or to the super-heater section. as described by the equation for the net energy leaving a surface.

. In steady state heat transfer. • Black surfaces: A black. a surface cannot receive net energy if it is insulated. Insulated surfaces are often termed as re-radiating surfaces. will have no surface resistance: ⎡1 − ε ⎤ ⎡1 − 1 ⎤ ⎢ ε ⋅ A ⎥ = ⎢1 ⋅ A ⎥ = 0 ⎦ ⎣ ⎦ ⎣ In this case the nodal Radiosity and emissive power will be equal. or ideal surface. 1 A2 ⋅ F2→3 J2 R2 J3 R12 1 A1 ⋅ F1→ 2 ⎡ 1− ε 2 ⎤ ⎢ ⎥ ⎣ ε 2 ⋅ A2 ⎦ Eb2 R13 J1 ⎡ ε ⋅ A1 ⎤ ⎢ 1− ε ⎥ ⎣ ⎦ 1 A1 ⋅ F1→3 Eb3 ⎡ ε ⋅ A1 ⎤ ⎢ 1− ε ⎥ ⎣ ⎦ R1 Eb1 • Insulated surfaces. Because the energy cannot be stored by a surface in steady state.It remains to evaluate the net heat flow leaving (entering) nodes 2 and 3. all energy must be re-radiated back into the enclosure.

2. The system will consist of two objects. irrespective of the actual surface properties: ⎡1 − ε ⎤ ⎡ 1 − ε ⎤ ⎢ε ⋅ A ⎥ = ⎢ε ⋅ ∞ ⎥ = 0 ⎣ ⎦ ⎣ ⎦ Consider the case of an object. (1-ε1)/(ε1⋅A1) Eb1 σ⋅T14 R1 J1 1/(A1⋅F1→2) J2 (1-ε2)/(ε2⋅A2) R2 Eb2 σ⋅T24 J1 1/(A1⋅F1→2) J2 R12 .• Large surfaces: Surfaces having a large surface area will behave as black surfaces. placed inside a large enclosure. Now we ground both Radiosity nodes through a surface resistance. so we proceed to construct a circuit with two radiosity nodes. 1.

let us assume that we know (a) the total heat being produced by the coal bed. . R2 = 0. F1→2 = 1 (1-ε1)/(ε1⋅A1) Eb1 σ⋅T14 R1 J1 1/(A1⋅F1→2) R12 J2 Eb2 σ⋅T24 Sum the series resistances: RSeries = (1-ε1)/(ε1⋅A1) + 1/A1 = 1/(ε1⋅A1) Ohm’s law: i = ∆V/R or by analogy: q = ∆Eb/RSeries = ε1⋅A1⋅σ⋅(T14 – T24) Returning for a moment to the coal grate furnace.Since A2 is large. The view factor. (b) the temperatures of the water walls and (c) the temperature of the super heater sections.

Apply Kirchoff’s law about node 1. matrix methods may be used: 1 R12 1 1 1 − − − R2 R12 R13 1 R23 . Since they are each linear in J. for the coal bed: q1 + q 2→1 + q3→1 = q1 + Similarly. for node 2: J 2 − J1 J 3 − J1 + =0 R12 R13 q 2 + q1→ 2 + q3→ 2 And for node 3: Eb 2 − J 2 J 1 − J 2 J 3 − J 2 = + + =0 R2 R12 R23 q3 + q1→3 + q 2→3 ⎡ 1 1 − ⎢− ⎢ R12 R13 1 ⎢ ⎢ R12 ⎢ 1 ⎢ ⎢ R13 ⎣ Eb 3 − J 3 J 1 − J 3 J 2 − J 3 = + + =0 R3 R13 R23 ⎤ ⎡ ⎤ 1 ⎥ ⎢ ⎥ − q1 ⎥ R13 ⎥ ⎡ J1 ⎤ ⎢ 1 ⎥ ⋅ ⎢ J ⎥ = ⎢ − Eb 2 ⎥ ⎥ ⎢ 2⎥ ⎢ R ⎥ R23 2 ⎥ ⎥ ⎢ ⎥ ⎢ 1 1 1 ⎥ ⎣ J 3 ⎦ ⎢ Eb 3 ⎥ − − − − ⎢ R3 ⎥ R3 R13 R23 ⎥ ⎦ ⎦ ⎣ The three equations must be solved simultaneously.

given the heat flow: Eb1 − J1 σ ⋅ T14 − J1 ⎡q ⋅ R + J ⎤ q1 = = ⇒ T1 = ⎢ 1 1 1 ⎥ σ R1 R1 ⎦ ⎣ 0.Surface 1: Find the coal bed temperature. given the water wall temperature: Eb 2 − J 2 σ ⋅ T24 − J 2 q2 = = R2 R2 Surface 3: (Similar to surface 2) Find the water wall heat input.25 Surface 2: Find the water wall heat input. given the water wall temperature: Eb3 − J 3 σ ⋅ T34 − J 3 q3 = = R3 R3 .

.

.

.

.

.

.

.

• Increase the convection coefficient h. This can be accomplished by increasing the fluid flow over the surface since h is a function of the flow velocity and the higher the velocity. the higher the h. Example: a cooling fan.Extended Surfaces / Fins Convection: Heat transfer between a solid surface and a moving fluid is governed by the Newton’s cooling law: q = hA(Ts-T∞). Therefore. one can • Increase the temperature difference (Ts-T∞) between the surface and the fluid. • Increase the contact surface area A. to increase the convective heat transfer. Pradeep datta HT/M9/L1/V1/2004 1 . Example: a heat sink with fins.

A C are all constants.Extended Surface Analysis Tb x dT q x = − kAC dx AC is the cross-sectional area P: the fin perimeter Ac: the fin cross-sectional area q x + dx dq x = qx + dx dx dq conv = h( dAS )(T − T∞ ). where dA S is the surface area of the element dq x Energy Balance: q x = q x + dx + dq conv = q x + dx + hdAS (T − T∞ ) dx d 2T − kAC 2 dx + hP(T − T∞ )dx = 0. Pradeep datta dx 2 HT/M9/L1/V1/2004 . if k.

Extended Surface Analysis (cont. we need to specify two boundary conditions: The first one is obvious: the base temperature is known as T(0) = Tb The second condition will depend on the end condition of the tip Pradeep datta HT/M9/L1/V1/2004 3 . A second .order. ( D 2 − m 2 )θ = 0 dx 2 kAC Characteristics equation with two real roots: + m & . where m = . so that d 2θ hP 2 2 − m θ = 0.m The general solution is of the form θ ( x ) = C1e mx + C2 e − mx To evaluate the two constants C 1 and C 2 .) d 2 T hP − (T − T∞ ) = 0. ordinary differential equation 2 dx kAC Define a new variable θ ( x ) = T ( x ) − T∞ .

4 in HT textbook. p.Extended Surface Analysis (cont. 118.T∞ cosh m( L − x ) θ = = Tb − T∞ cosh mL θb The fin heat transfer rate is dT q f = − kAC ( x = 0) = hPkAC tanh mL = M tanh mL dx These results and other solutions using different end conditions are tabulated in Table 3.) For example: assume the tip is insulated and no heat transfer dθ/dx(x=L)=0 The temperature distribution is given by T ( x ) . the following fins table Pradeep datta HT/M9/L1/V1/2004 4 .

Temperature distribution for fins of different configurations Case Tip Condition A Convection heat transfer: hθ(L)=-k(dθ/dx)x=L B C Adiabatic (dθ/dx)x=L=0 Given temperature: θ(L)= θL Temp. Note: This table is adopted from Fundamentals of Heat and Mass Transfer by Frank Incropera and David DeWitt Pradeep datta HT/M9/L1/V1/2004 5 . Distribution cosh m( L − x ) + ( h Fin heat transfer ) sinh m( L − x ) sinh mL + ( h ) cosh mL mk mk M cosh mL + ( h ) sinh mL cosh mL + ( h ) sinh mL mk mk M tanh mL cosh m( L − x ) cosh mL ( θL θb ) sinh m( L − x ) + sinh m( L − x ) sinh mL M (cosh mL − θ L sinh mL θb ) D Infinitely long fin θ(L)=0 e − mx M θ ≡ T − T∞ . m2 ≡ hP kAC M = hPkA C θ b θ b = θ ( 0 ) = Tb − T ∞ .

The handle of the pot is 20-cm long.Example An Aluminum pot is used to boil water as shown below. and the convection coefficient is 5 W/m2 °C. 3-cm wide. Question: can you touch the handle when the water is boiling? (k for aluminum is 237 W/m °C) T∞ = 25 °C h = 5 W/ m2 °C x 100 °C Pradeep datta HT/M9/L1/V1/2004 6 .5-cm thick. The pot is exposed to room air at 25°C. and 0.

h=5 W/ m2 °C.) We can model the pot handle as an extended surface. k=237 W/m °C. T ( x ) = 25 + 62.138.2) .T∞ θ cosh m( L − x ) = = cosh mL Tb − T∞ θb cosh[3138(0.Example (cont.2(m) Therefore.00015(m2). T − 25 . L=0. M=√(hPkAC)(Tb-T∞)=0. = cosh(3138 * 0.2 − x )] Pradeep datta HT/M9/L1/V1/2004 7 .111θb=0. m=(hP/kAC)1/2=3.111(100-25)=8.005)=0. P=2W+2t=2(0. 100 − 25 .03+0.32 * cosh[3138(0. case B. AC=Wt=0.325(W) T ( x ) . Assume that there is no heat transfer at the free end of the handle. The condition matches that specified in the fins Table.07(m).2 − x )] .

2)=87.Example (cont.2 As shown. it should not be safe to touch the end of the handle Pradeep datta HT/M9/L1/V1/2004 8 .15 0.) Plot the temperature distribution along the pot handle 100 95 T( x ) 90 85 0 0.1)=90. At the midpoint T(0.3°C.1 x 0. temperature drops off very quickly.05 0. At the end T(0.4°C. Therefore.

If a stainless steel handle is used instead.632(W) Very small amount: latent heat of evaporation for water: 2257 kJ/kg. Use the same parameter as before: ⎛ hP ⎞ m=⎜ ⎟ ⎜ kA ⎟ ⎝ C⎠ 1/ 2 = 12.2)=4.0281 Pradeep datta HT/M9/L1/V1/2004 9 . qf=Mtanh(mL)=8. Therefore. the amount of heat loss is just enough to vaporize 0.138*0.325*tanh(3.Example (cont. what will happen: For a stainless steel.) The total heat transfer through the handle can be calculated also. M = hPkAC = 0. the thermal conductivity k=15 W/m°C.47.007 kg of water in one hour.

47( L − x )] 100 75 T( x) 50 25 0 0 0.3 cosh[12.) T ( x ) − T∞ cosh m( L − x ) = Tb − T∞ cosh mL T ( x ) = 25 + 12.1 x 0.05 0. This example illustrates the important role played by the thermal conductivity of the material in terms of conductive heat transfer. not hot at all.2 m) is only 37.3 °C.Example (cont. Pradeep datta 10 HT/M9/L1/V1/2004 .2 Temperature at the handle (x=0.15 0.

Recall: h=5 W/ m2 °C. mss=(hP/kAC)1/2=12.2 − x )] . = = cosh mL Tb − T∞ θb Tss − 25 cosh[12.66(W) Tss ( x ) .03+0.1(W) Mcu= √(hPkssAC) θb=0.2(m) Therefore.Fins-2 If the pot from previous lecture is made of other materials other than the aluminum.00015(m2).47 * 0.142(100-25)=10.47(0. = cosh(12.47(0.K) and copper (385 W/m. L=0. mcu=2.3 * cosh[12.028(100-25)=2.2 − x )] Pradeep datta HT/M9/L1/V1/2004 11 .2) 100 − 25 Tss ( x ) = 25 + 12. P=2W+2t=2(0.005)=0. what will be the temperature distribution? Try stainless steel (k=15 W/m.07(m) AC=Wt=0.K).46 Mss=√(hPkssAC) (Tb-T∞)=0.47.T∞ θ cosh m( L − x ) For stainless steel.

12 0.76 * cosh[2.2 12 HT/M9/L1/V1/2004 .16 0.46(0.46(0.04 0.T∞ θ cosh m( L − x ) For copper.46 * 0.Fins-2 (cont. = = Tb − T∞ θb cosh mL Tcu − 25 cosh[2.2) Tcu ( x ) = 25 + 66.2 − x )] = .) Tcu ( x ) .2 − x )] 100 95 90 T( x ) T ss( x ) T cu( x ) copper aluminum 85 80 75 Pradeep datta stainless steel 0 0. 100 − 25 cosh(2.08 x 0.

Temperature at the end of the handle is 37. correspondingly.3°C. This is because the stainless steel has low thermal conductivity and heat can not penetrate easily into the handle.) • Inside the handle of the stainless steel pot.Fins-2 (cont. • Copper has the highest k and. the temperature inside the copper handle distributes more uniformly. • Question? Which material is most suitable to be used in a heat sink? Pradeep datta HT/M9/L1/V1/2004 13 . temperature drops quickly. Heat easily transfers into the copper handle.

h=5 W/m2.138. m=3.00672 sinh[3138(0.09{cosh(0.K.00672 sinh(0.6276 − 3138 x )} . = cosh(0. . Pradeep datta HT/M9/L1/V1/2004 14 .2 − x )] + 0. .) How do we know the adiabatic tip assumption is good? Try using the convection heat transfer condition at the tip (case A in fins table) We will use the aluminum pot as the example.2 − x )] .00672 sinh(0.T∞ Tb − T∞ = θ cosh[ m( L − x )] + (h / mk )sinh[ m( L − x )] = θb cosh mL + (h / mk )sinh mL cosh[3138(0. M=8.K.6276) T ( x ) = 25 + 62. k=237 W/m.6276 − 3138 x ) + 0.Fins-2 (cont.6276) + 0.325W Long equation T ( x ) .

16 0.75 85 T: adiabatic tip Tc: convective tip T(0. correct choice of boundary condition is not that important here.04 0.2 Note 1: Convective tip case has a slightly lower tip temperature as expected since there is additional heat transfer at the tip.2)=87.08 x 0. (especially for thick fins) Pradeep datta HT/M9/L1/V1/2004 15 . Note 2: there is no significant difference between these two solutions.09 °C 0 0.25 T( x ) 92. therefore.32 °C Tc(0.Fins-2 (cont. sometimes correction might be needed to compensate the effect of convective heat transfer at the end.5 T c( x ) 88.12 0. However.2)=87.) 100 96.

one would like to avoid using the long equation as described in case A. Pradeep datta HT/M9/L1/V1/2004 16 . However. fins table. t With convection LC=L+t/2 insulation Original fin length L L t/2 Then apply the adiabatic condition at the tip of the extended fin as shown above.Fins-2 (cont.) In some situations. The alternative is to use case B instead and accounts for the convective heat transfer at the tip by extending the fin length L to LC=L+(t/2). it might be necessary to include the convective heat transfer at the tip.

75 85 Pradeep datta T(0. 100 96.12 0.2+0.2025)=87.09 °C Tcorr(0.08 0.2025 − x )] .5 T( x ) T c( x ) T corr( x ) 88.2025) .) Use the same example: aluminum pot handle.05 °C slight improvement over the uncorrected solution 0 0. = .16 0. m=3.2)=87.0025=0.04 0. the length will need to be corrected to LC=l+(t/2)=0.25 92.05 * cosh[3138(0.2025 − x )] . 100 − 25 cosh(3138 * 0.32 °C Tc(0.2025(m) Tcorr ( x ) .2 HT/M9/L1/V1/2004 17 . Tcorr ( x ) = 25 + 62.2)=87.T∞ cosh m( Lc − x ) θ = = Tb − T∞ θb cosh mLc Tcorr − 25 cosh[3138(0.Fins-2 (cont.138.

• Thin rectangular fin: Ac=Wt. since t << W Lc=L+(Ac/P)=L+(Wt/2W)=L+(t/2) • Cylindrical fin: Ac=(π/4)D2.Correction Length The correction length can be determined by using the formula: Lc=L+(Ac/P). Lc=L+(Ac/P)=L+(W2/4W)=L+(W/4) Pradeep datta HT/M9/L1/V1/2004 18 . Lc=L+(Ac/P)=L+(D/4) • Square fin: Ac=W2. where Ac is the cross-sectional area and P is the perimeter of the fin at the tip. P=4W. P= πD. P=2(W+t)≈2W.

for an infinitely long fin Their ratio: R(mL)= qf ( q f )∞ = tanh mL Note: heat transfer increases with mL as expected.8 0. Pradeep datta HT/M9/L1/V1/2004 19 .6 R( mL ) 0.8% of the fin heat transfer. a long fin means more material and increased size and cost.762. Question: how do we determine the optimal fin length? Use the rectangular fin as an example: q f = M tanh mL. However. Initially the rate 0 of change is large and slows down 0 1 2 3 4 mL drastically when mL> 2.4 0. R(1)=0. for an adiabatic tip fin 1 0. the longer the fin. the higher the heat transfer.2 ( q f ) ∞ = M . means any increase beyond mL=1 will increase no more than 23.Optimal Length of a fin In general.

8 0. poor fin heat transfer R θ ( x ) 0.2 20 Pradeep datta 0 HT/M9/L1/V1/2004 .15 0. and L=0.05 0.648054 0. good fin heat transfer 1 1 Use m=5.Temperature Distribution For an adiabatic tip fin case: T − T∞ cosh m( L − x ) Rθ = = cosh mL Tb − T∞ High ∆T.2 0.2 as an example: Low ∆T.6 0 0.1 x 0.

Lc=L+(Ac/P)=L+(D/4) • Square fin: Ac=W2. • Thin rectangular fin: Ac=Wt. P=2(W+t)≈2W.Correction Length for a Fin with a Nonadiabatic Tip The correction length can be determined by using the formula: Lc=L+(Ac/P). since t << W Lc=L+(Ac/P)=L+(Wt/2W)=L+(t/2) • Cylindrical fin: Ac=(π/4)D2. P= πD. Lc=L+(Ac/P)=L+(W2/4W)=L+(W/4) Pradeep datta HT/M9/L1/V1/2004 21 . where Ac is the cross-sectional area and P is the perimeter of the fin at the tip. P=4W.

f is the thermal resistance of the fin. the fin resistance can be expressed as (Tb − T∞ ) 1 Rt . f = = qf hPkA C [tanh(mL)] Pradeep datta HT/M9/L1/V1/2004 22 .Fin Design T∞ Tb Total heat loss: qf=Mtanh(mL) for an adiabatic fin. f where Rt . and M= hPkA Cθ b = hPkA C (Tb − T∞ ) kAc Use the thermal resistance concept: (Tb − T∞ ) q f = hPkA C tanh( mL)(Tb − T∞ ) = Rt . or qf=Mtanh(mLC) if there is convective heat transfer at the tip hP where m= . For a fin with an adiabatic tip.

4) kP k⎛ P k ⎛ P ⎞⎞ εf → = ⎜ ⎟⎟ ⎜ hAC h ⎜ ACC ⎟⎠ h⎝ A ⎠ ⎝ In order to enhance heat transfer. However. it can be considered an infinite fin (case D of table3. For an adiabatic fin: εf = qf q = qf hAC (Tb − T∞ ) = hPkA C tanh( mL) kP = tanh( mL) hAC hAC If the fin is long enough. tanh(mL) → 1. mL>2.Fin Effectiveness How effective a fin can enhance heat transfer is characterized by the fin effectiveness εf: Ratio of fin heat transfer and the heat transfer without the fin. ε f ≥ 2 will be considered justifiable If ε f <1 then we have an insulator instead of a heat fin Pradeep datta HT/M9/L1/V1/2004 23 . ε f > 1.

•Conclusion: It is preferred to use thin and closely spaced (to increase the total number) fins. • It seems to be counterintuitive that the lower convection coefficient. Therefore. Fins are usually placed on the gas side. k.Fin Effectiveness (cont. heat fins are more effective if h is low. the higher εf. the higher the P/AC. the fin’s material should have higher thermal conductivity. Pradeep datta 24 HT/M9/L1/V1/2004 . The smaller W. h. But it is not because if h is very high. and the higher εf. AC=W2. Observation: If fins are to be used on surfaces separating gas and liquid.) εf → kP k⎛ P ⎞ = ⎜ ⎟ hAC h ⎝ AC ⎠ • To increase εf. Use a square fin with a dimension of W by W as an example: P=4W. it is not necessary to enhance heat transfer by adding heat fins. (Why?) • P/AC should be as high as possible. P/AC=(4/W).

f It is a ratio of the thermal resistance due to convection to the thermal resistance of a fin.h εf = = = = q hAC (Tb − T∞ ) (Tb − T∞ ) / Rt .h Rt . f Pradeep datta HT/M9/L1/V1/2004 25 .Fin Effectiveness (cont. the fin's resistance should be lower than that of the resistance due only to convection. In order to enhance heat transfer.) The effectiveness of a fin can also be characterized as Rt . qf qf (Tb − T∞ ) / Rt .

q max = hA f (Tb − T∞ ) T(x)<Tb for heat transfer to take place For infinite k T(x)=Tb.Fin Efficiency Define Fin efficiency: η f = qf q max where q max represents an idealized situation such that the fin is made up of material with infinite thermal conductivity. the fin should be at the same temperature as the temperature of the base. Therefore. the heat transfer is maximum Tb x Total fin heat transfer qf x Ideal heat transfer qmax Ideal situation Real situation Pradeep datta HT/M9/L1/V1/2004 26 .

f 1 h Ab T h e f in h e a t tr a n s f e r : q f = η f q m a x = η f h A f ( T b − T ∞ ) qf = T h e r m a l r e s is ta n c e f o r a s in g le f in . b > R t .) Use an adiabatic rectangular fin as an example: η = f = qf q m ax = M ta n h m L = h A f (Tb − T∞ ) h P k A c ( T b − T ∞ ) ta n h m L h P L (Tb − T∞ ) ta n h m L ta n h m L = ( s e e T a b le 3 . w h ere Rt. f o r A b < η f A f Pradeep datta HT/M9/L1/V1/2004 27 .5 f o r η f o f c o m m o n f in s ) mL hP L k Ac 1 Tb − T∞ T − T∞ = b .Fin Efficiency (cont. A s c o m p a r e d to c o n v e c tiv e h e a t tr a n s f e r : R t . b = In o r d e r to h a v e a lo w e r r e s is ta n c e a s th a t is r e q u ir e d to e n h a n c e h e a t tr a n s f e r : R t . f = 1 /(η f h A f ) η f hA f Rt.

Overall Fin Efficiency Overall fin efficiency for an array of fins: Define terms: Ab: base area exposed to coolant qb Af: surface area of a single fin At: total area including base area and total finned surface. At=Ab+NAf N: total number of fins qt = qb + Nq f = hAb (Tb − T∞ ) + Nη f hAf (Tb − T∞ ) = h[( At − NAf ) + Nη f Af ](Tb − T∞ ) = h[ At − NAf (1 − η f )](Tb − T∞ ) = hAt [1 − NAf At (1 − η f )](Tb − T∞ ) = ηO hAt (Tb − T∞ ) NAf At (1 − η f ) 28 qf Define overall fin efficiency: ηO = 1 − Pradeep datta HT/M9/L1/V1/2004 .

f That is.f is the base area (unexposed) for the fin To enhance heat transfer AtηO >> A =Ab+NAb. f )(Tb − T∞ ) = hA where Ab. to increase the effective area ηO At . Pradeep datta HT/M9/L1/V1/2004 29 .Heat Transfer from a Fin Array Tb − T∞ 1 qt = hAtηO (Tb − T∞ ) = where Rt .O = Rt .O hAtηO Compare to heat transfer without fins 1 q = hA(Tb − T∞ ) = h( Ab + NAb.

f Rb=t/(kbA) T1 T∞ T1 T2 Tb T∞ R1=L1/(k1A) Rt .O HT/M9/L1/V1/2004 30 Pradeep datta .O = 1 /( hAtηO ) T2 Tb T1 − T∞ T1 − T∞ q= = ∑ R R1 + Rb + Rt .Thermal Resistance Concept L1 t A=Ab+NAb.

Sign up to vote on this title