Design and Construction of an Induction Furnace (Cooling Sys

YANGON TECHNOLOGICAL UNIVERSITY DEPARTMENT OF MECHANICAL ENGINEERING

DESIGN AND CONSTRUCTION OF AN INDUCTION FURNACE: COOLING SYSTEM

BY

MAUNG THANT ZIN WIN

Ph.D. THESIS

NOVEMBER, 2005 YANGON


DESIGN AND CONSTRUCTION OF AN INDUCTION FURNACE: COOLING SYSTEM

BY MAUNG THANT ZIN WIN

A THESIS SUBMITTED TO THE DEPARTMENT OF MECHANICAL ENGINEERING IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (MECHANICAL ENGINEERING)

NOVEMBER, 2005 YANGON


We certify that we have examined, and recommend to the University Steering Committee for Post Graduate Studies for acceptance of the Ph.D. thesis entitled: "DESIGN AND CONSTRUCTION OF AN INDUCTION FURNACE: COOLING SYSTEM" submitted by Maung Thant Zin Win, Roll No. Ph.D. M.7 (October, 2003) to the Department of Mechanical Engineering in partial fulfilment of the requirements for the degree of Ph.D. (Mechanical Engineering).

Board of Examiners:

1. Daw Yin Yin Tun Associate Professor and Head Department of Mechanical Engineering, Y.T.U.

…………………. (Chairman)

2. Dr. Mi Sandar Mon Associate Professor Department of Mechanical Engineering, Y.T.U.

…………………. (Supervisor)

3. Daw Khin War Oo Lecturer Department of Mechanical Engineering, Y.T.U.

…………………. (Co-supervisor)

4. Dr. Sandar Aung Associate Professor Department of Mechanical Engineering, Y.T.U.

…………………. (Member)

5. Dr. Kyaw Sein Professor and Advisor Ministry of Science and Technology

…………………. (External Examiner)

i

ACKNOWLEDGEMENTS First and foremost, the author sincerely wishes to express my deep gratitude to His Excellency Minister U Thaung, Ministry of Science and Technology, for opening special intensive courses leading to Ph.D Degree in Yangon Technological University. Special thanks are extended to Minister Dr. Chan Nyein, Ministry of Education, for his guidance and kind help, and deep thanks are due to Deputy Minister U Kyaw Soe, Ministry of Science and Technology, for his advice and keen interest to produce the cooling system of induction furnace. The author also wishes to thank Daw Yin Yin Tun, Associate Professor and Head of Department of Mechanical Engineering, for her invaluable guidance and helpful suggestions throughout the study. Associate Professor Dr. Mi Sandar Mon, my thesis supervisor, provided me with expert guidance throughout the study and the author is deeply grateful for it. She was very helpful. Also, Daw Khin War Oo, my thesis co-supervisor, supported me with the helpful suggestions in improving the thesis. Sincere thanks are then extended to Associate Professor Dr. Sandar Aung, for her critical review and inspiring guidance. Special thanks are extended to Professor Dr. Kyaw Sein for his participation in the Board of Examiners of my thesis. His help and advice are gratefully acknowledged. The author shall not forget Ko Cho Min Han, who skillfully drew the necessary figures for my thesis. Furthermore, the author would like to express my heart felt gratitude to my parents and to all my teachers who taught me everything from childhood till now. Finally, thanks are to the persons who contributed directly or indirectly towards the success of this thesis.

ii

ABSTRACT In coreless induction furnaces, water cooling system is the heart of the induction coil which consists of a hollow section of heavy duty and high conductivity copper tubing, and the coil must be water-cooled because of its high temperature about 78ºC. The purpose of this thesis is to prevent overheating and damage to the induction coil due to heat generated by the passage of alternating current to induce the charge around the coil and heat transferred through the refractory lining from the molten metal. For this reason, cooling pond system is theoretically designed and practically constructed for 0.16 ton coreless induction furnace. It is used in two induction furnaces for the alternative melting in foundry shop. The calculations of required pond area and volume are carried out according to the temperature difference between the hot water and cold water. The mass flow rate passing through the inside of induction coil is mainly calculated according to the increasing temperature. For 0.16 ton melting capacity of electric induction furnace, the centrifugal pump, the size which is of 11 kW and pumping capacity 0.69 m3/min is used to suck the amount of water sufficiently. To be a free flow of water, the size of 2.5 inches diameter galvanized iron pipes for inlet and outlet section of water from cooling pond, and 1.5 and 2 inches diameter polyvinyl chloride plastic (PVC) pipes have been used for the connection of pipelines to induction coil, capacity bank and control panel. Moreover, cooling tower system with induced draft counter flow type has been designed for the continuous operating time and mass production in the melting process. In addition, cooling tower is more efficient rather than cooling pond in that the duration of operating time is limited with its volume. As a result, cooling pond surface area 1,000 ft2 and volume 6,000 ft3 are obtained for 0.16 ton melting capacity of two induction furnaces. Finally, their influences and operating capacity on cooling system of induction furnace have been discussed with the recommendations.

iii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS

i

ABSTRACT

ii

TABLE OF CONTENTS

iii

LIST OF FIGURES

vi

LIST OF TABLES

viii

NOMENCLATURE

CHAPTER 1

2

ix

TITLE INTRODUCTION

1

1.1 Objective

2

1.2 Outline of Thesis

2

LITERATURE REVIEW

3

2.1 Electric Melting Furnaces

3

2.1.1 Arc Furnace

4

2.1.2 Induction Furnace

6

2.1.3 Resistance Furnace

9

2.2 Operating Principle of Coreless Induction Furnace

10

2.3 Features of Induction Melting Furnace

12

2.4 Energy Requirements and Coil Cooling Energy Losses

13

2.5 Heat Balance of Induction Furnace

15

2.6 Water Cooling System

17

2.6.1 Water Requirements

19

2.6.2 Effects of Water Quality

20

2.6.3 Water Purification/ Maintenance

20

2.6.4 Filtration

21

2.6.5 Effects of Impurities

21

2.6.6 Energy Water Supply and Cooling System

22

2.7 Types of Cooling Water System for Electric Induction Furnace

23

2.7.1 Cooling Pond System

23

2.7.2 Spray Pond System

24

iv

3

2.7.3 Evaporative Cooling Tower-Open Circuit System

25

2.7.4 Fan-Radiator Closed-Circuit System

26

2.7.5 Water/Water Heat Exchanger Dual System

27

2.7.6 Dual System with Closed-Circuit Cooling Tower

28

2.8 Selection of Cooling System

28

FLOW CALCULATION AND PUMP SELECTION

30

3.1 Consideration of Flow Velocity

30

3.1.1 Specifications of Induction Coil

31

3.1.2 Effect of Electrical Resistance in Induction Coil

32

3.1.3 Heat Generation Rate Calculation

34

3.1.4 Calculation of Heat Transfer Rate in Composite

36

Refractory Shells 3.1.5 Flow Velocity Designation 3.2 Pump Selection

4

5

38 39

3.2.1 Essential Parameters Required in Selection

40

3.2.2 Selection Procedures

40

3.2.3 Calculations for Pump Selection

44

COOLING POND DESIGN

57

4.1 Pond Design Parameters

57

4.2 Conceptual Study for Steady-State Cooling Pond Design

58

4.2.1 Classification of Ponds

59

4.2.2 Equilibrium Temperature and Surface Heat Flux

61

4.2.3 Traditional Model

67

4.3 Design Model Consideration

69

4.4 Design Calculation

72

EVAPORATIVE COOLING TOWER SYSTEM

80

5.1 Cooling Tower Fundamentals

80

5.1.1 Principal Criteria

81

5.1.2 Classification of Cooling Towers

81

5.1.3 Main Components and Tower Operation

84

5.1.4 Cooling Tower Fill

87

5.2 Conceptual Study for Induced Draft Cooling Tower System

89

5.2.1 Cooling Tower Theory

89

5.2.2 Heat-Balanced Process

91

v

6

7

5.2.3 Tower Coefficients

92

5.2.4 Factors Affecting on Cooling Tower Performance

93

5.3 Design Calculations

94

5.4 Operation Considerations

99

RESULTS AND DISCUSSIONS

101

6.1 Flow Velocity Calculation Results

101

6.2 Cooling Pond Performance

102

6.3 Cooling Tower Performance

106

6.4 Process Influence on Tower

107

CONCLUSION, RECOMMENDATION AND

110

FURTHER SUGGESTIONS 7.1 Conclusion

110

7.2 Recommendation

111

7.3 Further Suggestions

112

REFERENCES

113

APPENDICES

117

APPENDIX A PROGRAM

117

APPENDIX B

GRAPHS

121

APPENDIX C TABLES

125

vi

LIST OF FIGURES Figure

Page

2.1. Electric Arc Furnaces

4

2.2. Pictorial Diagram of Coreless Induction Furnace

7

2.3. Pictorial Diagram of Channel Induction Furnace

8

2.4. Pictorial Diagram of Electric Resistance Furnace

10

2.5. Simplified Cross Section of Coreless Induction Furnace

10

2.6. Melting Design Difference between Heel Method and Batch

11

2.7. Heat Balance Diagram of Crucible Type Induction Furnace

15

2.8. A Sample Induction Coil with Cooling Water

18

2.9. Sample of Damaging Induction Coil

18

2.10. Typical Sketch of Cooling Pond System

23

2.11. Sample Spray Pond System

25

2.12. Open-Circuit System with Evaporative Cooling Tower

25

2.13. Fan-Radiator Closed-Circuit System

26

2.14. Dual System with Water/Water Heat Exchanger

27

2.15. Dual System with Closed-Circuit Cooling Tower

28

3.1. Internal View of 0.16 ton Coreless Induction Furnace

30

3.2. Variation of Resistance with the Temperature

33

3.3. Temperature Distribution for a Composite Refractory

36

Cylindrical Shell 3.4. Approximate Relative Impeller Shapes and Efficiency Variations

43

for Various Specific Speeds of Centrifugal Pumps 3.5. Functional Layout Diagram of 0.16 ton Cooling Pond System

45

3.6. Sketch of Flow Branches in Pipes

45

3.7. Pipe Network for Joint E

46

3.8. Sketch of Suction and Discharge Line in Pumping System

49

4.1. Correlation between Pond Number, IP and Normalized

60

vii Temperature Gradient, ∆T v / ∆To 4.2.

Components of Surface Heat Transfer

63

4.3. Example of Plug-Flow Pond

67

4.4.

Schematic Elevation View of Completely Mixed Pond

67

4.5.

Illustrative Example of Cooling Pond Model

69

4.6.

Illustration for the Equilibrium Condition

70

4.7.

Heat Transfer Mechanism in Cooling Pond and

72

the Symbolic Notations 5.1.

Mechanical Draft Cooling Towers

82

5.2.

Natural Circulation Cooling Towers

83

5.3.

Cutaway View of Induced Draft Counterflow Cooling Tower

85

5.4.

Drift Eliminator used in Induced Draft Counterflow Cooling Tower

86

5.5.

Water Distribution System

86

5.6. Illustration of Typical Splash Fill

87

5.7.

Illustration of Typical Film Fill

87

5.8. Typical Film Fill Shape and Texture

88

5.9. Process Heat Balance Diagram of Counterflow Cooling Tower

91

5.10.

Enthalpy-Temperature Diagram of Air and Water

96

5.11.

Toolkit Software Dialog Box

98

5.12.

Output Results Comparison

98

6.1. Cooling Pond Performance Curve

102

6.2. Effect of Cooling Pond Configurations

104

6.3.

107

Comparison of Different Temperature Ranges at Constant Water Quantity

6.4.


108

by Changing L/G Ratio 6.5.


109

at the Close Approach Condition B.1. Skin Effect in Isolated Rounded Copper Tubings

121

B.2. Composite Rating Chart for a Typical Centrifugal Pump

122

B.3. Moody's Diagram

123

B.4. Nomograph of Cooling Tower Characteristics

124

viii

LIST OF TABLES Table 2.1. Induction Furnace Categories

Page 8

2.2. Electricity Use in Electric Melting Furnaces

14

3.1. Specifications of Induction Coil

32

3.2. Pumps Classes and Types

42

3.3. Total Losses for Pipe Sections

50

3.4. Operating Speed versus Required Specific Speed

52

3.5. Pump Types Listed by Specific Speed

53

3.6. Atmospheric Pressures at Various Altitudes

55

4.1. Iterative Solutions of Equilibrium Temperature

74

4.2. Resulting Values of the Water Temperature and the Operating Time

78

5.1. Enthalpy Difference by Using the Numerical Integration Method

95

5.2. Enthalpy Difference by Using the Chebyshev Method

97

6.1. Comparison of Process Variables in Tower Design

108

C.1. Pipe Roughness - Design Values

125

C.2. Resistance in Valves and Fitting expressed as Equivalent Length

125

in Pipe Diameters C.3. Properties of Water at Various Temperatures

126

C.4. Comparison of Different Roofing Materials

126

C.5. Characteristics of Modern Pumps

127

ix

NOMENCLATURE

A

area of pipe line, m2

Ap

pond surface area, m2

A1, Ai

copper conductor area in general, and for inner area, respectively, cm2

a, b

regression coefficient

C

cloud cover of the sky

Cp

specific heat of constant pressure, kJ/kg K

D

diameter of a pipe, m

Do

outer diameter of induction coil, cm

Di

inner diameter of induction coil, cm

Dv

vertical dilution

ea

vapour pressure, mmHg

esat

saturation vapour pressure, mmHg

E

thermal energy, W

f

rated frequency, Hz

f

fraction factor f′

internal fraction factor

f (W1 ), f (W2 )

wind speed function for analytical, and empirical, respectively

Fo′

densimetric Froude number

g

gravitational constant

G

air loading, kg/(hr m2)

h

loss, m

hl

energy losses from the system, m

H

enthalpy of air-water vapor mixture at the wet bulb temperature, J/kg

H'

enthalpy of air-water vapor mixture at the bulb water temperature, J/kg

Ha

atmospheric pressure, m

x Hf

total friction-head loss, m

Hn

net heat exchange rate, W

Hp

pond depth, m

Hs

total suction head or lift, m

Ht

actual total head on the pump, m

Hts

total static head, m

HDU

height of a diffusion unit, m

I

rated alternating current, A

IP

pond number

Isc

solar constant

kA, kB, kC

thermal conductivity for silica lining, for asbestos sheet, and for asbestos cloth, respectively, W/mºC

kr

water retention rate, m/min

kT

thermal rate, min-1

K

heat exchange coefficient, W/m2ºC

Kx

overall enthalpy transfer coefficient, kg/(hr m2)

KxaV/L

tower coefficient

l

length of copper conductor, m

L

liquid loading, kg/(hr m2)

L1

height of crucible, m

L′

length of flow path, m

m

slope of the straight-line portion of the curve

m&

water mass flow rate, kg/s

n

Julian day number

nd

number of diffusion unit

N

pump rotative speed, rpm

Ns

pump specific speed, rpm

p

pressure, Pa

ps

possible sunshine hour, hr

Pg

power loss of induction coil, kW

qr

heat transfer rate, kW

Q

water outflow rate, m3/min

Qt

total heat transfer rate, W

Qv

volume flow rate, m3/min

xi QT

total volume flow rate, m3/min

r

pond cooling capacity

r1, r2, r3, r4

radii at various interfaces, m

R

water inflow rate, m3/min

Re

Renold number

RH

relative humidity, %

R1 , R DC1

resistance at temperature t1, and at temperature 20ºC, respectively, Ω

R2 , R DC2

resistance at temperature t2, and at temperature 60ºC, respectively, Ω

S

heat transfer surface, m2

S

monthly average of the sunshine hours per day at the location, hr

So

monthly average of the maximum possible sunshine hour per day at the same location, hr

t

operating time, hr

t1, t2

temperature of the copper tubing related to the resistance R1, and R2, respectively, ºC

tc

coil thickness, cm

T

temperature, ºC

∆Tv

average temperature difference between the surface and bottom of the pond, ºC

∆To

temperature difference between the surface and the bottom of the pond, ºC

Ti ∗

normalized intake temperature, ºC

v, vE, vi

flow velocity, for joint E and for the inside of induction coil, respectively, m/sec

V

volume, m3

w

pond width, m

Wc

circulating water flow rate, m3/min

Wd

drift loss, m3/min

We

water evaporative loss, m3/min

xii Wm

make-up water, m3/min

W2

wind speed at two meters above the water surface, mph

z, zE

elevation in general, and for joint E, respectively, m

Z

height of cooling tower, m

Greek Letters

α1

temperature coefficient of resistance

αE

kinetic energy coefficient

β

coefficient of thermal expansion

β∗

proportional factor

cˆ

specific heat of water, J/kgºC

∈

roughness, mm

υ

kinematic viscosity, m2/min

ρ

water density, kg/m3

ρ1

resistivity, µΩcm

φ

latitude of the location, degree

φn

net solar heat flux, W/m2

φ sn

net solar (short-wave) radiation, W/m2

φ an

net atmospheric (long-wave) radiation, W/m2

φbr

back (long-wave) radiation, W/m2

φe

evaporative heat flux, W/m2

φc

conductive heat flux, W/m2

φs

solar radiation at water surface, W/m2

φ sr

reflected solar radiation, W/m2

φa

atmospheric (long-wave) radiation, W/m2

φ ar

reflected atmospheric radiation, W/m2

φ sc

extraterrestrial solar radiation, kJ/m2. day

φ sc

clear sky solar radiation, kJ/m2.day

ωs

sunset or sunrise angle, degree

δ

declination angle, degree

o

xiii Subscripts a

ambient air

atm

atmosphere

AC

alternating current

b

pond number

c

copper conductor material

d

dew point

DC

direct current

E

equilibrium

i

inlet into the pond

m

major

m,i

entering water into the coil

m,o

leaving water from the coil

n

minor

o

outlet from the pond

p

pond

s

surface

sd

static discharge head

sl

static suction lift

s,1

molten metal

s,2

silica lining

s,3

asbestos sheet

s,4

asbestos cloth

t

tower

w

wet bulb

1

hot water

2

cold water

1

CHAPTER 1 INTRODUCTION

The basic metal melting processes require application of heat to raise the metals to their respective melting points. The major melting processes available for foundry industries include electric induction furnace, arc furnace, resistance furnace, gas furnace and cupola furnace. Among them, the electric induction furnace is suitable for not only ferrous and non-ferrous applications but also high temperature melting because of its energy concentration, and installation space is reduced as compared with other types of melting furnace. Especially, coreless induction furnaces are used for the various types of metal. An induction furnace consists of a refractory structure surrounded by high conductivity copper tubing with the cooled water in which the alternating current is passed. This current generates a magnetic field that induces a current on the surface of the metal. The heat generated by this current is conducted into the metal, causing melting. Heat carried away through the refractory lining due to the molten metal inside the crucible, and heat generated by the magnetic field (frequency of the power) and its intensity (power input) inside the induction coil itself, are simultaneously conducted and reach the water-cooled coil which is wound into a helical coil. Its heat causes the melting effect to the water-cooled coil. Not to be damaged and not to melt the induction coil, it is essential for the water cooling system to feed the cooling water to the coil. There are different varieties of cooling system used in induction furnaces. Most of the newer coreless induction melting system uses a recirculating system for getting a great quality of cooling water. To be more efficient and effective, some foundry industries are using the cooling ponds, cooling towers, fan radiators, and heat exchangers for operating continuous batch method during the day. Nowadays, industrial zones are rapidly growing and the demand of coreless induction furnace for foundry industries is also increasing. In Myanmar, it has the promising regions for installing and setting up the induction furnaces to produce the good quality products more efficiently. If the induction furnaces can be built in foundry industries locally and commercially, it will save cost, and improve the

2 productivity towards the industrialized nation. Thus, the design and construction of an induction furnace essentially requires careful selection, installation, and maintenance of the water cooling system. Here, the further investigations of mostly used cooling system such as cooling pond and cooling tower system are of broad interest to design more compact and efficient in coreless induction furnace.

1.1. Objective The objectives of the present study are: (a) To design and construct the cooling pond system for 0.16 ton melting capacity. (b) To design the evaporative cooling tower (induced draft counterflow type) system for the continuous operating time and mass production in melting process. (c) To support the foundry industries in melting with coreless induction furnace where the cooling system is an essential part of furnace.

1.2. Outline of Thesis This research is directed to the understanding of the design and construction of an induction furnace with water cooling system. The objectives and outline of the thesis are expressed in chapter one. In chapter two, the relevant literature on cooling system of coreless induction furnace is reviewed. There are significant differences among cooling systems. Flow calculation and pump selection of cooling pond system are described in chapter three. In chapter four, design and calculation of cooling pond system is presented by using the concepts of equilibrium temperature and surface heat flux. Theoretically, it describes design processes of the evaporative cooling tower system (induced draft counterflow type) in chapter five. The results and discussions on the study with all the problems are presented in chapter six. Finally, conclusion, recommendation and further suggestions are expressed in chapter seven.

3

CHAPTER 2 LITERATURE REVIEW

This chapter covers the literature review of electric melting furnace essentially required in foundry sector without any calculation for design, and energy requirements and cooling coil energy losses. Various types of water cooling system mostly used in induction furnaces are described with the necessary diagrams. Water related problems and effects of impurities for induction melting system are presented in this chapter.

2.1. Electric Melting Furnaces In electric melting furnaces, energy is introduced by radiation, convection, or induction directly to the metal to be melted. Raw ferrous materials consist mostly of scrap and some cold pig iron. For this reason, the electric furnace plays an important role in the recovery and recycling of waste iron resources. In area where an abundant supply of scrap and electric power are available, the properties of steelmaking via the electric furnace route is relatively high, because both energy consumption and equipment investment are substantially smaller than via the integrated route using a blast furnace and blast oxygen furnace to produce steel from ore. They are being increasingly used for melting metal and many new and improved types of furnace have been produced in year by year and installed at foundries. Electric melting methods are flexible in terms of the metal charged and can have very high melting rates. Their relative importance and the various types can be seen in the order of their industrial significance. Electric melting furnaces are usually divided into three main classes according to the method of pouring the metal from the crucible, the heating method, and several configurations. They are:

1. Arc furnace 2. Induction furnace 3. Resistance furnace

4 2.1.1. Arc Furnace Electric arc furnaces are refractory-lined melting furnaces that obtain heat generated from an electric arc within the furnaces. They are used more extensively for steelmaking and the other majority of applications, including the melting of gray iron, brass, bronze and gunmetal, as well as many nickel alloys, because its capacity is large and production efficiency is high. They are also capable of melting a higher fraction of alloy scraps. There are two main types of arc furnace, the direct arc and the indirect arc, as shown in Figure 2.1.

Power lead Carbon electrodes

Door Spout

Slag Metal Rammed hearth

(a) Direct Arc Furnace

Water-cooled roof Upper electrode (cathode)

Water-cooled panel

Eccentric bottom taphole

Bottom electrode (anode) Tilting device

(b) Indirect Arc Furnace

Figure 2.1. Electric Arc Furnaces

5 In Figure 2.1. (a), direct arc furnace is so called because an arc is struck directly between the electrode and the metal to be melted. The electrodes are of graphite or amorphous carbon, and the furnaces are either single-phase unit for very small furnaces or more generally, three-phase unit with three overhead, vertically disposed electrodes suspended over what is normally a bowl-shaped refractory hearth. Practically all modern arc furnaces are circular in plan, the kettle-shaped structure with a removable lid, with refractory sidewalls and a domo-shaped roof provided with holes for inserting the electrodes. The carbon electrodes provide the current for the process. They are totally removable in an upward direction to allow the top of the furnace to be removed. The tapping spout is used at the end of the process to allow the molten steel to be poured from the furnace. During the process it is sealed to keep the heat in. The operating door on a top-charged furnace is used for making alloy of slag additions, for rabbling the molten metal and for removing the slag if necessary. The furnace can usually be tilted backwards to assist this operation. Direct arc furnaces are either acid or basic-lined, depending on the melting operation to be carried out. Basic linings are used for steelmaking when sulphur and phosphorus removal are required and are generally recommended for high-alloy steels, such as stainless and manganese steels. Acid linings consist entirely of siliceous materials and are restricted to the melting of cast iron and the production of steel castings from scrap requiring no removal of sulphur and phosphorus. The changing process to the furnace is in itself damaging the refractory lining by both impart and the chilling effect of the cold scrap. The aggressivity to the refractory lining is further increased by rapid temperature increase during melting, combined with the attack by slag fluidizers such as fluorspar. Preferential attack of the refractory lining occurs in the hot spot areas (opposite the electrodes) caused by flare, and at the slag line, owing to low basicity slags, and high FeO slags, often employed to aid phosphorous removal. Indirect arc furnaces are so called because the arc is struck between two carbon electrodes and is therefore independent of the charge, which is heated indirectly by radiation. A typical indirect electric arc furnace is shown diagrammatically in Figure 2.1. (b). The efficiency of heating, melting, and decarburization in the indirect arc furnace has been substantially increased by adopting an ultra high-power transformer and an oxy-fuel burner, as well as by supplying coal power and pure oxygen gas.

6 Cooling the furnace walls and ceiling with water-cooled panels have also been enhanced, enabling an increase in production efficiency from 80 to 120 ton/h. The indirect arc furnace offers lower unit consumption of power, electrodes, and refractories, and both noise and flicker are also lower. The preheating and continuous charging equipment for scrap decrease the energy consumption because preheating is carried out by the high temperature exhaust gas, and heat loss by opening the furnace lid during conventional scrap charging can be prevented. The eccentric bottomtapping allows efficient tapping without tilting the vessel, and is desirable for maintaining the cleanliness of the molten steel, because the carry over of oxidizing slag into the ladle during tapping can be prevented.

2.1.2. Induction Furnace Electric induction furnace is used in both ferrous and nonferrous melting applications. It is also an AC electric furnace in which the primary conductor generates, by electromagnetic induction, a secondary current that develops heat within the metal charge. Many small furnaces are being used by the foundry can be operated in several configurations, including single furnace system, tandem operation, melter and holder configuration, and power sharing. In the conventional single furnace system, each furnace body is supplied from its own power supply. In tandem operation, two furnace bodies (usually identical) are fed from a single power supply that is switched from one furnace to the other. In melter/holder systems, an additional small power supply is used for holding requirements. The power sharing configuration is similar to melting/ holding except that a single power supply simultaneously provides melting power to one furnace and holding power to the second. In both these configurations, the two furnaces alternate in their melting and pouring roles. Metal production can be increased by up to 20 percent with this type of operation presented by Mortimer [1]. The advantages and disadvantages of induction melting systems are: Advantages -

The system permits but does not require the use of a slag.

-

The system exhibits good melt agitation, relatively easy fume control and rapid heat-up.

-

It is not as inherently dusty as electric arc melting, producing only 20 percent as much effluent dust.

7 Disadvantages -

There is an increased risk of cross-contamination between melts due to reactions between refractory lining and the metal and also the slag.

-

Molten slag is removed by skimming for which the furnace may be opened releasing fumes and dust. There are two main types of induction furnace. They are coreless type

induction furnace and core or channel type induction furnace.

(i) Coreless Induction Furnace In a coreless induction furnace, a water-cooled helical copper coil surrounds a refractory-lined cavity containing the charge material, as shown in Figure 2.2. An induced current is produced in the charge material by an alternating current in the coil. Once the charge is molten, stirring action occurs as a result of the interaction of currents in the melt with the magnetic field.

Steel shell Cooling coil Magnetic yoke Power coil Refractory lining Cooling coil

Figure 2.2. Pictorial Diagram of Typical Coreless Induction Furnace

Stirring velocity increases at high powers and lower frequencies. The amount of stirring is characterized by the velocity of the molten metal circulation as well as the resulting height of the molten metal meniscus. Horwath et al. [2] classified three categories of induction furnace depending on the capacity and melting rate required, and the frequency of the current supplied as shown in Table 2.1.

8 Table 2.1. Induction Furnace Categories Frequency Designation

Frequency (Hz)

Mains (or line)

50-60

Low

150-500

Medium or high

500-10,000

For melting high melting point alloys, all grades of steels and irons as well as many non-ferrous alloys, the coreless induction furnace has been widely used in foundry as the crucible furnace. This furnace can be used for remelting and alloying because of the high degree of control over temperature and chemistry while the induction current provides good circulation of the melt.

(ii) Core or Channel Induction Furnace Another type of induction melting furnace is the channel furnace or core type induction furnace. The configurations may be horizontal drum type furnace or semidrum or low-profile furnace with removable cover or vertical type furnaces. In a coreless induction furnace, the power coil completely surrounds the crucible. In a channel furnace, a separate loop inductor is attached to the upper-body, which contains the major portion of the molten metal bath.

Movable lid Cover plate

Upper case lining Back-up castable Insulating brick

Pouring spout Furnace platform Upper case assembly Upper case hearth Throat Blasch inductor lining Transformer Back-up castable Coil core Hydraulic cylinder Bushing Inductor assembly

Figure 2.3. Pictorial Diagram of Channel Induction Furnace

9 Attached to the steel shell and connected by a throat is an induction unit which forms the melting component of the furnace. The induction unit consists of an iron core in the form of a ring around which a primary induction coil is wound. This assembly forms a simple transformer in which the molten metal loops comprise the secondary component. The heat generated within the loop causes the metal to circulate into the mail well of the furnace. The circulation of the molten metal effects a useful stirring action in the melt. A vertical channel furnace may be considered a large bull ladle or crucible with an inductor attached to the bottom. In Figure 2.3, it is illustrated that the furnace has insoluble components, such as slag, accumulate over time in the induction loop or throat area. Buildup on the sidewalls of channel furnaces is also a common occurrence. Channel induction furnaces are commonly used for melting low melting point alloys and or as a holding and superheating unit for higher melting point alloys such as cast iron. They can be used as holders for metal melted off peak in coreless induction units, thereby reducing total melting costs by avoiding peak demand charges. Channel induction melting furnaces have been built with capacities exceeding 100,000 pounds. Overall required efficiency should be around 75 percent. Channel induction furnaces have capacities in the range of 1 ton to 150 tons.

2.1.3 Resistance Furnace The electrical- and heat-resistance reverberatory melting furnace is used for zinc and aluminum melting. This furnace is constructed with an aluminum-resistant refractory lining and a structural steel shell. The furnace is heated by silicon carbide or carbon electrode or other resistance elements mounted horizontally above the both. Heat is transferred through direct radiation from the refractory roof and sides. The details are seen in ACMA et al. [3]. Another type of electric resistance furnace uses electric immersion-type elements. The elements are inserted into silicon carbide tubes that are immersed in the molten aluminum. Through radiation, the element passes its heat to the silicon carbide tube. Through conduction, the tube releases its heat into the bath. To clarify the structure of electric resistance furnace, the example of electricresistance ash melting furnace is shown in Figure 2.4 and it uses carbon electrodes and performs the reduction melting treatment of ash in a fully closed structure. Molten

10 slag and molten metal are separated by the difference in specific gravity and each has a separate discharge port. Molten slag is discharged utilizing the head pressure. Power supply Incineration ash + Fly ash

Exhaust gas

Ash layer

Radiated heat transfer

Molten slag layer

Heat convection Molten metal

Molten metal layer Molten slag

Figure 2.4. Pictorial Diagram of Electric Resistance Furnace

2.2. Operating Principle of Coreless Induction Furnace The principle of operation of the coreless induction furnace is the phenomena of electromagnetic induction. Many induction furnaces are widely constructed by using the phenomena of electromagnetic induction. All electrically conductive materials can be heated quickly and cleanly with pollution free induction heating. A simplified cross section of a coreless induction furnace with the molten charge and the crucible lining is shown in Figure 2.5. It is composed of a refractory-lined container with electrical current carrying coil that surrounds the refractory crucible. Holding the molten container which is surrounded by a water cooled helical coil is connected to a source of alternating current. A metallic charge consisting of scrap, pig iron and ferroalloys are typically melted in such a container. Electrical current in the coil forms a magnetic field, which in turn creates thermal energy, melting the charge.

Figure 2.5. Simplified Cross Section of Coreless Induction Furnace

11 Otherwise, the induction (generation) of the electrical current in a conductive metal (charge) placed within a coil of conductor carrying electrical current is known as electromagnetic induction of secondary current. The magnetic currents in the molten metal cause an intense stirring action, thus ensuring a homogenous liquid. During the melting process, slag is generated from oxidation, dirt, sand and other impurities. Slag can also be generated from the scrap, erosion and wear of the refractory lining, oxidized ferroalloys and other sources. It normally deposits along the upper portion of the lining or crucible walls and above the induction coils. The hottest area of high frequency coreless induction furnaces is at the mid-point of the power coil, where insufficient metal turbulence from magnetic stirring occurs. Two methods or melter are used for operating a coreless induction furnace. In the heel method (also called “tap and charge”), a portion of the liquid charge is retained in the furnace and solid charge material is added. The batch method requires the furnace to be completely emptied between melts. Batch melting on a large has become more common for the development of reliable high-power components for variable frequency equipment and technology that allows utilization of full power input during the entire melting cycle. The energy losses associated with holding iron between melts, as well as the larger overall furnace sizes resulted in high overall energy consumption rates. The basic design differences between heel melt and batch melt induction furnaces are shown in Figure 2.6.

Metallic Charge Water-Cooled Induction Coils Molten Metal Heel

Heel Melter

Refractory Lined Steel Shell

Batch Melter

Figure 2.6. Melting Design Difference between Heel Method and Batch Method

12 The older power supplies were also very inefficient, with losses approaching 40 percent. The heel was used primarily to help reduce stirring associated with line frequency melting, and it also required that charges be preheated to ensure that no wet charges were put into the molten iron in the furnace heel. As more sophisticated solid-state power supplies with increasingly higher power ratings become available, the “batch” furnace increases in numbers. A batchmelting furnace empties the furnace after each melting cycle, reducing the holding power requirements. Over time, methods were developed to increase the frequency of the power supplies, allowing for increased power densities and smaller furnace sizes. Another inherent advantage of the batch induction melter is that when a magnetic charge such as solid scrap iron and cold pig iron are melted, the coil efficiency can be as high as 95 percent, compared to 80 percent when heating the molten bath in a heel melter. Hysteresis losses associated with induction heating of a solid ferrous material are responsible for this increased coil efficiency during the first part of the melting cycle.

2.3. Features of Induction Melting Furnace In metallic material placed in magnetic field generated by the current in induction coil of the furnace, electromotive force is induced by the action of electromagnetic induction, and induced current flows to heat up the material by its Joule’s heat. Compared to other types of melting furnace, induction furnace has the following features: 1. Its heat efficiency is high because the material is directly heated by electromagnetic induction. 2. No carbon dioxide is produced and little smoke and soot is emitted because cokes are not used as fuel. 3. Metal loss by oxidation is little, thus little contamination of metal because of heating without air. 4. Temperature control is simple, uniform composition of metal product is attained by agitation effect and alloyed cast iron is easily produced. 5. Induction melting is suitable for high temperature melting because of its energy concentration, and installing space is reduced as compared with other types of melting furnace.

13 6. It is possible to melt not only steels very low in carbon but also ferrous and non-ferrous metals because there are no electrodes in arc furnace and resistance furnace. 7. As the electricity causes heat in an induction furnace, and the molten metal/air interface is relatively small, off-gas volumes are smaller for induction furnaces than for electric arc furnace given by A.D. Little [4].

2.4. Energy Requirements and Coil Cooling Energy Losses The overall efficiency of coreless induction furnaces depends on furnace operating parameters and factors related to the charge. Energy consumption in coreless induction furnaces is affected by the contaminants (e.g. rust, sand, oil, water, coatings) on the charge since these materials contribute to slag formation. Removing the slag requires additional time during the melt cycle, thereby lowering the efficiency. About 20 percent more energy is required to melt virgin gray iron in coreless induction furnaces than using scrap metal. Researchers theorize that it takes a higher temperature and longer melting time to melt the virgin material to produce carbon. These differences between virgin materials and scrap have not been shown, however, for carbon and low-alloy steel. Further details can be found in Horwath et al. [2]. Other variables affecting energy use during coreless induction melting include the melting method (heel versus batch); power application (step power versus full power); use of covers; and furnace condition (e.g. hot, medium, or cold). For ferrous materials, heel melting typically requires less energy than batch melting (in the order of 5 percent less for stainless steel), as does the use of a hot furnace (about 2 percent to 4 percent less for gray iron and low-alloy steel compared to cold conditions). Coreless induction melting furnaces have electrical efficiencies in the range of 76 percent to 81 percent although the efficiency of an inductor is around 95 percent. Induction furnaces operated in tandem can achieve a maximum electric power utilization exceeding 80 percent (excluding power plant losses). About 75 percent of the energy delivered to the furnace is used for increasing the temperature of the metal. The main source of energy loss is via the coil water cooling system, typically a 20 percent to 30 percent loss. The above energy percents are given by ACMA et al. [3], and Smith and Bullard [5]. Other energy loses in a coreless induction furnace come from -

conductive losses through the lining,

14 -

heat losses associated with the slag, and

-

radiation losses when the furnace lid is open. Heat losses associated with slag are a function of the temperature and

composition of the slag produced. The heat content of a typical slag in furnace is about 410 kWh/ton at 1,538ºC. Unless large quantities of slag are produced, the heat loss due to slag does not detract substantially from the overall performance of the furnace [6]. Radiation heat loss from an uncovered molten bath and the bottom of an opened cover can reach 130 kW for a 10-ton furnace. However, radiant heat loss caused by iron melting is less than that by aluminum melting. Table 2.2 summarizes the energy requirements for various types of electric melting furnaces.

Table 2.2. Electricity Use in Electric Melting Furnaces Electricity Use in Electric Melting Furnaces (kWh/metric ton of metal) Induction

Electric Arc

Electric-Resistance

[106 Btu/tona]

[106 Btu/tona]

Furnace [106 Btu/tona]

520 – 800b [5.0 – 7.6]

500 – 600 [4.3 – 5.2]

600 – 825 [5.2 – 7.9]

500 – 550c [4.3 – 4.8] Sources: Smith and Bullard (1995), Booth (1996) and Process Metallurgy International (1998) a

Using electricity conversion factor of 10,500 Btu/kWh.

b

Ferrous melting. Medium frequency coreless. When an ancillary equipment energy use is included, the tool ranges from 550 to 650 kWh/metric ton of metal.

c

Molten, efficient furnaces. Energy consumption for medium-frequency induction melting is generally in

the range of 520 to 800 kWh/metric ton. The use of furnace covers can reduce melting-rated energy consumption to as low as 500 kWh/metric ton. Allowing for holding power requirements and ancillary equipment, overall energy consumption is reported to be in the range of 550 to 650 kWh/metric ton.

15 With modern, efficient, solid state power electronics, the energy required in many induction furnaces can be as low as 500 kWh/metric ton for aluminum or iron at high utilization rates. Energy consumption for electric arc furnaces ranges from 450 to 550 kWh/ton of charge, depending on the scrap type and length of time heat is applied. For the electric resistance furnace, the only heat loss is through the shell and from exposed radiant metal surfaces.

2.5. Heat Balance of Induction Furnace As the induction furnace is operated with the large amount of temperature, heat balance of the furnace must be understood fully to make the proper decision about cooling effects inside the induction coil to resist the overheating condition and power source side such as frequency conversion equipment and power-factor improving capacitor. Efficiency of induction furnace is expressed as a total, deducting electrical and heat transfer losses. Heat balance diagram of crucible type induction furnace is shown is Figure 2.7. Input 100% Transformer (1)

Input 100% Water-cooled Transformer cable (1) (1.5)

Inverter (4)

Coil (16)

Bus bar

Slag, etc. (1.5)

condenser

(2) Total efficiency 69%

(a) Distribution of losses in high-frequency of furnace. Heat loss (%) is given in ( ).

Heat conduction (3) Heat radiation (2)

Water-cooled cable (1.5)

Bus bar condenser (2)

Coil (17)

Total efficiency 67%

Heat conduction (7) Heat radiation (4.5)

(b) Distribution of losses in low-frequency furnace. Heat loss (%) is given in ( ).

Figure 2.7. Heat Balance Diagram of Crucible Type Induction Furnace Source: Energy Conservation in Iron Casting Industry (1998)

In above figure, 100 percent of input energy is used fully in both of these furnaces; high-frequency and low-frequency crucible type furnace which have electrical and heat losses. Electrical losses consist of transformer, frequency converter, water-cooled condenser, bus bar, wiring, cable and coil. Loss in coil is an essential factor, on which the furnace capacity depends. Heat losses in induction furnace

16 consist of conduction loss of heat escaping from furnace wall to coil side, radiation loss of heat released from melt surface, absorption loss in ring hood and slag melting loss. Heat efficiency of high-frequency furnace (69%) is slightly larger than that of low-frequency furnace (67%). Low-frequency furnace is larger in heat loss (conduction and radiation) due to long melting time, while high-frequency furnace is larger in electrical loss (transformer, inverter and bus bar) due to short melting time. To improve heat efficiency of furnace, the proper decision about the kind of material, size and shape of charging materials to be melted, melting amount, connection with pouring line and layout of the melting shop should be made and adjusted carefully by user’s side. Induction furnace equipment should be melted with minimum distance between each of equipment to reduce wiring losses. To reduce the wiring losses remarkably, it is essential to shorten the distance between furnace body and power-factor improving capacitor as very large current flows between them. Moreover, skin effect and effect of agitation are considered to improve the heat efficiency and induction current flows concentratedly in the surface of material to be melted. This concentration of current becomes more remarkable as the frequency become higher, resulting in better heating efficiency. Diameter or thickness of material to be melted in the furnace may be decreased accordingly as the frequency becomes higher when cast iron is melted in high-frequency induction furnace, there is practically no limitation in its size, but in low-frequency furnace when starting with cold metal, melting has to be started only by the use of starting block. Continuous melting is to be preformed with residual molten metal. In the effect of agitation, molten metal is agitated to raise its surface in the center because molten metal is excited by current opposite to current flowing in induction coil. Surface of molten metal is raised higher as frequency becomes lower. So, agitation of molten metal occurs stronger in low-frequency furnace than in highfrequency. This effect of agitation makes it possible to ensure uniform temperature of molten metal and its uniform quality as well as to promote entrapment of material charged and fusion of chemical composition adjusting agents, specially carbon addition. In this respect, as compared with low-frequency furnace, high-frequency furnace can be charged with larger electric power at the same agitation degree, which will speed up the melting and improve the furnace heat efficiency because highfrequency furnace can be operated with power density about three times larger than low-frequency furnace.

17 To improve the heat efficiency in operating condition, the following should be carried out as: (a) Lower tapping temperature To keep the tapping temperature lower, it is necessary to take care throughout measurement such as ladle traveling distance and preheating and covering of ladle. (b) Close furnace cover In practice of furnace operation, especially in case of small-sized furnace, furnace cover sometimes remains open carelessly. It is important to train personnel and make necessary preparation so as to charge materials and adjusting agent regulator as quick as possible. (c) Required temperature and duration for melting metal Molten metal should be held, when required, at low temperature, or turn off power supply. Preparatory operations should certainly be performed so that there is no unmatching with mold assembly or waiting for crane. (d) Dust collecting hood Dust collecting degree and time should be controlled according to furnace running conditions. (e) Clean of sand, rust and other dirts Sand or rust adhered to cast iron or steel scrap may react with furnace refractory to form slags. Power loss at 1500ºC is about 10 kWh/ton if slags are formed about 1 percent in melting of 3 tons iron.

2.6. Water Cooling System In coreless induction melting systems, water is vital to the success of a complete operating system. It needs the high quality water to maximize system reliability and component longevity for the cooling of power supplies and furnaces. In a coreless induction furnace much of the heat loss by the metal passes through the furnace lining. Heat is also generated in the power coil or induction coil itself by the passage of current. To prevent damage and overheat to the coil it must be water cooled. A sample of the cooling water passing through inside the thick-walled copper tubing is shown in Figure 2.8.

18

Figure 2.8. A Sample Induction Coil with Cooling Water

Figure 2.9. Sample of the Damaging Induction Coil

Flow velocity and monitoring of all water circuit should be considered for the cooling of induction coil. Bailey [10] recommended that all cooling-passages should be designed so that the flow velocity is not less than 1 meter per second, to prevent any suspended solids settling-out in the system. All complete water circuits should be designed so that the flow can be monitored, either by open-ended pipes or by instrument indication. Monitoring with instrument indication may be expensive, but accuracy is good and reliable for the whole system. Temperature should also be monitored at each outlet. Flow switches should be provided at each outlet to ‘trip out’ the furnace power supply in the event of a failure. The over flow-bucket types are preferred in an open system. If the cooling water cannot be sufficiently provided to the induction coil and the necessary components in some installations such as the frequency-conversion equipment, the power cables, the control panel and the capacitors, the coil may be damaged and exploded to the surrounding where the employees will be working

19 inside the foundry shop. Simultaneously, it will affect the productivity, the mental and physical power of workers and all works of industry. A sample of the damage of induction coil is shown in Figure 2.9. Sometimes, it may also be necessary in some installations to cool the water in the frequency-conversion equipment, the capacitors and power cables. In channel furnaces the coil and the inductor casing are usually water-cooled. The cooling water supply temperature should not be below 25ºC, to prevent condensation on the cooled components. The upper limit of water temperature leaving the coil should be no more than 70ºC, and that from the capacitors and frequency-conversion equipment should not exceed the value specified by the manufacturers. If too cold water is allowed to return to the system (cold temperature is defined as water temperature lower than the ambient air temperature), condensation will then form on the electrical parts and the coil. The life expectancy of these components is related to their operating temperature and maintenance. There are various types of cooling system to support the induction coil, frequency- conversion equipment, the capacitors and the control panel. They are installed and constructed in many foundry shops according to the requirements of installation space, the annual operating costs, the furnace sizes and capacities, and the environmental conditions, and the area of the industry. The types of water cooling system used in most of the application for coreless induction melting systems will be described in section 2.7.

2.6.1. Water Requirements The quality and quantity of water required to cool a coreless induction melting system should be specified in the equipment manufacturers’ literature or quotation. If a new coreless induction melting system is proposed to be installed in an existing facility with established plumbing in place, several design factors relating to water flow and pressure must be considered. Additional water supply must exist within the plant. Then, there is adequate flow and pressure to satisfy the equipment manufacturers’ specifications. The present water quality characteristics do meet the specifications of the induction furnace manufacturer. The addition of the new system will affect flow and pressure to the existing and new system may be required. If a new line is required, it should be designed to eliminate friction losses along with assuring that there is an adequate supply of emergency water.

20 2.6.2. Effects of Water Quality There are three detrimental effects of poor water quality in melting equipment cooling paths are: (a) The reduction in the ability to transfer heat that leads to subsequent damage to the components from overheating. (b) Electrochemical corrosion of tubing. (c) Degradation of the electrical performance of the melting equipment due to the water having too high an electrical conductivity. All of these effects are directly related to impurities in the water scale formation, fouling due to products of corrosion or fouling due to biological growth. When this fouling does occur, in order to maintain the same heat transfer, the temperature difference between the water and the component will increase. As the fouling continues to build up, the temperature increases and the components fails. This process is further aggravated by the reduction of water flow caused by the reduction in the cross sectional area of the path. Electrochemical corrosion is the deterioration of solids by liquid electrolytes. In this case, the electrolyte is the contaminated cooling water, which attacks metal components in the system. Under severe corrosion conditions the components can corrode or rust in less than a year time. High electrical conductivity is directly related to the amount of dissolved solids in the water. The resulting problems are the distortion of the electrical control signals to solid-state devices and the desensitizing of the ground detector circuits.

2.6.3. Water Purification/Maintenance The highly de-ionized water has very corrosive properties and it can cause damage to the induction coils. Corrosion of iron in the piping can add enough iron in suspension to affect conductivity. Therefore, the newer water systems will usually include a de-ionizer to main the conductivity of the water at acceptable levels. The deionizers are used to maintain a water conducting level of 50 micromhos/cm or lower. It is generally accepted that an operational water conductivity range of 100 to 300 micromhos/cm is adequate for operation for the water system. In a closed water system if the water is not changed periodically a microscopic organism will develop. This organism will attack the copper surfaces of the water system and if not addressed will eventually lead to water leaks throughout the system.

21 By removing a hose on the furnace coil and inspecting the inside diameter of the copper tubing it can be determined if there are microorganisms present. The inside of the copper tubing will show a shiny black surface and will be very slipping. Treatment for microscopic organisms can be done by draining the system of all water, then acid wash the entire system with water. Then refill the system, making sure to remove all of the entrapped air.

2.6.4. Filtration Many filtration units have been used with high maintenance requirements. The centrifugal separator, one of the filtration units, is used in water systems to remove solids from liquids. Many advantages of using these devices are as follows: 1. No moving parts to wear out 2. No screens, cartridges, cones or filter elements to replace 3. No backwashing 4. No routine maintenance or downtime requirements 5. No standby requirement needs 6. Low and steady pressure loss 7. Easily automated By removing the solids from the water, the life of the pumps can be extended, fouling of cooling towers and heat exchangers can be virtually eliminated and allow for optimum efficiencies.

2.6.5. Effects of Impurities It is important that there are the effects of impurities in circulating water system. Typical water impurities affect water quality. High water conductivity can result in distortion of control signals and it can lead to corrosion of pipe nipples. If the water is over saturated with calcium bicarbonate, calcium carbonate will form on the piping interior. This deposited scale will restrict water flow and decrease heat transfer. The suspended solids can also accumulate in equipment, particularly at low points, causing clogging and reducing heat transfer. Suspended solids in makeup and circulating water can be removed by either filtration or centrifugal separation. Water that contains a high amount of free mineral acid is required. Acidity is evidenced by effervescence when in contact with carbonate. This makes the water very corrosive. The measure of pH of a solution is a measure of acidity of the solution.

22 Acid solutions have a pH of less than 7. Other effects of impurities are alkalinity, slime and algae biological fouling, and dissolving oxygen and corrosion. If the alkalinity is determined to be in excess, treatment of water with acid may be necessary. Slime and algae biological fouling can offer and occur in once through and open circulating systems. It is formed by the excessive growth or accumulation of lower forms of plant life. Chemical treatment, usually chlorine, may be used for control of these growths to avoid loss in heat transfer and to minimize biological fouling on metal surfaces. Dissolving oxygen and corrosion is accelerated by dissolved gases such as oxygen, ammonia, carbon dioxide or sulfur dioxide, dissolved solids and high temperature. The gases mentioned cannot be removed by mechanical means because they tend to ionize in the water. The life of electrical conducting components in induction systems relies heavily on the quality of the water supplied by the water system. Nevertheless, the selection of a high quality cooling system for coreless induction melting systems is of prime importance.

2.6.6. Emergency Water Supply and Cooling System In all coreless induction furnace systems, a source for emergency water must be used to supply cooling water to the furnace during times when the water system loses power or has a pump failure. Many water systems are provided with a standby pump in case of primary pump failure; but in a case where there is a power outage and the recirculating pumps cannot be run, an emergency water system is the only alternate source for cooling water. This is due to the fact that both the molten metal in the furnace and the refractory system have significant amount of stored energy that must be removed through the recirculating water at all times. Energy transfer to unrecirculated water in the coil will cause the temperature of the water contained within it to rise. The temperature will continue to elevate until the water turns to steam where it will expand in volume. Since the water is closed, the pressure in the coil will increase until hoses blow off of the coil and all of the water contained within will be expelled. At this point there is nothing to remove the stored energy in the furnace and it will transfer to the coil and raise its temperature to that exceeding the ratings of materials in contact with it. This will result in a significant expense to the foundry as regards to equipment damage as well as loss of production due to loss of service of the equipment. In this

23 situation, if possible, there should be a procedure to empty the furnace immediately of molten metal, thereby eliminating the largest amount of the stored energy that needs to be removed. The emergency cooling system should be provided to cool the furnace coil in the event of power failure. The emergency water should be gravity-fed from a highlevel storage tank, supplied from the mains, and connected directly to the furnace coil via a check valve that should be opened automatically when the pressure in the normal, pumped supply falls. The emergency water will flow through the coil to the buffer tank, and then to the drain through an overflow pipe.

2.7. Types of Cooling Water System for Electric Induction Furnace Various types of cooling water system for electric induction furnace are as follows: 1. Cooling pond system 2. Spray pond system 3. Evaporative cooling tower-open circuit system 4. Fan-radiator closed-circuit system 5. Water/water heat-exchanger system 6. Dual system with closed-circuit cooling tower

2.7.1. Cooling Pond System Cooling pond system is one of the cooling systems of induction furnace melting. When large ground areas are available, cooling ponds offer a satisfactory method of removing heat from water. A pond may be constructed at a relatively small investment by pushing up on earth dike 1.8 to 3.1 m (6 to 10 ft) high. For a successful pond installation, the soil must be reasonably impervious, and location in a flat area is desirable. Typical sketch of cooling pond is shown in Figure 2.10. Hot water inlet

Cool water outlet Water surface

Pond

Figure 2.10. Typical Sketch of Cooling Pond System

24 In many cases, the pond water must be treated with chlorine, thus it is more economical to use an open loop for the treated water. Acceptable circulation rates vary from hour by hour for a complete change of water. They should be considered to resist the corrosive effects of the chlorine in the pond water and scaling or corrosion. Four principal heat-transfer processes are involved in obtaining cooling from an open pond. Heat is lost through evaporation, convection, and radiation and is gained through solar radiation. The required pond area depends on the number of degrees of cooling required and the net heat loss from each square foot of pond surface.

2.7.2. Spray Pond System The hot water from the induction coil needs to be cooled to the desirable temperature before pumping it. The cooling process is carried out in spray ponds after which the water is pumped back to the induction coils. In spray ponds, the exchange of heat between the hot water and ambient air is performed by conduction process between the fine droplets of water and the surrounding air. The efficiency of the system is mainly dependent on the relative humidity of the air. Due to loss of water from the pond, fresh water makes up system operating on pond level is required. Spray ponds provide an arrangement for lowering the temperature of water by evaporative cooling and, in so doing, greatly reduce the cooling area required in comparison with a cooling pond. A spray pond uses a number of nozzles which spray water into contact with the surrounding air. A well-designed spray nozzle should provide fine water drops but should not produce a mist which would be carried off as excessive drift loss. The pond should be placed with its long axis at right angles to the prevailing summer wind. A long, narrow pond is more effective than a square one, so that decreasing pond width and increasing pond length will improve performance. Performance can also be improved by decreasing the amount of water sprayed per unit of pond area, increasing the height and fineness of spray drops, and increasing nozzle height above the basin sides. A typical spray pond system with evaporative cooling, which is by far the most effective factor, is shown is Figure 2.11.

25

Figure 2.11. Sample Spray Pond System

2.7.3. Evaporative Cooling Tower-Open Circuit System An induction furnace requires a great quality of cooling water, so a recirculating system should be used to conserve water and save cost. In this system, water from the furnace coil and, if necessary, the other ancillaries cascades through the splash matrix of an evaporative cooling tower are cooled by a counter-current of air supplied by a fan. The water gravitates to a sump, from which it is pumped through the coil and other circuits before being returned to the tower via a buffer tank. Simplified schematic arrangement of this system is shown in Figure 2.12.

Figure 2.12. Open-Circuit System with Evaporative Cooling Tower

This type of system has advantages and disadvantages as follow: Advantages -

Simplicity.

-

Low capital cost.

26 -

Cooling water with the ambient wet-bulb temperature.

Disadvantages -

Water is lost by evaporation, so that solids dissolved in the system concentrate and cause electrical conductivity problems.

-

Airborne dust and impurities are drawn into the tower and cause corrosion and fouling problems.

-

If the make-up water is hard, scaling can result, reducing heat transfer and even causing total blockage.

-

Cooling towers are temperature and humidity dependent; in conditions of high temperature and high humidity their efficiency will be decreased.

2.7.4. Fan-Radiator Closed-Circuit System This system provides an essentially closed-circuit system which prevents entrainment of dust particles and other atmospheric pollutants. It consists of a heat exchanger in the form of a fan-blown radiator, a circulating pump, and a buffer tank to allow for expansion. Schematic diagram of fan-radiator (closed-circuit) system is shown in Figure 2.13.

Figure 2.13. Fan-Radiator Closed-Circuit System

Advantages and disadvantages in this system are as follow: Advantages -

Water circuit can be made completely enclosed.

-

Loss of water is slight, so expense for water is lower than in evaporative towers.

27 Disadvantages -

Radiators are large for a given thermal duty.

-

Radiator fins are subject to blockage by atmospheric dust, and may be difficult to clean.

-

Radiators are ambient temperature dependent and are less effective in warm ambient conditions.

2.7.5. Water/Water Heat Exchanger Dual System This system is shown in Figure 2.14. It consists of two circuits: primary open circuit and secondary closed-circuit. 1. Primary open circuit _ with cooling tower, circulating-pump and heat exchanger. 2. Secondary closed circuit _ with furnace coil and other circuits, buffer tank and circulating-pump.

Figure 2.14. Dual System with Water/Water Heat Exchanger

The primary system supplied cooled water at near ambient temperature to the heat exchanger, where heat is removed from the secondary circuit and returns to the cooling tower. The secondary circuit carries heat away from all furnace circuits to a buffer tank, from which the water is pumped back through the heat exchanger. Its advantages and disadvantages are as follows: Advantages -

The water/water heat exchanger is more compact and easier to clean and maintain than the fan-radiator system.

28 Disadvantages -

A primary source of cooling-water is required.

2.7.6. Dual System with Closed-Circuit Cooling Tower In this arrangement, the splash system of the normal evaporative cooler is replaced by a tube bundle, through which the furnace cooling-water is circulated. The primary water trickles over the bundle against the flow of air provided by a fan, and so it is cooled at the same time as heat is transferred from the secondary water to the primary water. Schematic arrangement of this system is shown in Figure 2.15. Its advantages and disadvantages are as follows: Advantages -

Water/water heat exchanger is eliminated.

-

Piping and pumping costs are lower than in conventional tower with heat exchanger.

Disadvantages -

Slightly more expensive than conventional tower with heat-exchanger.

Figure 2.15. Dual System with Closed-Circuit Cooling Tower

2.8. Selection of Cooling System It depends upon: 1. Furnace size 2. Furnace environment 3. Local water board regulations 4. Nature of water supply available

29 5. Local noise-control requirement, particularly at night 6. Cost To eliminate noise level in a furnace environment, cooling pond system gives a satisfactory solution. This system reduces the maintenance costs compared with other types of cooling system. Although it is suitable for small furnaces, the space available in foundry for pond surface area becomes the major factor for the larger furnaces. For small furnaces, it is often more economical to use a sample, open recirculating system with a cooling-tower. For larger furnaces, a fan-radiator system or dual system with a water/water heat exchanger is preferable. Fan radiators should not be used in a dusty environment, or where noise is likely to be nuisance, particularly at night. Noise can be reduced by installing fans at ground level, wherever possible, and by using foundry buildings to screen the noise. A closed-circuit coolingtower may be useful for larger furnace, where it could be smaller than the normal tower in a dual system.

30

CHAPTER 3 FLOW CALCULATION AND PUMP SELECTION As the flow velocity of induction coil (power coil) and the feasible pump of pumping the water sufficiently are the important factors, the considerations and calculations based on these factors are solved analytically by using the solution procedures. To obtain the prefect flow rates, pump selection should be carried out for the cooling system. The required flow rate and pump for 0.16 ton coreless induction furnace are focused in this chapter by using the equation of heat transfer and fluid mechanics.

3.1. Consideration of Flow Velocity To consider the flow velocity inside the induction coil, there are two portions: heat transfer due to the effect of heat generated by the alternating current and transferred through the refractory lining from molten metal and heat carrying from fluid flow due to the pumping device. Before considering the flow velocity of the induction coil, the internal structure of 0.16 ton coreless induction furnace is shown in Figure 3.1. Trunion Molten metal

Shell Pouring spout

Refractory cement Crucible Copper induction coils Rammed refractory

Tilting bail

Water cooling hoses Power leads Stand

Figure 3.1. Internal View of 0.16 ton Coreless Induction Furnace

31 Firstly, the temperature of molten metal in the crucible is approximately about 1,600ºC according to the melting points of various types of metal. This crucible is made up of silica lining, which is surrounded by an asbestos sheet, which is again surrounded by an asbestos cloth. Heat from molten metal passes through the silica lining, asbestos sheet and asbestos cloth, and then it conducts to induction coil. The temperature of coil will be maintained at about 78ºC because of the effect of cooling water and the high flow velocity. In accordance with the temperature of molten metal in the crucible, the flow velocity of induction coil is considered for the cooling system. It should be selected for the suitable pump corresponding to the designative flow velocity. Flow velocity may affect not only the service life of high conductivity copper coil but also overall system of furnace. It is also the main point among the most important design parameters. Nevertheless, the flow velocity for all cooling passages, especially the induction coil, should be designed more than 1 meter per second that had been met as described in the aforementioned chapter. 3.1.1. Specifications of Induction Coil The design of induction coil is typically manufactured with a copper tube wound with a carefully selected tubing profile and number of turns on the coil to match the melting process into the power supply used. It may be either flattened, round, or elongated vertically [11]. The round section allows the large water passages within the coil and assures maximum water circulation together with efficient cooling, but the flatted section permits a higher input per unit of coil height. The use of heavy copper tubing prevents coil distortion when the coil is positioned and clamped immovably inside the casing. The power for the coil is carried in flexible water cooled leads which can be connected either left hand or right hand side of the coil. One of the induction coils recommended by low power transmission resistance is produced from copper material for 0.16 ton induction furnace made in Russia. The specifications of induction coil concerning with the physical and electrical parameters are described in Table 3.1. The electrical parameters such as input power, rated voltage and frequency may be varied throughout the melting and pouring time. The maximum possible ratings for the specifications of induction coil are also described in Table 3.1.

32 Table 3.1. Specifications of Induction Coil Physical Parameters Material

Electrical Parameters

Copper

Input power

95 kW

2.0828 cm

Frequency

880 Hz

1.7018 cm

AC current

1,500 A

Outer surface area

3.4071 cm2

Rated voltage

650 V

Inner surface area

2

2.2741 cm

Water inlet temperature

28ºC (82.4ºF)

Number of turns

16

Water outlet temperature

Coil height

46.228 cm

Water pressure

2 to 4 MPa

Total length

20.96 m

Estimated melting time

1.56 hr

Coil outer diameter Coil inner diameter

54ºC (129.2ºF)

3.1.2. Effect of Electrical Resistance in Induction Coil The electrical resistance due to the heat generating rate is formed inside the induction coil itself while passing through the alternating current. It is a measure of the degree to which a body or tubing opposes the passage of an electric current. The electrical resistance of high conductivity copper tubing is similar to the hydraulic resistance of a pipe and it varies directly with the length and inversely with the crosssectional area. This relation proposed by Loew [12] can be expressed as follow: RDC =

ρ1l A1

Equation 3.1

where, l = length of conductor in direction of current, cm A1 = area of conductor normal to direction of current, cm2

ρ1 = resistivity, µΩcm The resistivity is also called specific resistance of conductor material which depends upon the chemical and physical properties and measured in micro ohmcentimeters and micro ohm-millimeters. Resistivity is always expressed as at the standard temperature 20ºC (68ºF). When the resistivity of copper tubing is known, the total resistance of its material may readily be computed from its dimension. The electrical resistance of a pure metal is directly varied with the temperature, as

33 illustrated in Figure 3.2 for the case of copper, and its resistance would be reduced to zero when the temperature reached -234.5ºC.

Resistance

R2 R1 Tc

0

-234.5

t1

t2

Temperature, ºC

Figure 3.2. Variation of Resistance with the Temperature Since the usual range of interest runs from perhaps 20ºC to a few hundred degrees above zero, a straight-line law of variation may be assumed for the usual condition. The resistance-temperature relationship is apparent from Figure 3.2 that the rule of similar triangles may be applied to find the resistance R2 of copper tubing at any temperature t2, if the resistance R1 at some other temperature t1, and the temperature intercept Tc of the conductor material of copper are known. From similar triangle,

R2 Tc + t 2 = R1 Tc + t1 R2 =

R1(Tc + t 2 ) Tc + t1

Equation 3.2

In another consideration of this relationship, if the slope of the straight-line portion of the curve in Figure 3.2 is designed as m, the equation from analytic geometry may be written as follows: R2 = R1 + m(t 2 − t1 )

where m =

R1 and therefore Tc + t1 R2 = R1 +

⎡ ⎤ R1 1 (t 2 − t1 ) = R1 ⎢1 + (t 2 − t1 )⎥ Tc + t1 ⎣ Tc + t1 ⎦

34 The fraction 1/ (Tc+t1) is usually considered as α1 and is called the temperature coefficient of resistance. Ultimately, the relationship of resistance variation in a copper metal with temperature is shown as follow: R2 = R1 [ 1 + α1(t 2 − t1 )]

Equation 3.3

Because the temperature Tc for copper is 234.5ºC the temperature coefficient of resistance can be described as

α1 =

1 234.5 + t1

Equation 3.4

On the other hand, the calculation formulas of the electrical resistance for various conductor materials can be seen in electrical handbooks. 3.1.3. Heat Generation Rate Calculation In the flow velocity consideration, the generation of heat in induction coil with respect to the electrical resistance is one of the important factors. Paschkis and Persson [13] studied the common feature in induction heating in which heat generation is always localized, whereas in dielectric heating the generation of heat may be uniform. The locality of temperature in the induction coil can be approached by heat generation rate according to the supplied power, rated voltage and the usage of frequency. Table 3.1 given by the specifications of induction coil for 0.16 ton melting capacity will be used for the calculation of heat generation rate. The area of copper tubing is π(Do2 − Di2 ) π( 2.0828 2 − 1.7018 2 ) = = 1.1325 cm 2 4 4

A1 =

The resistivity of copper at 20ºC from Marks [14], ρ1 = 1.7 µΩcm By using the Equation 3.1, the DC resistance inside the coil is computed as R DC1 =

ρ1l 1.72 × 2096 = = 0.003183 Ω A1 1.1325

From the Equation 3.4, the temperature coefficient of resistance is calculated at t1 = 20ºC, 1 234.5 + 20 = 0.00393

α1 =

35 Another electrical resistance with the Equation 3.3 is calculated by using the induction coil temperature, t2 = 60ºC which is maintained by the effect of cooling water. This temperature measures from the operating condition of induction furnace. R DC2 = R DC1 [ 1 + α1(t 2 − t1 )] = 0.003183[ 1 + 0.00393( 60 − 20 )] = 0.00368

AC resistance of induction coil RAC is taken from R AC /R DC ratio that is read from the curve for skin effect and proximity effect. The skin effect is the phenomenon where the apparent resistance of copper tubing increases as the frequency increases. For round copper tubing, the Figure B.1 of the Appendix taken from Dwight [15] plotting the root of the ratio, frequency: DC resistance (ohm per 1,000 ft), versus RAC/RDC can be easily given to determine the AC resistance RAC. To compute AC resistance of copper tubing, the ratio of coil thickness divided by outer diameter tc/d must be known: t c 0.1905 = = 0.09146 d 2.0828

Based on DC resistance (ohm per 1,000 ft), RDC2 Ω

0.00368 Ω × 1000 ft × 0.3048 m 1000 ft 20.96 m = 0.0535 Ω =

f ⎛ RDC2 ⎞ ⎜ ⎟ ⎜ 1000 ⎟ ⎝ ⎠

=

880 0.0535

= 128

From Figure B.1 of the Appendix, R AC = 1.04 RDC R Ac = 1.04 × (RDC2 ) = 1.04(0.00368) ∴ R AC = 0.0038272 Ω Finally, power losses or heat generation rate of induction coil due to the effect of electrical resistance can be expressed as follows: Pg = I 2 R AC = ( 1500 2 ) × 0.0038272 = 8.61123 kW

36 3.1.4. Calculation of Heat Transfer Rate in Composite Refractory Shells In this calculation, a system with cylindrical symmetry and having heat conduction only in the radial direction is considered as this situation occurs in refractory shell in electric furnace. From the point of view of steady state heat conduction, the heat transfer rate through the refractory lining from the molten metal can be solved by using the prescribed physical conditions of induction coil. The schematic diagram of temperature distribution for a composite refractory cylindrical shell is shown in Figure 3.3.

ln( r2 / r1 ) ln( r3 / r2 ) ln( r4 / r3 ) 2πk A L 2πk B L 2πk C L

Figure 3.3. Temperature Distribution for a Composite Refractory Cylindrical Shell Thermal insulations in refractory shell of furnace comprises low thermal conductivity materials such as silica lining, asbestos sheet and asbestos cloth combined to achieve an even lower system thermal conductivity. The radii at various interfaces should be defined as - r1, r2, r3, the temperatures at these interfaces - Ts,1, Ts,2, Ts,3, Ts,4 and the conductivities of these materials - kA, kB, kC, respectively. The heat transfer rate of the surface of induction coil is considered at maximum melting point. In solving these heat transfer rates, all pertinent simplifying assumptions are carefully listed as follows:

37 1. Steady state conduction. 2. One-dimensional heat transfer by conduction across the cylindrical walls. 3. Using pure metals with the maximum point (1,600ºC). 4. Supplying the induction coil with the electrical current 1,500 A, AC voltage 650 V, power 95 kW and frequency 880 Hz. 5. Constant properties. 6. Neglecting the interfacial contact resistance. 7. Neglecting the radiation heat transfer. As the heat flow through the composite refractory shells is considered to be under steady state, whatever heat enters into a layer at one end must also leave it at the other. In the calculation of heat transfer, the values for the thermal conductivity of asbestos sheet, asbestos cloth and silica lining have been read from Phelps [16], Marks [14], and Kern [17]. From Figure 3.3 together with the thermal equivalent circuit, the temperature at one side of induction coil is available at 74ºC measuring the practical result and the other side is potentially maintained 60ºC from the practical measurement owing to the cooling water. Fourier's law expressed in Incropera and DeWitt [18] is used for the heat transfer rate as follow:

qr =

Ts,1 − Ts,2 Ts,2 − Ts,3 Ts,3 − Ts,4 = = ln (r2 /r1 ) ln (r3 /r2 ) ln (r4 /r3 ) 2πk A L1 2πk B L1 2πk C L1

Equation 3.5

By substituting the desired values: three different types of thermal conductivity and length and various radii in Equation 3.5, the heat transfer rate will be given by

qr =

1600 − 74 ln(0.193336/0.12) ln(0.194836/0.193336) ln(0.1973336/0.194836) + + 2π (7.2663)(0.46228) 2π (5.5189)(0.46228) 2π (1.5116)(0.46228)

According to the calculation, q r = 58.7354 kW

Finally, the heat transfer rate passing through the refractory lining and insulation materials has been obtained as the amount of heat flowing. Using the result of heat transfer rate, the flow velocity inside the induction coil will be determined in the next section.

38 3.1.5. Flow Velocity Designation In the previous subsections, the heat transfer rates have been computed by applying the effect of electrical resistance in induction coil and steady state heat conduction equation for composite refractory cylindrical shell. Now, the total heat transfer rate will be considered for the designation of flow velocity circulated in induction coil or power coil. The total heat transfer rate can be expressed as follows: Total Heat Transfer Rate = Heat Generation Rate + Heat Transfer Rate and substituting the required value from the previous subsections, it has been obtained as follows: Qt = (8.6112 + 58.7357) kW = 67.3469 kW

In case the analytical form of the total heat transfer rate, Qt is for 0.16 ton induction furnace system, heat flow and generating heat are not uniform throughout the operating condition. Both of these are dependent upon the frequency, supplied power and current, rated voltage given by operator, and thermal conductivity of insulation materials changing from the temperature. Assuming the maximum possible condition in this furnace, it will be continued to solve the flow velocity. According to the practical measuring result, the inlet temperature or cool water temperature into the tubing is about 28ºC and the outlet temperature or hot water temperature from the tubing is about 54ºC at the highest melting conduction in induction furnace. In order to take into account of the flow velocity of induction coil, Newton's law of cooling will be used for the heat exchange. Accordingly, Newton's law of cooling for incompressible fluid presented in Incropera and Dewitt [18] may be expressed as

Qt = m&C p (Tm,o − Tm,i )

Equation 3.6

where, Qt = total heat transfer rate or quantity of heat or heat exchange rate, kW m& = the mass flow rate, kg/s

Cp = specific heat at constant pressure, 4.179 kJ/kgºK Tm,o = the leaving temperature of water from the coil, ºC and Tm,i = the entering temperature of water into the coil, ºC Hence, using Equation 3.6 and putting in the desired values, the mass flow rate is

39

m&=

Qt 67.3469 kW = C p (Tm,o − Tm,i ) 4197 J/kg οK (54 − 28) ο C

= 0.6172 kg/s From Mott [19], the mass flow rate, m& is related to the volume flow rate, Qv by

m&= ρQv

Equation 3.7

where ρ = the density of the fluid, 986.4 kg/m3 at 54ºC

The volume flow rate, Qv can be driven as follows:

Qv = Ai vi

Equation 3.8

where, Ai = the inner area of the tubing, m2

vi = the average velocity of flow, m/s The inner area of the induction coil is calculated as:

Ai =

πDi2 π × 0.017018 2 = = 0.00022746 m 2 4 4

Using Equation 3.7 and Equation 3.8, the average velocity of flow can be solved as follows: vi =

m& 0.6172 kg/s = ρA i 986.4 kg/m 3 × 0.00022746 m 2

= 2.7507 m/s

Ultimately, the designation of flow velocity inside the induction coil can be defined satisfactorily as 2.7507 m/s.

3.2. Pump Selection

When selecting a pump for the application of cooling system in induction furnace, the possible factors must be considered to be economical and successful throughout the melting cycle. These factors involve the nature of liquid to be pumped, the required capacity to be sufficient, the total head on the pump and others. Especially, there are two pumps used in induction furnaces. For the induction coil to be circulated with cooled-water, one pump is operated continuously during the melting process. If the emergency cases such as power failure and unexpected error occur in the running condition, another pump or stand-by pump will be used not to be lack the cooled-water for the furnace ancillary and equipment. Hence the pumps are one of the equipments essential in cooling system.

40 During the pump selection process, the alternative pumping station layouts should be developed in sufficient detail so that the cost of pumps over the life of the project can be determined. The cost of pumps should include the capital cost and the operating costs which include cost of energy, maintenance, and replacement costs. It is usually best to consider all types of pumps when developing the pumping station layout unless it is obvious that certain ones are not applicable. The pump selection corresponding to the cooling system with the pipe line arrangements should be done in sufficient detail to follow the process without reference to additional catalogs or other such sources. 3.2.1. Essential Parameters Required in Selection The essential parameters required in selection of pumps are summarized as follows: 1. Number of units required. 2. The nature of the liquid to be pumped. 3. The required capacity (volume flow rate) as well as the minimum and maximum amount of liquid. 4. The conditions on the suction (inlet) side of the pump. 5. The conditions on the discharge (outlet) side of the pump. 6. The total head on the pump. 7. The continuous or intermittent service. 8. The type of system to which the pump is delivering the fluid. 9. The type of power source (electric motor, diesel engine and stream turbine) 10. Space, weight, and transportation limitations. 11. Location of installation. 12. Environmental conditions. 13. Special requirements of marked preferences with respect to the design, construction, or performance of the pump. 14. Cost of pump operation and pump purchase. 3.2.2. Selection Procedures There are various types of selection procedure to use a pump in pumping system. Chopey [20] gave a step-by-step procedure for choosing the class, type, capacity, drive and materials for a pump that will be used in an industrial pumping

41 system. Later, this procedure will be used for the pump selection in cooling system of induction furnace. Solution procedures for any pumping system are expressed as follows:

1. Sketch the proposed piping layout In the first procedure, both single-path and multiple-path diagram of the piping system should be sketched on the actual job conditions. Showing all the piping, fittings, valves, equipment, and other units in the system and marking the actual and equivalent pipe length on the sketch are involved in this procedure.

2. Determine the required capacity of the pump The required capacity is the flow rate that must be handled in gal/min, m3/min, or some similar measure. It has been obtained from the process conditions such as boiler feed rate, cooling water flow rate and chemical feed rate. The required flow rate for any process unit is usually given by the manufacturer. Once the required flow rate is determined, a suitable factor of safety is applied. Typical safety factors are in the 10 percent range.

3. Compute the total head on the pump The most common way of expressing the total head on a pump is the result in meter or feet of water. To compute the total head on the pump, the total static head; in meter or feet, must be considered from the pump piping arrangements with static suction lift and static discharge head. When both the suction and discharge surfaces are open to the atmosphere, the total static head equals the vertical difference in elevation. When the supply source is below the pump centerline, the vertical distance is called the static suction lift and otherwise, above the pump centerline, is called static suction head. The total static head, as described above, refers to the head on the pump without liquid flow. The friction losses in the piping system during liquid flow must also be considered to determine the total head on the pump. Thus, the actual total head on the pump is the sum of total static head and total friction-head loss.

4. Analyze the liquid condition The liquid conditions on a pump-selection should include the name and chemical formula of the liquid, maximum and minimum pumping temperature,

42 corresponding vapor pressure at these temperatures, specific gravity, viscosity at the pumping temperature, pH, flash point, ignition temperature, unusual characteristics (such as tendency to foam, curd, crystallize, become gelatinous or tacky), solids content, type of solids and their size, and variation in the chemical analysis of the liquid. Such these data are available from many pump manufacturers or can be prepared to meet special job conditions.

5. Select the class and type of pump Three classes of pumps: centrifugal, rotary and reciprocation are generally used today. Each class of pump is further subdivided into a number of types as shown in Table 3.2. Table 3.2. Pump Classes and Types Class

Type • Volute

Centrifugal

Rotary

Reciprocating

(or) Radial-flow Diffuser or turbine pump • Regenerative - turbine • Vertical - turbine • Mixed - flow • Axial - flow (propeller) • Gear • Vane • Cam and piston • Screw • Lobe • Shuttle - block • Direct - acting • Power (including crank and flywheel) • Diaphragm • Rotary - piston •

Note that these terms apply only to the mechanics of moving the liquid – not to the service for which the pump was designed. A general guide to the characteristics of various classes and types of pumps to be used in industrial process is shown in Table C.5 of the Appendix. This table describes that a centrifugal pump would probably be best when a large capacity at moderate pressure is required. It is also needed to be considered all the operating factors related to the particular pump. These factors include the type of service (continuous or intermittent), operating-speed preferences, future load expected and its effect on pump head and capacity, maintenance facilities available, possibility of parallel or series hookup, and other conditions peculiar to a given job. Once the class and type of pump are selected, pump selector should be consulted and adjusted with a rating chart, as illustrated in Figure B.2. of the

43 Appendix, or a rating table to determine if a suitable pump is available from the manufacturer whose unit will be used. When the hydraulic requirements fall between two standard pump models, it is usual practice to choose the next larger size of pump, unless there is some reason why an exact head and capacity are required for the unit. Some pumps are constructed for custom-built of a given job when precise head and capacity requirements must be met. In the engineering information of manufacturer; the characteristics curves for various diameter impellers in the same casing, variablespeed head-capacity curves for an impeller of given diameter, and other supporting data are included to be satisfied for the customers. More completed selection procedures can be found in Church [21], Peerless [22], Dickinson [23], Walker [24], Stepanoff [25], and Hicks and Edwards [26].

6. Evaluate the pump chosen for the installation In evaluating the pump chosen for the installation, the specific speed of a centrifugal pump is important to classify the impellers on the basis of their performance. It should be checked by using one of the suitable methods. Once the specific speed is known, the impeller type and approximate operating efficiency can be found from Figure 3.4

Figure 3.4. Approximate Relative Impeller Shapes and Efficiency Variations for Various Specific Speeds of Centrifugal Pumps (Worthington Corp.) Source: Daily (1950)

44 .

Then, it must be considered to see if the available net positive suction head

(NPSHA) is equal, or greater than, the required net positive suction head (NPSHR) of the pump. In addition to the checking of net positive suction head (NPSH), horizontal or vertical design of pump may be determined. From the stand point of floor space occupied, required NPSH, priming, and flexibility in changing the pump use, vertical pumps may be preferable to horizontal design in some installations. But where headroom, corrosion, abrasion, and ease of maintenance are important factors, horizontal pumps may be preferable. Lastly, before making a final purchase decision, all above presented steps must be checked and determined to be successful for the pumping system. In the next section, the necessitated pump for cooling pond system used in 0.16 ton coreless induction furnace will be chosen correctly according to the selection procedures. 3.2.3. Calculations for Pump Selection In this section, the candidate pump in cooling pond system is calculated by using the selection procedures for the given application. Calculation mainly involves the determination of pumping capacity, total head, specific speed and net positive suction head. Many manufacturers and system engineers currently use computerized procedures to select a pump that is most suitable for each given application. Such procedures are simply automated versions of the traditional selection method. Nevertheless, the calculations for the selected pump must be carried out accurately and correctly to meet the satisfactory system.

(i) Sketching the Piping Layout Before selecting the pump, the piping layout diagram is sketched on the actual job condition. The functional layout diagram of 0.16 ton coreless induction furnace has been shown in Figure 3.5. The different diameters of pipe are used in the piping system. The representation of bend or curve on the piping arrangements is illustrated as fittings such as elbow, valve, sudden enlargement and sudden contraction.

(ii) Determining the Pumping Capacity Based on the designation of flow velocity inside the induction coil, the required flow rate for pumping is considered for 0.16 ton coreless induction furnace. 0.16 ton coreless induction furnace in Figure 3.5 consists of two induction furnaces.

45

Furnace No.2 Furnace No.1

Control Panel

Capacitor Bank

Pump Discharge Pipeline

Suction Pipeline Cooling Pond

Figure 3.5. Functional Layout Diagram of 0.16 ton Coreless Induction Furnace

B

Q2 QT

A

Q4 Induction Coil

Q5 Q6

Water Cooled Cable

C

Q7

D

Capacitor Bank

Q1 Q8

Induction Coil

Q3 Control Panel

E

Q9 Q10

Water Cooled Cable

Q11

E

Figure 3.6. Sketch of Flow Branches in Pipes

46 According to the calculation of flow velocity designation in previous section, the velocity 2.75 m/s is essentially needed for both furnace 1 and furnace 2 because the high velocity rate causes the effective cooling. Omitting the frequency-conversion equipment, induction furnaces, cooling pond, emergency pump and all return pipes, the connections of pipe lines are schematically shown in Figure 3.6. To determine the required pumping capacity, the pipe joint A, B, C, D, E and F are defined as node points. First of all, the pipe from the outlet of pump holding the total flow rate, QT is in contact with the pipe joint A and it separates into two branches, Q1 and Q2. Similarly another pipe joints are also to go out in different directions as shown in Figure 3.6. Cooling system of coreless induction furnace has been constructed by using the multiple-path system in the pipe line arrangement. As the multiple-path system with pipe branches, the pipe flow regime is determined by the starting point of joint E. Joint E is defined as the junction of the flow rate Q9 and Q10. Pipe network for joint A is shown schematically in the following Figure 3.7.

(a) Representation of Pipe System with Three Branches

(b) Illustration of Flow Direction

Figure 3.7. Pipe Network for Joint E The general energy equation to each branch recommended by Fox and McDonald [28] is applied. This equation for the representation of the mechanical energy per unit mass at a cross section expresses as follows: 0

0

0

0

p v2 v2 pE + α E E + gz E = 9 + α 9 9 + gz 9 + hL ρ ρ 2 2

47 The following assumptions are used in the above equation. 1. v E2 << v92 , v102 2. z E ≅ z1 3.

p9 = p10 = p atm

4. Neglect loss due to flow split at

E

In the above equation, the kinetic energy coefficient, α varies with Reynolds number. For pipe flow calculations, α is often assumed unity because it is reasonably close to one for large Reynolds number. Hence p E v92 = + hL ρ 2

Equation 3.9

To know the corresponding flow rate for branch 8 and 9, it must be used one of the assuming velocities. For the case of Joint E, the flow velocity of water passing through the induction coil is known as 2.75 m/s in branch 9. It would be rounded of to 2.8 m/s throughout the calculation. For branch 9, kinematic viscosity, υ = 5.04 × 10 −5 m2/min at 28ºC Viscosity may be changed with temperature and is given by Mott [19] according to the different temperatures. Using the interpolation method, the kinematic viscosity is driven at 28ºC because of the cooled water temperature from the outlet of cooling pond. Determine the pipe Reynolds number. Re9 =

v9 D9 = 4.2 × 10 4 υ

To find the friction factor, f 9 , pipe wall roughness, ∈9 is read from Table C.1 of the Appendix. For leather tubing pipe, ∈9 = 0.0000015 mm ∴

∈9 = 0.0001 D9

From Moody's diagram given in Figure B.3 of the Appendix, f 9 = 0.022 The major is defined as hm and is solved by using the following equation given in Mott [19]. L9 v92 2.6975 168 2 × = 0.022 × × hm = f 9 × D9 2 0.0127 2

48 If the valves and fittings are included in pipe branches, the minor loss, hn must be determined with the equivalent length of pipe diameter, Le/D by using the Table C.2 of the Appendix. There are no minor losses due to valves and fittings in branch 9.

∴ hn = 0 Total head loss, hL = hm + hn Substituting the calculating values into Equation 3.9 yields pE

ρ

= 80055

Equation 3.10

For branch 2, Similarly,

Re10 = 503.96v10 ∈10 = 0.006 D10

hn = 0 hm = f 10 × and

0.1524 v102 × 0.0254 2

p E v102 ⎛ 0.1524 ⎞ = ⎜1 + f10 × ⎟ 2 ⎝ 0.0254 ⎠ ρ

Equation 3.11

From the continuity equation, Q8 = Q9 + Q10 A8 v8 = A9 v9 + A10 v10 0.0005v8 = 0.0212 + 0.0005v10 ∴ v8 = 42.4 + v10

or v10 = v8 − 42.4

Using the above velocity relation, one of the flow velocities is assumed to be substituted in Equation 3.11 and to know the fraction factor. When the calculated value of p E / ρ from Equation 3.11 is approximately balanced to the value of Equation 3.10, the flow velocities can be determined for branch 8 and branch 10. According to the calculating result, the velocity v8 (6.8 m/s) and v10 (6.1 m/s) are satisfactorily obtained for pipe branch 8 and 10 with respect to the flow rate Q8 (0.2064 m3/min) and Q10 (0.1852 m3/min). By using the same solution procedure, the flow rates in each branch could be determined for each pipe joint. As a result, the flow rate Q4 is obtained as the value 0.2742 m3/min, and the volume flow rate of Q1 and Q3 entered into the capacitor bank and control panel are given the range of 0.1 to 0.2 m3/min and 0.1 to 0.25 m3/min

49 respectively by the equipment manufacturer. So, the average flow rate of Q1 and Q3 become 0.15 m3/min and 0.175 m3/min. From the continuity equation, Q2 = Q3 + Q4 QT = Q1 + Q2 Finally,

∴ QT = 0.5992 m 3 /min

Therefore, the required flow rate for 0.16 ton coreless induction furnace is sufficiently given by 0.5992 m3/min as the pumping capacity. A pump of 0.69 m3/min (182 gal/min), which is larger than the calculated flow rate, capacity would be selected to be safe for providing flow circulated into the cooling pond. Generally, many manufacturers may be given for all their equipments. (iii) Computing the Total Head To determine the total head on the pump, the unnecessary pipes from piping arrangement must be omitted on the sketch of layout as shown in Figure 3.8.

Ø 0.0127 m (Copper tubing) L 32.8011 m Elbow – 4 No: Valve – 1 No:

6 Ø 0.0254 m (GI) L 2.4384 m Sudden contraction

5 4

Ø 0.0635 m (GI) L 11.5782 m Elbow – 8 No:

3 Ø 0.0508 m (GI) L 2.1334 m Elbow – 4 No: Valve – 1 No: Sudden enlargement

2

Ø 0.0635 m (GI) L 12.1914 m Elbow – 7 No:

Ø 0.0508 m (PVC) L 3.6574 m 1 Elbow – 4 No: Valve – 2 No: Sudden enlargement

Figure 3.8. Sketch of Suction and Discharge Line in Pumping System In the above figure, the bend or curve is represented by the elbow due to the small scale-size. Pipe sections are also defined as the number 1 and 2 for suction

50 pumping line, and 3, 4, 5 and 6 for discharge line according to its diameter, length, usage of valves and fittings, and type of pipe. Gate valve for pipe section 3 and 6, ball valve for pipe section 1, and 90º long radius elbow for all sections are used in suction and discharge line of pumping system. Using the volume flow rate of each pipe line, Reynolds number has been computed first. From Moody's diagram, friction factor is determined according to their relative roughness. All major and minor loss calculating procedures are similar to the determination of pumping capacity. The total losses resulting from the calculation are summarized in Table 3.3. Table 3.3. Total Losses for Pipe Sections Pipe

Flow Rate

Section

(m3/min)

1

0.69

2

Friction Factor

Major Loss Minor Loss

Total Loss

(m)

(m)

(m)

0.0143

1.6892

3.9199

5.6092

0.69

0.0245

3.1613

2.3052

5.4665

3

0.69

0.0256

1.7640

3.7830

5.5471

4

0.622

0.024

2.3899

4.4870

6.8769

5

0.205

0.032

7.1188

0.7878

7.9067

6

0.015

0.022

11.2795

0.3624

11.6420

To calculate the total loss that is considered for each pump selection, the computerized procedure with MATLAB program given in Appendix A could be applied. As the continuity, the volume flow rate of pipe section 1, 2 and 3 are equal to the required flow rate, 0.69 m3/min given by the determination of pumping capacity. Since the water passes through a sudden enlargement or sudden contraction, the equivalent length is also determined from Mott [19], Streeter [29], and Fox and McDonald [28]. The kinematic viscosity is based on cooling water temperature (28ºC) for pipe section 1, 2, 3, 4 and 5, and hot water temperature (54ºC) for pipe section 6. Now, total friction-head loss, Hf or the total loss of pipe and fittings for suction and discharge line computed above is 43.0485 m (141 ft) which is the sum of each of total losses from Table 3.3. Another consideration of total head is the total static head, Hts that is the sum of static suction lift, hsl and static discharge head, hsd. Because the vertical difference between the pump center line and the water surface level in the pond is approximately about 0.3047 m (1 ft), the total static head becomes

51 as Hts = hsl + hsd = 0.3047 + 1.5239 = 1.8286 m (6 ft). Note that the static discharge head is computed between the pump centerline and the pipe outlet with a free discharge. Then, the actual total head on the pump Ht = Hts + Hf. For the selected pump, Ht = 1.8286 + 43.0485 = 44.8772 m (147 ft). Thus an appropriate pump pumping total head 44.8772 m (147 ft) or above these meters must be chosen correctly for cooling system of 0.16 ton coreless induction furnace. (iv) Analyzing the Liquid Condition Water which is pumped to all parts of the equipment is an essential part. Especially, the water passing through the induction coil is very important to purify throughout the melting process. Involving solid contents, foreign matters and other suspended materials in water must be analyzed with the chemical technologies. (v) Selecting the Class and Type of Pump For the cooling system of induction furnace, centrifugal pump is selected because there is a strong dependency between capacity and the pressure that must be developed by the pump. Other characteristics of three classes of pumps are shown in Table C.5 of the Appendix. Among types of centrifugal class of pumps, centrifugal volute or radial flow type should be chosen as an appropriate pump for cooling system. In actual practice, it has been found that in volute type of casing there is only a slight increase in the efficiency of pump, because considerable loss of energy takes place in eddies developed in the casing. Volute type or radial flow based on the geometry of the flow path is significantly used where a small to medium quantity of liquid is to be discharged to high head. Moreover, centrifugal volutes are more economical compared with diffuser or turbine pump because there are no needs as diffuser ring and guide vanes. This reduces the frictional losses in the pump. When the required total head, pumping capacity, the class and type of pump are known, a rating table or rating chart as shown in Figure B.2 of the Appendix should be consulted to determine the exact horsepower requirement for that point of operation. For the capacity, 0.69 m3/min (182 gal/min) and the actual total head, 44.8772 m (174 ft), the specific Head-capacity point falls above 10 hp (7.5 kW) contour lines, the higher horsepower (11 kW) is selected in practice. In practice, more efficiently and safely, the followings are applied for more efficient and safe cooling system of 0.16 ton coreless induction furnace.

52 1. Total Head

46 m (160 ft)

2. Pumping Capacity

0.69 m3/min (182 gal/min)

3. Class and Type

Centrifugal Volute (Radial-flow Type)

(vi) Evaluating the Pump chosen for the Installation As discussed in selection procedure, the specific speed of a centrifugal pump should be checked by using the flow and head coefficients. Before determining the specific speed, it is assumed that the motor driving this pump might operate at 1160, 1750, 2900, or 3500 rpm. Note that alternating-current motors can operate at a variety of speeds, depending on the number of poles. Impeller proportions run at the specific speed. It is an important parameter for the installation of pump. It may be expressed by a relation given as Church [21]. Ns =

NQ 0 .5 H 0 .75

Equation 3.12

where, N s = pump specific speed N = pump rotative speed, rpm

Q = flow at optimum efficiency, m3/min

H = total head at optimum efficiency, m Using Equation 3.12, the specific speed N s = N( 182 )0 .5 / 160 0 .75 = 0.2999 N is computed. For each of the assumed speeds, the calculating results are tabulated in Table 3.4 as follows: Table 3.4. Operating Speed versus Required Specific Speed Operating speed (rpm)

Required specific speed (rpm)

1,160

346

1,750

522

2,900

865

3,500

1,045

The specific speed at each of the various operating speeds is analyzed using the data in Table 3.5. This table shows that volute and diffuser types of pump are pointed out the specific speed range below 2,000. In the practical installation, the specific speed (2,900 rpm) and radial-flow type (centrifugal volute) are selected as the operating speed at which a pump produces unit head at unit volume flow rate. Fox and

53 McDonald [28] recommended that low specific speeds are produced efficiently by radial-flow type and high specific speeds are produced efficiently by axial-flow type. Table 3.5. Pump Types Listed by Specific Speed Specific Speed Range

Type of Pump

Below 2,000

Volute, diffuser

2,000 – 5,000

Turbine

4,000 – 10,000

Mixed-flow

9,000 – 15,000

Axial-flow

Source: Peerless Pump Division, FMC Corporation (1979) When the specific speed is known, the impeller type and approximate operating efficiency can be determined from Figure 3.4. From the general relation between impeller shape, specific speed, pump capacity, efficiency and characteristics curve, the approximate operating efficiency (65 %) has been obtained at the specific speed (865) and pumping capacity (182 gal/min or 0.69 m3/min). Radial-type impeller in which the head is developed largely by the action of centrifugal force is selected and is used for medium and high heads (above about 150 ft) in cooling system of induction furnace. Under the Ns = 1000 ordinate, a cross section of impeller, as shown in Figure 3.4, appear directly and has a relatively moderate discharge area. Now, the net positive suction head (NPSH) is checked for the given piping system, to see if the available net positive suction head is equal, or is greater than, the required net positive suction head of the pump. Generally, net positive suction head required (NPSHR) is important in designing a specific pump. In simple terms it is the pressure, measured at the centerline of the pump suction, necessary for the pump to function satisfactorily at a given flow. The flow and temperature vary with NPSHR, but altitude is not affected. It may be measured in a pump test facility by controlling the input pressure.

NPSHR is usually determined by the pump manufacturer for the satisfactory operation and by running cavitations test on the pump. Typically, NPSHR can be calculated as follows: 4

⎡ NQ 0 .5 ⎤ 3 NPSHR = ⎢ ⎥ ⎣ 1200 ⎦

Equation 3.13

54 4

⎡ 2900( 0.69 )0.5 ⎤ 3 Using the above equation, NPSHR = ⎢ ⎥ = 2.5323 m (8.3085 ft) 1200 ⎣ ⎦ is obtained for the selected pump. In addition to NPSHR, the available net positive suction head is another parameter which is a characteristic of the system in which the pump operates. It depends upon the elevation or pressure of the suction supply, friction in the suction line, altitude of the installation, and the vapor pressure of the liquid being pumped. Both available and required NPSH vary with the capacity of a given pump and suction system. NPSHA is decreased as the capacity is increased due to the increased friction losses in the suction piping. The equation of NPAHA given in Evans [30] is as follows:

NPSHA = H a ± H s − H vp

Equation 3.14

where, H a = atmospheric pressure in meter

H s = total suction head or lift in meter H vp = vapor pressure in meter Atmospheric pressure as measured at sea level, is 14.7 PSIA because it is a measurement using absolute zero (a perfect vacuum) as base. The height of head is measured in foot. Head =

psi × 2.31 Specific Gravity

For water it is:

Head =

14.7 × 2.31 = 34 ft = 10.36 m 1.0

Thus 34 ft is the theoretical maximum suction lift for a pump pumping cold water at sea level. The relationship of altitude and atmospheric pressure is shown in Table 3.6. After knowing the atmospheric pressure at sea level, the relationship between vapor pressure and temperature for cold water is considered from the Table C.3 in the Appendix. From this table, vapor pressure can be given 1.28 ft at a temperature of 28ºC (82.4ºF). It follows then, that total suction head or lift (elevation difference) from the level of fluid in the cooling pond to the centerline of pump inlet is measured as negative sign because the pump is above the water surface of pond and it could be found illustratively in Figure 3.5. Hence it is ready to specify NPSHA. The available

55 NPSH, NPSHA = 34.1-1.28 = 31.72 ft (9.6678 m), is determined at a function of the

station layout and suction water levels. Often a two-foot safety margin is subtracted from NPSHA to cover unforeseen circumstances. When selecting a pump for the conditions above, the NPSHR as shown on the characteristics curve of pump or supply data from pump manufacturer should be 29.72 ft (31.72 – 2) or less to be operated satisfactorily throughout the running condition of cooling system. Table 3.6. Atmospheric Pressures at Various Altitudes Altitude in feet

Atmospheric pressure in feet of water

Deduction in feet of water

0

34.0

0

500

33.3

0.7

1000

32.8

1.2

1500

32.2

1.8

2000

31.6

2.4

2500

31.0

3.0

3000

30.5

3.5

3500

29.8

4.2

4000

29.4

4.6

4500

28.7

5.3

5000

28.2

5.8

5500

27.8

6.2

6000

27.3

6.7

6500

26.6

7.4

7000

26.2

7.8

7500

25.7

8.3

8000

25.2

8.8

8500

24.8

9.2

9000

24.3

9.7

10000

23.4

10.6

Source: http//www.pacificliquid.com/pumpintro.pdf Unless NPSHA at least equals NPSHR at any condition of operation, the fluid as water will vaporize in the pump inlet and bubbles of vapor will be carried into the impeller. These bubbles will collapse violently at some point downstream of the pump inlet (usually at some point within the impeller) and produce very sharp, crackling noises, frequently accompanied by physical damage of adjacent metal surfaces. These factors should be considered in the selection of pump. Horizontal type pumps are more desirable for the applications where the quantity of water to be pumped is large.

56 In the cooling system of 0.16 ton coreless induction furnace, the horizontal type pump for large quantity of water is used by providing a separate primary system. According to the above procedures, the selection used in cooling system of 0.16 ton coreless induction furnace has been satisfactorily solved with step-by-step solution. Moreover, one should consult with experienced system engineers to verify that the system operating condition has been predicted accurately and the required pump has been selected correctly.

57

CHAPTER 4 COOLING POND DESIGN

The design model consideration and design calculation of steady state cooling pond system required in 0.16 ton coreless induction furnace is the focus matter of this chapter. A few details of the conceptual study of cooling pond based on the solar heat flux and the equilibrium temperature are covered. The amount of solar heat flux relies on the meteorological conditions which are varied throughout the day. Hence, while dealing with the above topics, emphasis will be on cooling pond. The commonly used pond types are classified according to the degree of pond stratification (normalized temperature gradient) and the pond number.

4.1. Pond Design Parameters When designing the cooling pond for the application of induction furnace, the following factors must be mainly considered: 1. The hot water or inlet temperature into the pond 2. The cool water or outlet temperature from the pond 3. The operating time occupied in melting and number of heats 4. The solar heat flux or solar energy identified as the main heating mechanisms 5. The pond volume and size corresponding to the equilibrium temperature The hot water into the pond is characterized by the temperature at the total outlet of induction coil, control panel and capacitor bank. The outlet temperature of control panel and capacitor bank specified by the manufacturers is much lower than the outlet temperature of induction coil. It may be neglected in the consideration of total hot water outlet temperature into the pond, but the outlet temperature of induction coil should be examined as an important design parameter. The cool water temperature from the pond should not be below 25ºC (77ºF) recommended by Bailey [10] which has been discussed in the aforementioned chapter. To obtain the required cooled-water temperature, it is largely concerned on the capability of the pond to dissipate waste heat. A cooling pond should also damp out transient temperature fluctuations which arise due to the diurnal cycling or other

58 changes in meteorological conditions. On the other hand, the temperature of the inlet and outlet for cooling pond usually depends on the operating time according to the most economical production. Thus the operating time must be obeyed as the criterion of design parameter. Another parameter is solar heat flux or solar energy identified as the main heating mechanism for pond. The major cooling mechanism is evaporation, but convective cooling to the atmosphere is less significant. Net solar radiation can also account for a significant amount of cooling during night hours. Conductive heat transfer to the ground and pond wall is generally a relatively insignificant process, and it can be neglected. The amount of solar radiation greatly affects the equilibrium temperature, but variation of the amount is usually not known very accurately throughout the day. The equilibrium temperature at which there is no net heat exchange between the atmosphere and the water will be discussed in the next section. Based on the above parameters, the volume and size of the circulating cooling pond can be determined for 0.16 ton coreless induction furnace before constructing. For studying the performance of a cooling pond, it is found that in the heat exchanges occurring in the volume and on the pond surface area, and their influence on climatic fluctuations of location, summarily, the cooling ponds are sized and selected based on the requirement of cooling water supply temperature, the entering temperature into the pond, the number of melting time, solar heat flux received on the pond surface, land utilization and other economical considerations as well as constraints imposed by system components.

4.2. Conceptual Study for Steady-State Cooling Pond Design Generally, the conceptual study before designing and constructing the pond is important for the designer who faces with the problem of land area requirements and high investment costs, especially when diking is necessary. The cost of pond construction and the available water supply dictate that the pond should be built as small as possible. Maximizing the heat transfer from the pond surface, damping transient fluctuations and building a pond as small and as cheaply as possible should be aimed for the designer as the primary emphasis of conceptual study. In order to understand the basic hydrodynamic concepts of various cooling pond configurations as well as to provide the preliminary information to the designer of a cooling pond on the effects of pond volume and size, the water flow rate, the inlet and outlet

59 temperature, some simple, the studies for steady-state cooling pond design are reviewed here. Before discussing, however, an understanding of the classification of ponds is necessary.

4.2.1. Classification of Ponds Harleman et al. [31] classified as deep or shallow pond on their experimental investigation of emergency heat releases from floating nuclear power plants. This classification is dependent not only on the actual pond depth, but upon pond shape and size, the temperature rise in the pond, the flow rate, and the degree of mixing temperature at the pond entrance. Deep or shallow ponds are also distinguished by thermal stratification which a thin buoyant layer of heated effluent floats upon a cooler reservoir. The heated upper layer which is vertically isothermal exhibits only horizontal temperature gradients due to surface cooling while deep reservoir the temperature gradients, if any, are vertical and isotherms horizontal. The above saying from the laboratory temperature profiles is given by Cerco [32]. Hence the classified pond between deep and shallow may be expressed as the degree of pond stratification. The distinction of pond stratification may be quantified in several ways. A simple measure of stratification on pond temperature is shown as the normalized temperature gradient ∆T v / ∆To where ∆T v is the average temperature difference between the surface and bottom of the pond and ∆To is the temperature difference between the pond inflow and outflow. This is related to a parameter which is also called pond number, IP, defined by Harleman et al. [31]. Pond number is calculated from the relation.

⎡ f ′Q Dv L ′ ⎤ IP = ⎢ 4 2 ⎥ ⎢⎣ 4β ∆To gH p w ⎥⎦ 2

3

1 4

where, f ′ = internal fraction factor, Q = water flow rate, ft3/ min Dv = vertical dilution, L' = length of flow path, ft

β = coefficient of thermal expansion, 0.115 g = gravitational constant, 3.44 × 10-8 ft4/ lb. sec4

Equation 4.1

60 Hp = pond depth, ft w = pond width, ft A correction between pond number, IP and the normalized temperature gradient ∆T v / ∆To is shown in Figure 4.1.

∆T v ∆To

1

⎡ f ′Q 2 Dv 3 L ⎤ 4 IP = ⎢ 4 2 ⎥ ⎣⎢ 4β ∆To gH p w ⎦⎥

Figure 4.1. Correlation between Pond Number, IP and Normalized Temperature Gradient, ∆T v / ∆To Source: Cerco (1977) Vertical dilution, Dv may be calculated from Jirka et al. [33] and expressed in terms of discharge densimetric Froude number, Fo′ as follows: Dv = 1 + 1.2( Fo′ − 1)

Equation 4.2

The value of Dv, 1.5 investigated in Ryan and Harleman [34] is taken as the minimum which can be attained in a deep pond.

61 (i) Deep Pond The criterion of the deep pond is IP ≤ 0.3 and it is a desirable configuration in two respects. Firstly, the heated water in all parts of the pond spread out the buoyancy of the surface layer, insuring that the entire surface area is effectively utilized for heat dissipation. Secondly, the deep pond is isolated from diurnal and short-term surface temperature fluctuations and responds only to long term meteorological variations with time scales on the order of the cooling pond residence time. Therefore, it has high thermal inertia that is the capability of a pond to damp out the highly desirable transient conditions. Deep ponds are usually constructed by damming a river or stream creating an artificial impoundment for the electric generating plants. (ii) Shallow Pond Shallow pond is defined as having a pond number, IP ≥ 0.3 and it has the advantage of being free from temperature regulations if constructed off-stream. Based on the degree of stratification, shallow ponds may be further subdivided into partially stratified or vertically well-mixed. Partially stratified ponds exhibit slanted isotherms in a longitudinal cross-section and are characterized by having surface temperatures warmer than the underlying water. Vertically well-mixed ponds exhibit vertical isotherms and are characterized by uniform temperatures from surface to bottom at any location. Cooling pond used for the induction furnaces may be either partially stratified or vertically well-mixed. In shallow ponds, it does not predominate the buoyancy currents as in deep pond. The internal baffles are frequently used in an attempt to direct the throughflow to all portions of the pond.

4.2.2. Equilibrium Temperature and Surface Heat Flux Equilibrium temperature, defined by Cerco [32], is the theoretical temperature of a natural water body at which incoming heat fluxes due to solar and atmospheric radiation are exactly balanced by outward fluxes due to evaporation, back radiation, and conduction. Most methods proposed to determine TE are iterative. It may be calculated iteratively by the method of Ryan and Harleman [34] as TE =

φ sc + f ( W2 )[ β ∗Td + 0.255Ta ] − 1600 23 + f ( W )2 ( β ∗ + 0.255 )

where, φ sc = the clear sky solar radiation, Btu/(day.ft2)

Equation 4.3

62

β ∗ = the proportionality factor Td = the dew point temperature of the air, ºF The wind speed function, f ( W2 ) = 14W2 , is used in the calculation of equilibrium temperature. W2 is the wind speed in mph measured at two meters above the water surface. To iterate the equilibrium temperature, the proportionality factor is computed by using the following formula. ⎛ TE − Td ⎝ 2

β ∗ = 0.255 − 0.0085⎜

⎛ T + Td ⎞ ⎞ ⎟ ⎟ + 0.000204⎜ E ⎝ 2 ⎠ ⎠

2

Equation 4.4

Equilibrium temperature is obtained recursively by assuming a value TE, calculating β ∗ by Equation 4.4, and then calculating a new TE via Equation 4.3. The new value of TE is used to obtain a new β ∗ until the TE obtained from Equation 4.3 matches the value used in the previous iteration. When the equilibrium temperature is obtained, the heat exchange coefficient, K related to the pond surface temperature, Ts and net solar heat flux into the pond surface, φ n , may be calculated from the relation. K=

φn (Ts − TE )

Equation 4.5

At this equation, the net heat flux into the water surface is a non-linear and non-uniform process. The techniques and algorithms to model the net heat flux through the surface of a water body and the transfer of heat across the water surface have been well established in the most comprehensive references which are Ryan and Harleman [34], and Edinger et al. [35]. There are two basic methods to compute the net heat transfer: the complete heat budget and the linearized heat exchange method. The heat budget technique is more accurate than the linearized method based on the equilibrium temperature and surface heat exchange coefficient. In the calculation, an emphasis on the heat budget technique in surface water bodies is applied for the separate calculation of the individual heat flux components. Another method can be found in Shanahan [36]. Surface heat flux consisting of the five radiations and heat flux components is illustrated in Figure 4.2.

63 φsc

φbr φs

φa

φsr

φe

φc

φar

φsn

φan

Figure 4.2. Components of Surface Heat Transfer

The sum five components to arrive at the net surface heat flux are given by Shanahan [36] as follows:

φ n = φ sn + φ an −φ br −φ e − φc

Equation 4.6

where, φ n = the net heat flux into the water surface,

φ sn = the net solar (short-wave) radiation into the water surface, φ an = the net atmospheric (long-wave) radiation from the water surface, φbr = the back (long-wave) radiation from the water surface, φ e = the evaporative heat flux from the water surface, φ c = the conductive heat flux from the water surface, and From Figure 4.2, φ sc = the clear sky solar (short-wave) radiation,

φ s = the solar radiation at water surface, φ sr = the reflected solar radiation, φ a = the atmospheric (long-wave) radiation, and φ ar = the reflected atmospheric radiation All heat flux components have the English system units of Btu/(day.ft2). To change the unit in surface heat transfer calculations, the relationship (1 Btu/day.ft2 = 0.0131 W/m2) can be used because it is hardly completed for hand calculations. In Figure 4.2, the net solar radiation into the water surface is the incoming radiation from the sun, less that absorbed or scattered in the atmosphere, blocked by clouds and reflected at the water surface. The best solar radiation information is predicted from measurements at the site, however these are usually unavailable. If there is not required measuring instrument, calculations should be done based on the solar radiation. The clear sky solar radiation suggested to Ryan and Harleman [34]

64 from the empirical information is computed by accounting for reflectance and cloud cover as follows:

φ sn = 0.94φ sc (1 − 0.65C 2 )

Equation 4.7

where, C = the fraction of the sky covered by clouds (express as fraction) The cloud cover of the sky is usually recorded as the percent (or number of tenths) of the sky that is covered by clouds. Occasionally, the quantity percent possible sunshine will be given. The conversion from percent possible sunshine to cloud cover can be made using an equation derived from relations given by TVA [37] as follows: 2 3

C = [1.2(1 − p s )]

1 2

Equation 4.8

where, ps = the possible sunshine (expressed as a fraction) The calculation of the clear sky solar radiation φ sc is complicated for that the influences from the climatic condition should be determined as a function of the geographical latitude, the declination angle, the time of year, and the hour of the day. In order to obtain the clear sky solar radiation, the extraterrestrial solar radiation on a horizontal surface, φ sco is considered from the correlations. Various climatic parameters have been used in developing empirical relations for predicting the month average global solar radiation or the extraterrestrial solar radiation. Among the existing correlations, the following relation is the generally accepted modified form of the Angstrom-type regression equation from Duffie and Beckman [38].

φ sc S = a+b φ sc So

Equation 4.9

o

The regression coefficients, a and b using the method of least squares can have different varieties of value. An interesting correlation which is believed to be applicable anywhere in the world is given by Rietveld [39] as follows:

φ sc S = 0.18 + 0.62 φ sc So

Equation 4.10

o

where, φ sc = the clear sky solar radiation or the monthly average of the daily global radiation on a horizontal surface at a location, kJ/m2.day

φ sc = the extraterrestrial solar radiation or o

65 the monthly average of the daily global radiation on a horizontal surface at same location on a clear day, kJ/m2.day

S = the monthly average of the sunshine hours per day at the location

S o = the monthly average of the maximum possible sunshine hours per day at the location The extraterrestrial solar radiation on a horizontal surface is calculated from the following equation given as Magal [40].

φ sc = o

⎞ ⎛ n ⎞ ⎛ I sc ⎜⎜1 + 0.033cos⎜ 360 ⎟ × ωs sin φ sin δ + cosφ cosδ sin ω s ⎟⎟ π ⎝ 365 ⎠ ⎝ ⎠

24

Equation 4.11

where, I sc = the solar constant, 5141 W/m2

n = the Julian day number

ω s = the sunset or sunrise angle, degree φ = the latitude of the location, degree δ = the declination angle, degree Copper's equation from Magal [40] has given the following relation for calculating the declination. ⎡ 360 ⎤ δ = 23 .45 sin ⎢ ( 264 + n ) ⎥ ⎣ 365 ⎦

Equation 4.12

The hour angle corresponding to sunrise or sunset on a horizontal surface can be found as follows:

ω s = cos −1 (− tan φ tan δ )

Equation 4.13

The maximum possible sunshine duration is also given by Duffie [38] as follows: So =

2 × ωs 15

Equation 4.14

After determining the clear sky solar radiation and cloud cover in sky, the net solar radiation has been simplified by using Equation 4.7. As stated earlier, the net atmospheric radiation, φ an caused by water vapor, carbon dioxide, ozone and other atmospheric constituents has been formulated by Brunt [41]. It may be written as:

φ an = 2.05 × 10 −0.8 ( 1 + 0.149 ea )( Ta + 460 ) 4 ( 1 + 0.17C 2 )

Equation 4.15

where, ea = the vapor pressure two meters above the water surface, mm Hg Ta = the air temperature two meters above the water surface, ºF

66 The back radiation, φbr form the water surface is given as:

φbr = 4 × 10 −8 (Ts + 460) 4

Equation 4.16

where, Ts = the water surface temperature, ºF Of the various components of the heat budget, evaporation is the most uncertain. Evaporative heat flux is computed as the product of the latent heat of evaporation, the gradient in vapor pressure from the water surface to the overlying air, and an empirical wind speed function. It is recommended by Shanahan [36] as:

φ e = ( 1.03 − 5.1 × 10 −4 T s ) f ( W1 )( esat − ea )

Equation 4.17

where, the wind speed function, f (W )1 has been expressed in many different expressions. For the cooling ponds, Edinger et al. [35] is recommended owing to its through calibration and long standing as follows: f ( W1 ) = 70 + W2

2

Equation 4.18

The Equation 4.15 and 4.17, the vapor pressure two meters above the water surface, ea is given by the following conversion of TVA [37] from dew point temperature to air vapor pressure over water. This equation is valid for relative humidity less than about 95%. ⎡ ⎞⎤ ⎛ 7.5Td − 236.9 + 0.6609 ⎟⎟⎥ ea = exp ⎢2.3026⎜⎜ ⎠⎦ ⎝ Td + 395.5 ⎣

Equation 4.19

where, Td = the dew point temperature of the air, ºF The humidity conditions of the air must be expressed as the water vapor pressure in the equations for calculation of evaporative and conductive heat flux. The vapor pressure of the air is also computed by using either a psychometric chart from relative humidity or the following relationship. ea = RH e sat

Equation 4.20

where, RH = the relative humidity, % esat = the saturation vapor pressure, mm Hg Another heat flux in the calculation of the sum of the five component fluxes is conduction in which a heat diffusion process occurs similar to the moisture diffusion that drives evaporation. Thus, the equation for conductive heat flux is similar in form to that of evaporative heat flux Shanahan [36] suggested that as follows:

φ c = 0.255 f ( W1 )( Ts − Ta )

Equation 4.21

67 As stated previously, the net heat flux from the cooling pond surface is obtained theoretically on the overlying air temperature, incoming solar radiation, moisture contents, cloud cover and wind speed outlined above. Further details can be found in Shanahan [36], Ryan and Harleman [34], Rietveld [39], Magal [40] and Edinger et al. [35]. Note that all above equations concerned the equilibrium and surface heat flux will be used in the calculation of cooling pond design required in 0.16 ton coreless induction furnace.

4.2.3. Traditional Model Utilizing the concepts of equilibrium temperature and linearized surface heat flux reviewed in section 4.2.2, a number of analytical steady-state models for cooling pond design may be formulated by the recommendation of Cerco [32]. There are two of the simplest models: plug-flow and completely mixed ponds.

Water outlet Water inlet

Figure 4.3. Example of Plug-flow Pond

Water outlet

Water inlet

Figure 4.4. Schematic Elevation View of Completely Mixed Pond

In a plug-flow pond, as illustrated in Figure 4.3, there is no mixing between the heated discharge and the receiving waters and no dispersion along the flow path.

68 With long, narrow flow paths, lateral mixing as well as vertical mixing will be suppressed and it is similar to the dispersive model which is incorporated to account for the heat dispersion caused by cross-sectional velocity non-uniformities likely to be encountered in a shallow. The solution of the governing heat conservation equation for the plug-flow pond is analytically given as the classic exponential decay equation by Cerco [32] ∗

Ti = e − r

Equation 4.22

∗

where, T i is a normalized intake temperature. In contrast to the plug-flow pond, the normalized intake temperature is related to the equilibrium temperature as follows: ∗

Ti =

To − TE Ti − TE

Equation 4.23

where, To is the outlet temperature from the pond and Ti is the inlet temperature into the pond. The completely mixed pond may be schematized as in Figure 4.4. In this figure, as the width of a cooling pond increases relative to its length and width, the potential for lateral mixing and dilution is increased. It is also the fully mixed pond in which the inflow is immediately mixed and dispersed throughout the pond, corresponding to infinite entrance mixing and dispersion. For the completely mixed ∗

pond, the normalized intake temperature, Ti is given as: ∗

Ti =

1 1+ r

Equation 4.24

where, r is the pond cooling capacity or the characteristics of cooling pond that is calculated by Cerco [32]. r=

KA p

ρcˆQ

Equation 4.25

In Equation 4.25, Ap is the pond surface area and Q is the water flow rate. The density and specific heat of water are described as ρ and cˆ , respectively. Neither the plug-flow nor completely mixed models are realistic since some entrance mixing and dispersion is always present, but infinite mixing is unlikely. Although they were much used in the past, reflecting on incomplete understanding of cooling pond hydrodynamics and surface heat transfer, the present usefulness of these models is limited to providing theoretical limits on cooling pond performance.

69 ∗

Nevertheless, the plug-flow pond provides a best case, lowest estimate of Ti and the completely mixed pond provides a worst case, highest estimate, assuming the entire pond surface is utilized. The design model consideration of completely mixed pond on the cooling energy requirement is discussed in section 4.3.

4.3. Design Model Consideration

For the design consideration of cooling pond, the following model as well as a lumped parameter model is applied to a problem as shown in Figure 4.5.

Interchange with Atmosphere Hn

Vp = volume Tp = temperature Ap = area

Q, Tp

R = inflow rate

Q = Outflow rate

Cooling Pond

Ti, R

V, Tp

Induction Furnaces

(a)

(b)

Figure 4.5. Illustrative Example of Cooling Pond Model (a) Schematic Diagram of Pond Model (b) Illustration of Energy Balance in Cooling Pond Thermal energy gives as: E = ∫ ρcˆT p dV p v

where ρ is the density of water and cˆ is the specific heat of water. Energy balance equation for the cooling pond is d ( ρcˆTpV p ) = ρcˆTi R − ρQTp cˆ + H n dt = Energy inflow rate - Energy outflow rate + Net heat exchange rate The following assumption is used in the energy balance equation. Assumption: 1. The inflow to and outflow from the cooling pond is steady

70 2. R = Q 3. Vp = constant, ρ = constant 4. No seepage into or from ground water 5. Heat exchange occurs near the surface of the pond only 6. Neglect heat conduction between the surrounding soils The energy balance equation becomes dT p dt

=

Hn Q (Ti − T p ) + Vp ρcˆV p

Equation 4.26

The net heat flux is expressed as:

φn =

Hn Ap

where H is the rate of net heat transfer and Ap is the unit area as a function of wind speed, water surface temperature, air temperature and relative humidity.

φ n = φ sn + φ an −φ br −φ e − φc Formulas for each term are available in section 4.2.2. To illustrate the equilibrium condition in Figure 4.6, it is assumed there exist a simple linear relationship between the net heat exchange rate and surface temperature of the water in the cooling pond.

φn Heat loss

Heat gain

TE

Ts

Surface temperature

φn Net heat flux from pond to atmosphere = 0

Figure 4.6. Illustration for the Equilibrium Condition

Another net heat flux relation is

φ n = − K (Ts − TE )

71 where TE is the equilibrium temperature at which there is no net heat exchange between the atmosphere and the water. This has been discussed in section 4.2.2. Its value varies with locality, solar radiation, wind speed, meteorological conditions and

K is heat transfer coefficient, which has the unit of Btu/(ft2 day ºF). If Ts = Tp is assumed for completely mixed pond, the energy balance equation becomes as follows:

dT p dt

=

A p K (T p − TE ) Q (Ti − T p ) − Vp ρcˆV p

The water retention rate,

Equation 4.27

Ap K Q = kT = k r as constant and the thermal rate, Vp ρcˆV p

as constant are defined as in Equation 4.28. The energy balance equation takes a new form.

dT p dt

= k r (Ti − T p ) − k T (T p − TE )

Equation 4.28

At a steady-state energy system,

dT p dt

=0

Tp → ∞ :

at which T p = T∞

Assuming that Ti is constant, the final relationship is obtained as follows: k r (Ti − T∞ ) = k T (T∞ − TE ) k r [(Ti − TE ) − (T∞ − TE )] = k T (T∞ − TE ) T∞ − TE kr 1 = = Ti − TE k r + kT 1 + r

Equation 4.29

where r = kT/kr, which is a characteristics of a cooling pond. kr and kT can be estimated from temperature data of the site. In the above equations, the inlet and outlet temperature of the pond, the water flow rate, the pond volume, the pond area, and the operating time are related usually vice versa. The net solar heat flux has been always changed throughout the day according to the meteorological parameters. However, in general, the pond volume corresponded to the required operating time should be considered for the cooling capacity by using the assumptions. The correlation for which the temperature in the pond is significantly related to the operating allowable time must be determined to know the performance which would required for the cooling pond design. Based on the design model considerations, the design calculation of cooling pond required for 0.16 ton coreless induction furnace will be described in the next section.

72 4.4. Design Calculation

For design calculation, a schematic diagram of cooling pond has been illustrated in Figure 4.7. The weather condition to compute the solar heat flux is based on the location of Yangon in Myanmar. Yangon, the capital city of Myanmar, is situated at the latitude of 16.45 N. In the collection of the meteorological parameters from the weather station, the hottest month as March with respect to the location is used conveniently for the calculation of solar radiation.

Sun

Ta

φsc

Td

φa Wind W speed 2

φs

φbr φe φc φar

φsr

Q, Ti

Cool water R, To outlet

Ts

Hot water inlet

φsn

TE

φan

T Tb

Subsurface conduction

Ground

Figure 4.7. Heat Transfer Mechanism in Cooling Pond and the Symbolic Notations The circulating water flow rate at design condition is applied to 0.69 m3/min (24.37 ft3/min) which correspond to the sufficient amount of water to all components. In this case, there is no forced withdrawal of circulating water and thus the only make-up required is related to evaporated water, purge and other losses. Generally, the circulating water which enters the pond is assumed to be 33ºC (91.4ºF) and leaves at 28ºC (82.4ºF). To design the pond volume required in induction furnace, the necessary data is collected from the climatic parameters based on 15th March, 2004 and summarized as follows:

Known Data: •

Relative humidity,

•

Ambient air or dry-bulb air temperature,

•

Dew point temperature,

•

Water inlet or hot water temperature into the pond, Ti = 33ºC (91.4ºF)

RH = 62% Ta = 31ºC (88ºF)

Td = 22ºC (72ºF)

73

•

Water outlet or cool water temperature,

•

Latitude of Yangon, φ = 16.45 N

•

Wind speed,

•

Water flow rate,

To = 28ºC (82.4ºF)

W2 = 4 mph Q = 0.69 m3/min (24.37 ft3/min)

Assumption: 1. The water surface temperature on the pond, Ts = 29.25ºC (84.65ºF) 2. The pond temperature at the bottom, Tb = 28ºC (82.4ºF) 3. Steady-state (Completely mixed pond) 4. Inflow rate is equal to outflow rate (Q = R) 5. Volume, Vp = constant 6. Density, ρ = constant 7. Ts = Tp (Completely well-mixed system) 8. At time t = 0, Tp = 28ºC (82.4ºF) 9. No seepage into or from groundwater 10. Neglect heat conduction between the sounding soils 11. Heat exchange occurs near the surface of the pond only

Firstly, it is classified the pond type from Figure 4.1.

∆T v 84.65 − 82.4 = = 0.25 ∆To 91.4 − 82.4 From the correlation graph between IP versus ∆T v / ∆To ,

IP = 0.47

∴ The type of pond is shallow. (Θ IP ≥ 0.3) The average daily global radiation is estimated on a horizontal surface at Yangon (16.45 N) during the month of March, if the average sunshine hours per day are 9.5. Using Equation 4.12,

⎡ 360 ⎤ (264 + 74)⎥ = −10.51ο ⎣ 365 ⎦

δ = 23.45 sin ⎢ From Equation 4.13,

ω s = cos −1 (− tan(16.45) tan(−10.51)) = 86.85 ο = 1.5159rad Day length from Equation 4.14,

74 So =

2 × 86.85 = 11.58 hr 15

Substituting the desired values into Equation 4.11, 360 ⎛ ⎞ × 5141⎜1 + 0.033 cos × 75 ⎟ × π 365 ⎝ ⎠ ( 1.52 sin( 16.45 ) sin( −10.51 ) + cos( 16.45 ) cos( −10.51 ) sin( 86.85 ))

φ sc = o

24

= 34233.62 kJ/(m 2 day) From Equation 4.10,

φ sc

⎛ 9.5 ⎞ = 0.18 + 0.62⎜ ⎟ 34233.62 ⎝ 11.9 ⎠ ∴φ sc = 23106.25 kJ/(m 2 day) = 2034.62 Btu/(day ft 2 ) Then the equilibrium temperature is calculated iteratively by using Equation 4.3 and Equation 4.4. The results are tabulated in Table 4.1 as follows: Table 4.1. Iterative Solutions of Equilibrium Temperature Iterative Step

TE (ºF)

β*

1

Assume: 82

1.42216

2

63.50735

1.227568

3

62.63512

1.219258

4

62.59384

1.218867

5

62.59188

1.218848

6

62.59179

1.218847

7

62.59179

1.218847

∴TE = 62.5917 ≈ 62.59 ο F The possible sunshine hours convert to a fraction,

(9.5 / 24) × 100 = 39.58% = 0.39 The possible sunshine hours, 9.5 hrs is determined for the clear sky solar radiation. From Equation 4.8, 2 3

1 2

C = [1.2(1 − 0.39 ] = 0.7479

Substituting the desired values into Equation 4.7, ∴ φ sn = 0.94( 2035 )[ 1 − 0.65( 0.74 ) 2 ] = 1217 Btu/(day.ft 2 ) = 159 W/m 2

75 Using Equation 4.19, ⎡ ⎛ 7.5( 72 ) − 236.9 ⎞⎤ ea = exp ⎢2.3026⎜ + 0.6609 ⎟⎥ = 20.38 mm Hg ⎝ 72 + 395.5 ⎠⎦ ⎣ Then, the net atmospheric solar radiation is computed by using Equation 4.15. ∴ φ an = 2.05 × 10 −8 ( 1 + 0.149 20.38 )( 88 + 460 ) 4 [ 1 + 0.17( 0.7479 ) 2 ] = 3386 Btu/(day ft 2 ) = 442 W/m 2 From Equation 4.16, the back radiation is ∴ φbr = 4 × 10 −8 ( 84.65 + 460 ) 4 = 3519 Btu/(day ft 2 ) = 459 W/m 2 Also from the Equation 4.20, e sat =

20.38 = 32.87 mm Hg 0.62

From Equation 4.18, wind speed function is f ( W1 ) = 70 + (4) 2 = 86 The evaporative heat flux from Equation 4.17 is ∴ φ e = ( 1.03 − 5.1 × 10 −4 × 84.65 )( 86 )( 32.87 − 20.38 ) = 1060 Btu/(day ft 2 ) = 138 W/m 2

The conductive heat flux from Equation 4.21 is ∴ φ c = 0.255( 86 )[ 84.65 − 88 ] = −73 Btu/(day ft 2 ) = - 9 W/m 2 In the calculation of net solar and net atmospheric radiation, the value of φsn and φan have been determined on the heat gain of the clear sky without paying special attention to the roof. Significantly, the greatest source of heat gain is received from the roofs. Solar radiation coming through the roof can account for a third of the heat build-up in a house or building. For cooling pond, metal sheet is used as the roofing material. It is covered above three feet from the water surface of the pond. In general, the light colored roofing sheets can reflect 25-30% of the radiant heat of sun. Both the color and the type of material affect amount of radiant heat reflected which is measured in reflectance. The reflectance value of a number of roofing materials is described in Table C.4 in the Appendix. Using this table, the reflectance value for metal sheet is substantially recommended in designing cooling pond. The other factors such as shading from existing trees, bushes, hill and insulation of the roof are assumed as 15% reflectance. Hence the sum of reflectance is defined as 85% in preventing the large

76 amount of heat build-up for pond. Approximately, 15% of total heat flows into the pond, even when the clear sky solar radiation is fully gained. ∴ φ sn =

1217 × 15 = 182 Btu/(day ft 2 ) = 23 W/m 2 100

∴ φ an =

3386 × 15 = 508 Btu/(day ft 2 ) = 66 W/m 2 100

Summing up the resulting terms, the net heat flux is given by Equation 4.6. ∴ φ n = 182 + 508 − 3519 − 1060 + 73 = −3816 Btu/(day ft 2 ) Neglecting the negative term, φ n = 3816 Btu/(day ft 2 ) = 498 W/m 2 When the resulting value is substituted in Equation 4.5, the heat exchange coefficient becomes as follows: K=

3816 = 173 Btu/(day ft 2 ο F) = 22 W/m 2 ( 84.65 − 62.59 )

From Equation 4.23, the normalized intake temperature is ∗

Ti =

82.4 − 62.59 = 0.6876 91.4 − 62.59

For completely mixed pond, Equation 4.24 is used. 1 1+ r r = 0.4543

0.6876 =

Using ρ = 62.4 lb/ft3 and cˆ = 1.003 Btu/ (lb ºF), Equation 4.25 becomes as follow: 0.4543 =

( 173 / 1440 ) A ( 62.4 × 1.003 × 24.37 )

∴ A p = 5768 ft 2 = 536 m 2 Only the area availability for cooling pond is considered in all above cases. It is noted that the pond volume is realized in construction. For the cooling pond used in the cooling system of induction furnace, it is impossible without any depth in the pond construction in terms of cost-effectiveness and sufficiency of water. To be effective and to be built as small as possible, the aeration placed on the surface of the cooling pond has been implemented to unit depth, which is critical for uniform mixing and the prevention of short-circuiting especially in shaped cooling ponds. Thus the volume of cooling pond is required as follows: ∴V p = 5768 ft 3 ≈ 49 ft × 20 ft × 6 ft = 163 m 3 ≈ 15 m × 6 m × 1.8 m

77 In practice, the volume of cooling pond 6000 ft 3 ≈ 50 ft × 20 ft × 6 ft is used for the cooling system of 0.16 ton coreless induction furnace. Now, the relationship between the raising temperature on the pond surface and the time duration influenced on the operating limitation has been calculated by using the pond volume, Vp = 6000 ft3 and pond surface area A p = 50 ft × 20 ft = 1000 ft 2 . It is assumed that the temperature, Tp in the cooling pond is 28ºC (82.4ºF) at the operating time, t = 0 (without running induction furnaces). From Equation 4.26, the water retention rate and the thermal rate are kr =

24.37 = 4.06 × 10 −3 6000

kT =

1000 × 173 = 0.3199 × 10 −3 62.4 × 1.003 × 6000 × 1440

Substituting the kr and kT values into Equation 4.28, dT p dt

= −4.3799 × 10 −3 T p + 0.3911

Equation 4.30

Considering the rate of change of temperature, dTp/dt, Equation 4.30 is a linear first-order ordinary differential equation that is solved at each step using the exponential function as an integrating factor. The detailed solving is described as follows: Let

x1 = 4.3799 × 10 −3 , ∴

Let

x 2 = 0.3911

dT p

= − x1T p + x 2 dt T ′ = − x1T p + x 2

T ′ = y ′, T p = y

y ′ = − x1 y + x 2 y ′ + x1T p = x 2 It is compared with y ′ + f ( x) y = r ( x) ∴ f ( x) = x1 , r ( x) = x 2

For the solution, h( x) = ∫ f ( x)dx = x1 ∫ dx = x1 x

The general solution is given as:

[

y = e − h ( x ) ∫ e − h ( x ) r ( x)dx + C where, C = arbitrary constant

]

78

[

y = e − x1 x ∫ e x1x x 2 dx + C

]

⎡x ⎤ = e − x1 x ⎢ 2 e x1x + C ⎥ ⎣ x1 ⎦ x = 2 + Ce − x1x x1

Similarly,

Tp =

x2 + Ce − x1t x1

Equation 4.31

Defining the initial value, Tp(0) = 82.4ºF ∴ t = 0, Tp = 82.4ºF Substituting in Equation 4.31, ∴ C = 82.4 −

x2 x1

Again, it is substituted in Equation 4.31.

x2 ⎡ x ⎤ + ⎢82.4 − 2 ⎥Ce − x1t x1 ⎣ x1 ⎦ − 0.0043( t ) = 89.29 − 6.89e

Tp =

Finally, the relationship between the temperature and operating time is obtained as follows: ∴ T p = 89.29 − 6.89e −0.0043(t )

Equation 4.32

The tabulated form of Equation 4.32 is given as in Table 4.2.

Table 4.2. Resulting Values of the Water Temperature and the Operating Time Operating Time, t

Water Temperature, Tp

hour

ºF

ºC

0

82.4

28

1

83.99429958

28.88572199

2

85.2200326

29.56668478

3

86.16240343

30.09022413

4

86.88691911

30.49273284

5

87.4439429

30.8021905

6

87.87219519

31.04010844

7

88.20144508

31.22302504

79 From the above tabulated form, when the temperatures with the retention time have been limited approximately between 28ºC (82.4ºF) and 31.04ºC (87.87ºF), the volume of pond 170 m3 (6,000 ft3) would amount to six hours hold up to maintain fairly constant discharge temperature throughout the operating condition. Thus the trade-off between usage of the operating time and the temperature in the pond must be evaluated and adjusted for a pond designer to be optimized the efficiency of cooling pond required in 0.16 ton coreless induction furnace. Moreover, the effect of pond configurations will be described in the results of studies.

80

CHAPTER 5 EVAPORATIVE COOLING TOWER SYSTEM

The cooling system of induction furnace, treated in the preceding chapter, is considered with the evaporative cooling tower system. In this system, induced draft counterflow cooling tower, one of the classifications of cooling tower, is designed theoretically instead of cooling pond that requires the larger ground area. Cooling tower fundamentals and the concepts needed for the design calculation are introduced in this chapter. The removing heat amount, the required temperature range, wet bulb temperature and other related parameters in tower design have been used as the same data in the pond design. Tower coefficient is the primary factor that influences the effectiveness of the cooling tower. Operation considerations for cooling tower are covered in this chapter after designing.

5.1. Cooling Tower Fundamentals Cooling towers have been widely used to dispose of waste heat from industrial processes. It is cheaper or more convenient to reject heat to the atmosphere rather than to water in a nearby river, lake, or ocean. It is also one of the heat exchangers which are used in many commercial processes. The principle of cooling tower is very simple, but the heat transfer process is quite complex. When water changes its state from liquid to vapor or stream, an input of heat energy must take place which is known as the latent heat of evaporation. This input energy must either be supplied from fuel as in a boiler or be extracted from the surroundings. Cooling towers take advantage of this change of state by creating conditions in which hot water evaporates in the presence of moving air; by this means heat is extracted from the water and transferred to the air and the process is known as evaporative cooling. Any cooling towers consist of no more than a four-sided structure in which the hot water is introduced as a spray at the top of the tower, and mixed with the cooling air and drawn off from a sump at the bottoms. The water is thus cooled for return to the machine or process. This procedure can be seen in the

81 preceding chapter two in which the operating function of electric induction furnace with evaporative cooling tower was described in types of water cooling system. Other interesting fundamentals of cooling tower will be studied in the next sections.

5.1.1. Principal Criteria The principal criteria on which the design and manufacture of cooling tower are based on the followings. 1. Achieving maximum contact between air and water in the tower by the optimum design of tower packing and water distribution. 2. Assisting the flow of air by meant of fans. 3. Minimizing the loss caused by water spray escaping frown the tower; control of spray loss is also of great importance in eliminating the risk of infectious diseases being transmitted to people by warm moist air. 4. Relating the design of the tower to the volume flow rate of the water to be cooled and to the three critical temperatures (ambient air wet bulb, warm water input and cooled water output). 5. Ensuring that problems arising from the quality of water such as corrosion, fouling and growth of bacteria are properly understood and controlled. 6. Taking due account of space limitations at the tower location and of the possibility that noise from the tower may be a source of nuisance to those living or working in the vicinity.

5.1.2. Classification of Cooling Towers Cooling towers can be classified by the type of heat transfer; the type of draft and location of the draft, relative to the heat transfer medium; the type of heat transfer medium; the relative direction of air and water contact; and the type of water distribution system. All employ stacked horizontal rows of fill to provide increased contact surface between the air and water. Generally, the two main sub-divisions utilized the two different configurations are expressed as follows: 1. Mechanical draft cooling tower (a) Induced draft type - Counterflow - Crossflow (b) Forced draft type

82 - Counterflow - Crossflow 2. Natural circulation cooling tower (a) Atmospheric tower (b) Natural draft tower - Counterflow - Crossflow The diagrams of the various tower types and configurations are shown in Figure 5.1 and Figure 5.2. Air outlet

Air outlet

Fan

Water inlet

Water inlet

Fan

Fill

Fill

Air inlet

Air inlet

Air inlet Water outlet

Water outlet

(b) Forced Draft Counterflow

(a) Induced Draft Counterflow Water inlet

Air outlet

Fan Water inlet Fill

Air inlet

Fill Water inlet

Air outlet

Air inlet

Water outlet

Water outlet

(c) Induced Draft Crossflow

(d) Forced Draft Crossflow

Figure 5.1. Mechanical Draft Cooling Towers

In Figure 5.1, the two types of mechanical draft towers are the commonest in use today, and of these the vast majority is now induced draft towers. Since there are advantages to its use which exceed all others except under very special conditions, the induced draft tower has been pronounced and used for the designer, manufacturer and customer. In this system, the air discharges through the fan at a high velocity so that it is driven up into the natural air currents which prevent it from settling at the air intake. The induced draft counterflow cooling tower reduces the recirculation for the performance capability of cooling tower. However, the pressure drop is on the intake

83 side of the fan, which increases the total fan-power requirements. On the other hand, the air can enter along one or more entire lengths at wall, and as a result the height of tower required for air entry is very small. In the forced draft tower, the fan is mounted at the base, and air is forced in at the bottom and discharged at low velocity through the top. This arrangement has the advantage of locating the fan and drive outside the tower, where it is convenient for inspection, maintenance, and repairs. However, because of the low existing air velocity; the forced draft tower is subjected to excessive recirculation of the humid exhaust vapors back into the air intakes. A relatively large ineffective height and volume is used for opening a circular fan and air inlet space. Fill Water inlet

Air inlet

Air outlet

Water outlet

(a) Atmospheric Tower Air outlet Air outlet Water inlet Fill Air inlet

Canopy Fill

Air inlet

Air inlet

Air inlet

Water outlet

(b) Counterflow Natural Draft Tower

(c) Crossflow Natural Draft Tower

Figure 5.2. Natural Circulation Cooling Towers

The atmospheric tower, illustrated in Figure 5.2, avails itself of atmospheric wind currents. Shifting with the time of year and other atmospheric conditions, the air is blown through the louvered sides or fills in one direction at a time. Since the atmospheric currents must penetrate the entire width of the tower, the towers are made very narrow in comparison with other types and must be very long to afford equal capacity. Drift losses occur over the entire side and are greater than for other types. These towers make less efficient use of the available potential, since they operate in

84 crossflow. One great advantage of atmospheric tower is the elimination of fan power cost. The natural draft cooling towers are operated in the same way as a furnace chimney. Air is heated in the tower by the hot water it contacts, so that its density is lowered. It must be tall for sufficient buoyancy and must have large cross sections because of the low rate at which the air circulates compared with mechanical draft. Although the cost of fan power is eliminated in the natural draft tower, the pumping power is more consumed than the others. However, it may be reliable in some localities rather than atmospheric tower. Depending on the relative flow directions of water and air, it is further classified into counterflow and crossflow design as illustrated in Figure 5.1 and Figure 5.2. Thermodynamically, the counterflow arrangement is more efficient, since the coldest water contacts the coldest air, thus obtaining maximum enthalpy potential. Kern [17], Perry [42], Hensley [43] and London et al. [44] proposed that induced draft counterflow cooling tower, derive their primary cooling effect from the evaporation that takes place when air and water are brought into the direct contact and it gives the best efficiency. From the points above, the latter emphasis is based on the induced draft counterflow cooling tower, resulting in increase heat transfer efficiency.

5.1.3. Main Components and Tower Operation The schematic arrangement of a mechanical induced draft counterflow cooling tower with the cutaway section is shown in Figure 5.3. The typical layout of the main components which are defined below: 1. Casing or Shell: The structure enclosing the heat transfer process reinforced as necessary to carry the other main items. Traditionally, corrugated asbestos-cement board panels have been used for the large cooling tower casings, although their uses are fading rapidly. Currently, wood or steel framed, field-erected towers are similarly cased with free-retard and fiber-reinforced polyester corrugated panels, over lapped and sealed to prevent leakage for factory-assembled tower, as illustrated in Figure 5.3, the heavy gauge hot-dip galvanized steel is utilized for long life and durability. 2. Fan and Fan deck: The correct selection of fan according to the tower duty is the major importance. Volumetric air flow rate, fan pressure developed and noise from motor, and fan impeller must also be considered according to the duty and

85 location of the tower. Fan deck also provides a platform for the support of the fan cylinder, as well as an access way to the mechanical equipment and water distribution system.

2

5 1. Casing 2. Fan 3. Fill 4. Basin 5. Drift eliminator 6. Water distribution system 7. Hot water inlet 8. Cool water outlet

6 1

7

3 4 8

Figure 5.3. Cutaway View of Induced Draft Counterflow Cooling Tower Source: Hensley (1998) 3. Fill or Packing: It consists essentially of a system of baffles which slows the progress of the warm water through the tower. The baffles are ensured the maximum contact between water droplets and cooling air by maximizing surface and minimizing water film thickness. There are many different types of packing and these extensive representations are expressed in the latter section. 4. Cold Water Basin: It is also referred to as tank or sump. The cooled-water is collected before return to the edge of process. 5. Drift Eliminator: This is positioned in the outlet air stream as to prevent water droplets from being carried away from the tower by the air stream. The drift eliminators are constructed of an inert polyvinyl chloride (PVC) plastic material which effectively eliminates corrosion of these vital components. They are assembled in sections to facilitate easy removal for inspection of the water distribution system. An efficient drift eliminator is greatly illustrated in Figure 5.4. 6. Water Distribution System: Water entering the tower must be spread as evenly as possible over the cross section of the tower. This system uses the spray nozzles or diffusers, and made of PVC pipe and the plastic water diffusers for corrosion and protection. The water distribution system, internally as shown in Figure 5.5, is

86 simplified with only one header which is easily removable for cleaning. They also have an anti-sludge ring extending into the headers to prevent drift from building up in the diffuser opening. In general, the piping and distribution system of water within the envelope of the tower are responsibilities of the tower manufacturers.

Figure 5.4. Drift Eliminator used in Induced Draft Counterflow Cooling Tower

Figure 5.5. Water Distribution System

As shown in Figure 5.3, the operating principle of tower, the warm water from the heat source is pumped to the water distribution system at the top of the tower. The water is distributed over the wet deck fill by means of large orifice nozzles. Simultaneously, air is drawn in through the air inlet louvers at the base of the tower and travels upward through the wet deck fill opposite the water flow. A small portion

87 of the water is evaporated which removes the heat from the remaining water. The warm moist air is drawn to the top of the cooling tower by the fan and discharged to the atmosphere. The cool water drains to the basin at the bottom of the tower and returns to the heat source.

5.1.4. Cooling Tower Fill The fill within the cooling tower is the critical thermal component. In cooling tower, it is the medium over which the hot water is distributed as it is being cooled. Its primary function is to provide an environment for efficient heat transfer. The two basic fill classifications are splash type and film type. With splash fill typically as shown in Figure 5.6, the exchanger area for cooling is provided by water droplets, and heat exchange occurs on the surface of these droplets. Due to surface tension, these droplets are nearly perfect spheres, and therefore this type of fill does not provide maximum possible surface area. Current splash fill is primary various shapes of extruded PVC, and is still used for projects with major fouling concerns. Significant improvements in evaporative heat transfer efficiency were realized with the invention of film fills.

Figure 5.6. Illustration of Typical Splash Fill

Figure 5.7. Illustration of Typical Film Fill

88 Film fill of the form of Figure 5.7 consists of flat or formed sheets to provide a surface upon which water and air come in contact for heat exchange. Cellular fill stretches droplets of water into the thin films as the water proceeds vertically downward through the cells, thereby maximizing the surface area and permitting the available crossflow or counterflow air to cool the entire droplet more rapidly. Film fill creates a very large surface area on which the hot water disperses itself as a thin film that interacts with the air traveling through the tower, thus achieving the specified cooling. It is more compact than splash fill and can thereby be more cost effective. Film fill is used in both counterflow and crossflow cooling towers, from small applications to the large towers serving power plant around the world. Since film fills have defined the heart of the cooling tower, film fill technology had evolved from the use of wood panels, galvanized steel plates, brick and asbestos cement board to plastic film fills. Different shapes and configurations of fill materials are made by the manufacturers and sold by the companies. Cooling Technology Institute (CTI) recommended the standard for fill material according to their durability, long service life, self-extinguishing characteristics, and the ability to create a uniform water film (wetting). In the form of Rasching rings, Berl saddles, or other objects given in Kern [17], the fill characteristics can be found with the fixed liquid and gas loading. Many of the data for the fill or packing are available on laboratory-scale apparatus or in fill designer.

Water flow

Water flow

Water flow

Air

Air

(a) Cross-Corrugated Fill

Air

(b) Vertical-Offset Fill

Air

(c) Vertical Fill

Figure 5.8. Typical Film Fill Shape and Texture

Extensively, as shown in Figure 5.8, three broad categories: cross-corrugated, vertical-offset and vertical in film fill are described by the fill designer. Each fill category will have fill design with differing thermal performance. Cross-corrugated fills can be used in both counterflow and crossflow towers. They work by separating

89 the water and air paths in opposing angles through the pack. Vertical-offset fills are used only in counterflow cooling towers. Water enters the pack from the top and flows vertically. These patterns of angular transition occur continuously through the vertical pack depth. Vertical flow fill also are used only in counterflow towers. They keep the water and air path oriented vertically, with no offsets, through the pack. The benefits of high performance of film fill can occur in vertical flow fill designs which can accommodate poor water quality water characteristics and with a reduced potential for fouling. Nevertheless, the high performance film fills, especially either vertical fill or vertical-offset fill should be selected for the induced draft counterflow cooling tower, when good quality water makeup system are used with no contamination process and/ or low airborne contamination. By considering the thermal performance of fill products and matching it to the application, the cooling tower design can ensure high efficiency and optimizing costs. Further fill fundamentals can be studied in Henskey [43] and Kern [17].

5.2. Conceptual Study for Induced Draft Cooling Tower System In the preceding section the fundamentals and ideas of cooling tower have been considered as the important facts. Most cooling towers used in plants for industrial applications are mechanical induced draft cooling towers. The large surface area of water in this tower is provided by spraying the water through nozzle or splashing the water down the tower from one baffle to another. These baffles or fill materials or packing are essential in the designing of cooling tower. For design, the conceptual study for induced draft cooling tower is expressed in the following section, so that tower size and tower capacity can be calculated.

5.2.1. Cooling Tower Theory The most generally accepted concept of cooling tower heat-transfer process is that developed by Merkel [45]. This analysis is based upon enthalpy potential differences as the driving force and also combines the sensible and latent heat transfer into an overall process. Merkel demonstrated that the total heat transfer is directly proportional to the difference between the enthalpy of saturated air at the water temperature and the enthalpy of air at the point of contact with water. Qt = K x S(H ′-H)

Equation 5.1

90 where, Qt = the total heat transfer rate, W Kx = the overall enthalpy transfer coefficient, kg/ (hr.m2) S = heat transfer surface, m2

H ′ = enthalpy of air-water vapor mixture at the bulk-water temperature, J/kg dry air

H = enthalpy of air-water vapor mixture at the wet bulb temperature, J/kg dry air The heat transfer surface is equal to aV, in which a means the surface of packing or fill per unit of tower volume, ft2/ft3 and V is the effective tower volume, ft3. The water temperature and air enthalpy are being changed along the fill and Merkel relation can only be applied to a small element of heat transfer surface by dS. From Equation 5.1,

dQt = d[K x S(H ′-H)] ∴ dQt = K x (H ′-H)dS

Equation 5.2

The heat transfer from water side is as follow:

Qt = cˆL(T2 − T1 )

Equation 5.3

where, cˆ = the specific heat of water, J/kg ºC

L = water flow rate or liquid loading, kg/ (hr.m2) T2 = leaving water (cold-water) temperature, ºC T1 = entering water (hot-water) temperature, ºC The heat transfer rate, Qt from Equation 5.3 is substituted in Equation 5.2 and it becomes as:

dQt = d[cˆL(T2 − T1 )] = K x (H ′-H)dS ∴ cˆLdT = K x (H ′-H)dS

Equation 5.4

The heat transfer rate from air side is as follows:

Qt = G(H ′-H) where, G = air mass flow rate or gas loading, kg/ (hr.m2) Similarly,

dQt = d[G(H ′-H)] ∴ dQt = d[G(H ′-H)] = K x (H ′-H)dS ∴ GdH = K x (H ′-H)dS

By integrating to Equation 5.4 and 5.5,

Equation 5.5

91 K x S K x aV dT = =∫ L L H ′-H

K x S K x aV dH = =∫ G G H ′-H

and

By summarizing and rearranging, the Merkel equation for cooling tower theory becomes as follows: nd =

T1 dT K x aV =∫ T2 H ′-H L

Equation 5.6

nd =

H1 dH K x aV =∫ H 2 H ′-H G

Equation 5.7

The above equations are the key equation for the calculation of the design and the analysis of performance in cooling towers. The right-hand side of Equation 5.6 is entirely in terms of air and water properties and is dependent on tower dimensions. The left-side in it is defined as tower coefficient or the number of diffusion units, nd. For the detailed derivation of Merkel theory, the extensive explanations can be found in Kern [17], Wang [46], and Baker and Shryock [47].

5.2.2. Heat-Balanced Process The heat balance process of counterflow cooling tower is illustrated graphically in Figure 5.9.

H' (Hot water temperature)

A

H’-H

H (Air out)

D Air operating line

dH dT

C'

Saturation curve

C

L/G

Wet bulb temperature in

Approach

Range

Cold water temperature

H (Air in)

B

T2

Hot water temperature

H' (Cold water temperature)

Wet bulb temperature out

Enthalpy, Btu/ lb dry air

Water operating line

T1

Temperature, ºF

Figure 5.9. Process Heat Balance Diagram of Counterflow Cooling Tower

92 In a counterflow induced draft tower, water and air relationships as the driving potential exist at the place where air flows parallel but opposite in direction to water flow. In Figure 5.9, the water operating line is shown by line AB and is fixed by the inlet and outlet tower water temperature. Water entering the top of the tower at T1 is surrounded by an interfacial film that is assumed to be saturated with water vapor at the bulk water temperature. This corresponds to point A on the saturation curve. As the water is cooled to T2, the film enthalpy follows the saturation curve to point B. Air entering the base of the tower at wet bulb temperature TWB has an enthalpy corresponding to C ′ on the saturation curve. The initial driving force at the base of the tower is represented by the vertical distance BC which is the concept similar to

H ′ - H . Heat removed from the water is added to the air. So its enthalpy increases along the straight line CD, having a slope equaling the L/G ratio and terminating at a point vertically below point A. The area ABCD in the diagram is known as tower characteristic, varying with L/G ratio. This is also very key in the design of cooling towers. For fixed ambient and water conditions, the tower performance is predicted from the tower characteristic. The tower characteristic, KxaV/L, as described in Equation 5.6 or 5.7, can be determined by the various methods. Solving the tower characteristic will be presented in the calculation of the next section. From the heat balance diagram, the actual enthalpy can be determined at any point in the counterflow tower. It relates the change in enthalpy in the air phase dH to the accompanying change in the water temperature dT in contact with the air. This change can be represented by graphically in Figure 5.9. Thus

L dH = G dT

Equation 5.8

which is the equation of a straight line in Figure 5.9 whose slope is the ratio of the liquid to air loading L/G. The value of H at any point on the operating line is given by

H 2 = H1 +

L (T2 − T1 ) G

Equation 5.9

Since the enthalpy of the inlet air H1 is known or can be readily determined, all above concepts are used in the design consideration of cooling tower.

5.2.3. Tower Coefficients The dimensionless tower coefficients or tower characteristics, also known as the number of diffusion units, measures the degree of difficulty of the cooling process

93 required. The larger this value is, the more difficult the cooling duty. From the Merkel equation, Kxa is one of tower coefficients and it is also defined as the height of a diffusion unit (HDU) by Kern [17]. The HDU or Kxa are the performance characteristics of the given fill or packing. The performance of a variety of packing and fills has been given by the fill manufacturers and fill designers according to their liquid and air loading. To meet the theoretical tower characteristic, the suitable fill design must be used for the cooling duty. The Pritchard Corporation of California has developed performance data on various tower-fill designs. These data are too extensive to include here and those interested should consult this reference. In the calculation of cooling tower design, a particular fill design owing to Kxa 200 lb/ (hr ft3) (lb/lb), corresponded on L/G ratio 0.18, will be used from the available data reported on experiments of fill manufacturer. Many laboratory scale data have been reported with values of Kxa ranging from 200 to 5,000 lb/ (hr ft3) (lb/lb). Normally the air and liquid loading of from 200 to 5,000 lb/ (hr ft2) have been used. Many of the data for humidification characteristics such as Raschig rings and Berl saddles have been obtained in Kern [17], but liquid and air loading is limited between 500 and 3,000. The relation between the HDU and Kxa is given by Kern [17] as follows: HDU =

V Z = t Kxa L

Equation 5.11

where, HDU = the height of a diffusion unit, m

Z = the height of cooling tower, m Vt = tower volume, m3 L = liquid loading, kg/ (hr m2) of ground area When the value of Kxa is multiplied by the ratio of Vt/L, the number of diffusion units in a given tower height can be obtained for the fulfillment of process conditions, since Vt = 1Z. It has been applied in the design calculation of cooling tower.

5.2.4. Factors Affecting on Cooling Tower Performance The factors which influence the performance of a cooling tower may be summarized as follows: 1.

The cooling range (Hot water - Cold water temperature)

94 2.

The approach (Cold water - Wet bulb temperature)

3.

The ambient air wet-bulb temperature

4.

The flow of water to be cooled (or circulation rate)

5.

The rate at which air is passed over the water

6.

The tower or performance coefficients of the packing to be used

7.

The volume of packing (height multiplied by horizontal cross-sectional area)

8.

The tower sitting and orientation (It is the responsibility of the owner/ specifier) Item 6 depends on the how effectively the pack is designed and it follows that

the volume of the pack in items 7 will directly affect tower performance.

5.3. Design Calculations In the calculation of induced draft counterflow cooling tower, the heat transfer rate, the inlet and outlet temperature, and other design parameters given in the cooling pond design are used to meet the design equivalent. These parameters are as follows: Total heat transfer rate,

Qt = 67.3469 kW (229,843 Btu/hr)

Wet bulb temperature,

Tw = 25ºC (77ºF)

Hot water temperature,

T1 = 33ºC (91.4ºF)

Cold water temperature,

T2 = 28ºC (82.4ºF)

Obviously, water and air circulating rate are a key number in the original design. To specify the water and air ratio, the model ICT 4.59 for the product of EVPCO's ICT cooling tower, one of the original design models, is used and is also corresponded to the required parameter such as wet bulb temperature and hot and cold water temperature. This gives the fan capacity, 19,700 cfm and the total water flow rate, 200 gal/min. The cooling tower has been constructed on the ground area 8 ft × 4 ft and there is no change in the entering air mass flow rate.

The cooling range = T1 – T2 = 9ºF The approach = T2 – Tw = 5.4ºF Ground area = 8 ft × 4 ft = 32 ft2 Liquid loading,

L=

Total water flow rate = 3,125 lb/(hr ft 2 ) Ground area

From air-water vapor-mixture table recommended in Kern [17], The enthalpy at 77ºF,

H1 = 41.1 Btu/lb dry air

95 Specific volume (air + water) = 13.964 ft3/lb The density of the dry air in a cubic foot of mixture = G = 19,700 × 0.0716 ×

Air loading,

∴

1 = 0.0716 lb/ft3 13.964

60 = 2,645 lb/(hr ft 2 ) 576

L = 1.1813 G

Using the Equation 5.9,

H2 = 51.73 Btu/lb dry air By using the numerical integration method, cooling tower characteristics can be expressed in tabulated form as in Table 5.1 that describes the calculations for the process requirements.

Table 5.1. Enthalpy Difference by Using the Numerical Integration Method

(H ′-H)av

dT ( = 1ο F) (H ′-H)av

5.83

5.83

0.1715

43.46

5.83

5.83

0.1715

50.53

44.64

5.89

5.86

0.1706

86.4

51.87

45.82

6.05

5.97

0.1675

87.4

53.21

47.00

6.21

6.13

0.1631

88.4

54.55

48.18

6.37

6.29

0.1589

89.4

55.89

49.36

6.53

6.45

0.1550

90.4

57.30

50.55

6.75

6.64

0.1506

91.4

58.80

51.73

7.07

6.91

0.1447

T, ºF

H′

H

H ′-H

82.4

46.93

41.1

5.83

83.4

48.11

42.28

84.4

49.29

85.4

nd =

K x aV dT =∫ = 1.4534 L H ′-H

From the above table, the cooling tower must be capable of performing 1.4534 diffusion units to fulfill process requirements. On coordinates of enthalpy versus water temperature plotted the saturation line from air-water vapor-mixture table are shown in Figure 5.10 as the graphical representation of tower characteristic.

96

70

Enthalpy, Btu/lb dry air

65 H'2

60 55 50

H'1

H2=51.73


45

Air operating line

L/G

40

H1=41.1

35 30 25 93.4

92.4

90.4 91.4

89.4

88.4

87.4

86.4

85.4

84.4

83.4

82.4

81.4

80.4

20

Water temperature, ºF

Figure 5.10. Enthalpy-Temperature Diagram of Air and Water

Next requirement is the calculation of the required height of fill. To determine that, a particular fill at L = 3125 and G = 2645, Kxa is 200 lb/ (hr ft3) (lb/lb), is used from the manufacturers of cooling tower which guarantee the tower to cool a specified flow rate of water when the wet bulb temperature of air is 77ºF. A detailed fill performance can be found in the products of EVPCO's ICT cooling tower. On the loading is based on 1 ft2 of ground area, the tower height is determined by using Equation 5.11. nd =

Tower height, Z =

K x aV K x a( 1Z) = L L

n d L 1.4534 ÷ 3125 = = 22.7 ft = 6.91 m 200 Kxa

The height of a diffusion unit,

HDU =

Z 22.7 = = 15.6 ft = 4.75 m nd 1.4534

∴ Fill volume = 8 ft × 4 ft × 15.6 ft = 499 ft 3 = 14.12 m3

∴ Tower volume = 8 ft × 4 ft × 22.7 ft = 726 ft 3 = 20.55 m3 Finally, the tower height is given 22.7 ft that is approximately satisfy with the values of Perry [42] in which a relatively low cooling tower (15 to 20 ft) will suffice when a wide approach of 15 to 20ºF to the wet bulb temperature and 25 to 35ºF cooling range required. The L/G value used in the calculation is also between the ranging 0.75 and 1.5, and the value 1.1813 of KxaV/L is usually obtained. These values are corresponded to the values of Perry [42]. Perry [42] proposed that

97 mechanical draft cooling towers normally are designed for L/G ratios ranging from 0.75 to 1.50; accordingly, the values of KxaV/L vary from 0.50 to 2.50. The tower coefficient or characteristic KxaV/L can be calculated by using the various methods. With the similar range and approach temperature, KxaV/L is solved by using the Chebyshev method for numerically evaluating the integral as described in Perry [42], where by T1 K x aV T − T2 dT =∫ ≅ 1 T2 H ′ − H 4 L

⎛ 1 1 1 1 ⎞ ⎟⎟ ⎜⎜ + + + ⎝ ∆H 1 ∆H 2 ∆H 3 ∆H 4 ⎠

where, ∆H1 = value of (H'-H) at T2 + 0.1 (T1 - T2)

∆H2 = value of (H'-H) at T2 + 0.4 (T1 - T2) ∆H3 = value of (H'-H) at T1 - 0.4 (T1 - T2) ∆H4 = value of (H'-H) at T1 - 0.1 (T1 - T2) In Table 5.2, the calculations for the process requirement are described as follows: Table 5.2. Enthalpy Difference by Using the Chebyshev Method

T, ºF

H'

H

T2 = 82.4

46.93

h1 = 41.1

T2 + 0.1 (1) = 83.3

47.99

T2 + 0.4 (1) = 86

∆H

1/ ∆H

h1 + 0.1 L/G(9) = 42.16

∆H1 = 5.83

0.1715

51.34

h1 + 0.4 L/G(9) = 45.35

∆H2 = 5.99

0.1669

T2 + 0.4 (1) = 87.8

53.75

h2 - 0.4 L/G(9) = 47.47

∆H3 = 6.28

0.1592

T2 + 0.1 (1) = 90.5

57.45

h2 - 0.1 L/G(9) = 50.66

∆H4 = 6.79

0.1472

T1 = 91.4

58.80

h2 = 51.73 0.6448

∴

K x aV 91.4 − 82.4 = (0.6448) = 1.4508 4 L

Another method is nomograph approximation for a given cooling duty. It is provided the approximate degree of accuracy varied with changes in cooling as well as from tower to tower. The nomograph of cooling tower characteristics is shown in Figure B.4 in the Appendix. According to the same temperature range and approach, the straight line is drawn on the nomograph at the wet bulb temperature 77ºF. Because the L/G ratio is calculated to 1.1813, KxaV/L may be established by a line drawn through L/G = 1.1813 and parallel to the original line. Thus the tower characteristic,

KxaV/L is established at 1.62.

98 Another method is software method produced form the Cooling Technology Institute (CTI). Now CTI is given by the output of the calculation, KxaV/L with its Toolkit Software, Version 3.0 underlying the computer. The resulting dialog box is shown in Figure 5.11 as below:

Figure 5.11. Toolkit Software Dialog Box From Toolkit Software, KxaV/L, which is a measure of the required duty imposed on a cooling tower, is determined as 1.4922. Comparing the results among the various methods, the comparative visualization of the tower characteristics is shown in Figure 5.12. 1.7 1.62

1.65

1.45 1.4 1.35

1.4922 Software

1.4508

Nomograph

1.4534

Chebyshev

1.5

Numerical

KxaV/L

1.6 1.55

1.3

Figure 5.12. Output Results Comparison

All above solutions of the cooling tower design are based on the Mekel relation and are used for the size of tower. The main determinant of the size of the induced draft counterflow cooling tower is the cross-sectional area and the height of the fill; relative overall height and floor space requirements will be modified with different types of cooling towers. In the design calculation of induced draft tower, the

99 fill height 4.75 m (15.6 ft) and cooling tower height 7 m (22.7 ft) corresponding to the

L/G ratio are specified for the required temperature range. In practice the fill crosssectional area and pack height must be selected from standard pack module sizes.

5.4. Operation Considerations

The cooling towers for the induction furnace are installed outdoors location and layout of the cooling tower should be carefully chosen so that recirculation, interference, and air flow are minimized. Recirculation caused by insufficient exhaust air velocity, or insufficient vertical distance between the intake and the exit degrades the tower performance. When two or three cooling towers are used for the required process condition, interference is mainly considered for the tower performance. Interference can be minimized by considering the prevailing wind direction and by properly planning the layout of the cooling towers. Makeup water requirements are a special importance for the cooling tower operation. The total makeup water comprises the water loss by evaporation, drift loss and blowdown loss. It is given by Perry [42] as follows:

Wm = We + Wd + Wb

Equation 5.12

where Wm is the makeup water, m3/min, We = evaporation loss, m3/min, Wd = drift loss, m3/min and Wb = blowdown loss, m3/min. Evaporation loss depends on the cooling capacity and atmospheric conditions. In most cases, it can be estimated by the following equation.

We = 0.00085Wc (T1 – T2)

Equation 5.13

where, Wc = circulating water flow rate, gal/min Drift loss mostly occurs due to the air discharge of the cooling tower and the drift eliminator design, which typically varies between 0.1 and 0.2 percent of the water supplied to the tower. Now development in eliminator design makes it possible to reduce drift loss well below 0.1 percent. In order to avoid a too high concentration of dissolved solids in the circulating water, a part of the water circuit has to be drained continuously. The amount of blowdown can be calculated according to the number of cycles of concentration required to limited scale formation. Cycles of concentration, the ratio of dissolved solids in the recirculating water to dissolved solids in the makeup water, are determined from Perry [42]. They are normally accepted, ranging from three to five cycles.

100

Cycle of concentration =

Wb =

(We + Wb ) Wb

We (Cycles − 1)

Equation 5.14

For the calculation of cooling tower design, makeup water requirements are calculated as follow. From Equation 5.13, We = 0.00085 × 200(91.4 − 82.4) = 1.53 gal/min = 0.005 m 3 /min Using the cycles of concentration (5), blowdown loss is as follows: Wb =

1.53 = 0.3825 gal/min = 0.001 m 3 /min (5 − 1)

Drift loss at the air intake is 0.2 percent as follows: Wd = 200 ×

0.2 = 0.4 gal/min = 0.001 m 3 /min 100

From Equation 5.12, The total makeup water = We + Wd + Wb = 2.31 gal/min = 0.008 m3/min Therefore, the amount of makeup water required for the induced draft cooling tower in the design is 0.008 m3/min (2.31 gal/min).

101

CHAPTER 6 RESULTS AND DISCUSSIONS

6.1. Flow Velocity Calculation Results The first step in the induction furnace process was to calculate the flow velocity passing through the induction coil which is a hollow section of heavy duty and high conductivity copper tubing. In 0.16 ton coreless induction furnace, the diameter of the coil that must be water-cooled due to the highest temperature of the molten metal is 1.27 cm (0.5 in) for both furnace 1 and furnace 2. According to the size and capacity of the molten metal inside the crucible, coil diameter is varied such as 2.54 cm (1 in) for 1 ton furnace and the amount of tons melted is determined by the energy requirements. Schematic the flow branches in pipes were shown in Figure 3.8 for 0.16 ton coreless induction furnace except for the arrangement of actual furnace, capacitor bank and control panel. From the calculating results, the flow velocity passing through the induction coil is obtained 2.45 m/sec (9 ft/sec) for the furnace 1 and furnace 2 in chapter 3. The two velocities of induction furnaces are defined as the same velocity for this case. Bailey [10] and the broad sheet from British Cast Iron Research Association (BCIRA) recommended that all cooling-passages into the induction coil should be designed so that the flow velocity is not less than 1 m/sec (3.281 ft/sec), to prevent any suspended solids settling-out in the system. Therefore, the flow velocities supplied and passed to the induction coils are sufficed for 0.16 ton capacity furnace. Since the large amount of flow velocity passes through the coil, debris, foreign matters and suspended solids cannot be clogged and the cooled-water is continuously circulated at the required differential pressure and rated flow. The flow velocity from the electrical parts such as control panel and capacitor bank is not mainly considered in the calculation method. These flow rate and velocities are specified by the manufacturers. Also the temperature difference between inlet and outlet of these equipments is much lower than the leaving water temperature of the furnace coil. But the operator must monitor to make sure that they

102 do not clog other suspended solids. If the flow velocity and the required amount of water are insufficient or less than 1 m/sec, the temperature of induction coil is immediately raised and the coil may be damaged during the metal melting situation. Hence the great flow velocity and quantity of cooling water are very important to operate the coreless induction furnace and to maximize system reliability and component longevity.

6.2. Cooling Pond Performance The ideal cooling pond performance is defined as one which will maximize the operating time of the induction melting while minimizing the pond volume to remove from the heat of water. A review of the tabulated form of Equation 4.32 shows that the temperature variations within the pond corresponding to the operating time relatively increase; hence, the lumped parameter approach is appropriate. In designing cooling pond, the hot water or inlet temperature into the pond and the cold water or outlet temperature from the pond are used 91.4ºF (33ºC) and 82.4ºF (28ºC), respectively. According to the specified temperature range and the total heat transfer rate, the pond area and volume have been analytically obtained 1,000 ft2 and 6,000 ft3, respectively. By using the resulting data, the cooling pond performance curve is drawn to know the pond capability. This curve is shown in Figure 6.1.

Water temperature in the pond, Degree F

92

90

88

86

Designed Pond

84

82

80 0

1

2

3

4

5

6

7

8

Operating tim e, hr

Figure 6.1. Cooling Pond Performance Curve

103 From the above Figure, the water temperature in the pond is relatively increased when the operating time using 0.16 ton coreless induction furnace has increased, which is often an important consideration. Normally, the melting time for 0.16 ton coreless induction furnace is consumed 1 to 1.5 hours per 0.16 ton according to the supplied power, related voltage and the frequency variation. The designed pond curve on graph represents the alternative melting of two induction furnaces. To keep a fairly discharge temperature from the pond, the time limitation must be suitably chosen with regard to the particular metallurgical process in view. On the other hand, since the water surface temperature varies with the distance from the discharge point in the pond, the heat transfer rate from the pond surface is not uniform. For a given surface area 1,000 ft2, the efficient cooling pond is a shallow type enough to allow density currents to spread the warm water effluent efficiently over the surface of the pond. The water inlet should be located at the coolest point of the pond surface, and there should be minimal mixing of the hot water with the cooler deeper water to ensure the warmest surface layer temperature possible for highest surface cooling. The time required for water to pass through the pond depends on the rate of pumping and the pond volume while the temperature drop from the induction furnace discharge to the induction coil intake depends on the rate of pumping and the pond surface area. In this study the performance curve of cooling pond may be changed due to the effect that is meteorological parameters such as ambient air temperature, wet bulb temperature, relative humidity, wind speed and solar radiation. The performance curve from Figure 6.1 is considered based on a horizontal surface location of Yangon (16.45 N) during the month of March, one of the hottest months of Myanmar. For a given cooling load, the total volume of cooling pond is 6,000 ft3 with a depth of 6 ft. This is a practical constraint on the factor of cooling pond and the results indicate an increase in the performance curve due to the different temperature which corresponds to the cooling capacity. Another study on the performance curve is considered by using the various pond configurations for the given cooling duty. The pond models are denoted as type A, B, C, D, E and F according to their different area and depth. Firstly, Type A is defined as the designed pond at one condition owing to the pond volume 6,000 ft3 and surface area 1,000 ft2. For the design of this type, it has been calculated from the required temperature range and the climatic condition of the locating pond site. This

104 pond is built with the depth of 6 ft. Type B is referred to the pond surface 1,200 ft2 and depth of 5 ft. Similarly, type C has pond surface area 800 ft2 and depth of 7.5 ft. With the depth of 10 ft, type D is made of surface area 600 ft2 as the smallest model. Type E is created with the large pond surface area 1,400 ft2 and the depth of 4.285 ft. All above types have the fixed pond volume 6,000 ft3. Type F is referred to as the pond surface area 2,000 ft2 and the depth of 3 ft. This type also has the pond volume 6,000 ft3 and its area is larger than all above types. After that, the relationship between the increasing temperature within the pond and the operating time depending on the melting cycle is evaluated previously from the pond designing parameters. The correlations of different configurations have been replaced in the Equation 4.28 to determine the water temperature and the operating time. Using the correlations of the various conditions, the effect of cooling pond configurations is plotted on a graph as the operating time versus the water temperature in the pond and it is illustrated in Figure 6.2.

Water temperature in the pond, Degree F

92

90 Type A

88

Type B Type C

86

Type D Type E

84

Type F

82

80 0

1

2

3

4

5

6

7

8

Operating tim e, hr

Figure 6.2. Effect of Cooling Pond Configurations

From the above figure, the results comparing the configuration effects can be determined not only at the design condition of pond but also at the other conditions. The pond temperature is normally maintained about 82.4ºF before starting the induction furnace for melting process. During the one hour operating time, the temperature differences of all types are insignificant expect type F. At the six hours

105 operating time, the rising temperature 87.87 ºF, 87.59ºF, 88.15ºF, 88.43ºF, 87.32ºF and 82.57ºF are occurred in type A, B, C, D, E and F, respectively. From the effect of cooling pond configurations, the temperatures of type C and type D are slightly higher than the other four types because they are used for a small pond surface area with a relatively depth. The lesser surface area affects the evaporative heat loss that is the primary heat transfer mechanism contributing to pond cooling. The type B and Type E give the lower temperature rise because of the large surface areas. Compared with the temperature to type B and type E, the designed pond (type A) is slightly high but it is less than the type C and type D. Among them, type F gives nearly constant temperature in the pond because of the larger surface area. As a result, the varying pond surface area is an important factor and observed from the investigation of pond cooling. This result is substantially true to the recommendation of Cerco [32], in which the surface area extent of the pond affects the performance of cooling pond and it can improve the heat transfer from the pond surface. It can control the water temperature in the pond and maximize the evaporative cooling effect of the pond surface. Although the large amount of area should be used in the pond construction site, the land space condition and construction cost become a problem. By compromising the above factors, cooling pond should be constructed as economically efficient as possible. According to the layout of foundry, the surface area 1,000 ft2 and the depth of 6 ft are efficient for 0.16 ton coreless induction furnace. On the other hand, the amount of heat removed from the pond is relatively depended upon the meteorological parameters such as relative humidity, wet bulb temperature, dry bulb temperature, solar radiation and wind speed. Accordingly, the total volume of cooling pond 6,000 ft3 with a depth of 6 ft is applied for 0.16 ton coreless induction furnace. This is a practical constraint imposed by the process. By imposing constraints on the design factor of cooling pond, the cooling efficiency has been optimized, and the required cooling temperature is a fairly constant discharged to the induction coil and other equipment. Consequently, there is a major factor of cooling pond area to maintain a fairly constant discharge temperature throughout the operating time. By adjusting the operating time, the desired volume and area corresponding to the pond cooling capacity can be obtained. Note that there is a forced withdrawal of cooling water and further make up flow rate, caused by incapacity of the pond to provide the cooling

106 requirements. Subsequently, the water must be clean, since debris will create clogging in the small passages in the electrical parts. The poor water quality should not be used in cooling system of induction furnace because it is caused by scaling formation, product corrosion and fouling. All cooling paths always should be taken care and monitored throughout the induction melting process.

6.3. Cooling Tower Performance In the cooling system study for electric induction furnace, one design of cooling tower takes into account of the tower performance with the surrounding air by governing the ratio of the weights of air to water and the time of contact between water and air. The cooling tower performance influences on the wet bulb temperature which is the vital importance in design and it is based on relevant meteorological information, but must be modified to take account of the consequences of the design figure being exceeded under operating conditions. As the review of previous section 5.3, the water to air ratio (1.1813) and the tower characteristic KxaV/L (1.4534) have been determined based on the wet bulb temperature 77ºF (25ºC), the water flow rate 200 gal/min and the range 9ºF (5ºC) between the hot and cold water temperature. By varying the wet bulb temperature in the ambient air, the result of cooled-water putting out from the tower may be changed. For the designed cooling tower, the wet bulb temperature is plotted to yield a performance curve with the help of CTI Toolkit v.30 Software. A review of the wet bulb temperature plots in Figure 6.3 as cooling tower performance shows that the designed cooling tower with induced draft counterflow type falls to the temperature 9ºF (5ºC) range. From the performance curve, the cooling water temperature 82.4ºF (28ºC) can be occurred in the wet bulb temperature 77ºF (25ºC) as the design condition. By using the performance curve, it is determined to know the cooled-water temperature on the designed wet bulb temperature as well as the minimal allowable temperature. Moreover, the other different temperature ranges 8ºF, 10ºF and 11ºF are plotted similarly on the performance curve at the condition of constant water flow rate 200 gal/min, the water to air ratio and the tower characteristics. It is also illustrated in Figure 6.3 that is the type of performance curve furnished by the design of induced draft cooling tower.

107

Cold water temperature, ºF

87 86 85 84

8ºF Range

83

9ºF Range

82

10ºF Range

81

11ºF Range

80 79 78 77 74

75

76

77

78

79

80

81

Wet bulb tem perature, ºF (at 200 gal/m in)

Figure 6.3. Comparison of Different Temperature Range at Constant Water Quantity

From the Figure 6.3, the required temperature range of the designed tower is compared favorably to the variations with changes in wet bulb and hot water temperatures while the water quantity is maintained constant. Using the comparison of cooling tower performance curves, the hot water and cold water temperature corresponding to the wet bulb temperature can be also determined extensively for the induced draft cooling tower designs.

6.4. Process Influence on Tower To apply the counterflow cooling tower as well as the extracting of cooledwater across the fill, the design model has been described in section 5.3. For the continuous operating time and mass production in melting process, cooling tower is used instead of cooling pond. Based on the parameters of cooling pond, the design conditions of cooling tower have been known as a result which consists of tower height 22.7 ft, fill height 15.6 ft, fill volume 499 ft3, water to air ratio 1.1813 and approach temperature 5.4ºF. As shown in Figure 5.10, the area between H 1 H 2 H ′2 H 1′ is not only the tower characteristic but also the key in the design of tower. By varying this area, the tower design may be changed for the process requirements. The comparison of three statements: calculated design, changing the L/G ratio, and close approach are shown in Table 6.1. In this table, the wet bulb temperature is emphasized on 77ºF for all conditions.

108 Table 6.1. Comparison of Process Variables in Tower Design Calculated Design

L/G ratio Changing

Close Approach

Condition

Condition

Condition

77

77

77

91.4

91.4

88.4

82.4

82.4

79.4

Range, ºF

9

9

9

Approach, ºF

5.4

5.4

2.4

L/G ratio

1.1813

1.5

1.1813

KaV/L

1.4534

1.9027

3.7656

Tower height, ft

22.7

37.74

58.82

Fill height, ft

15.6

19.84

15.62

Fill volume, ft3

499

634

450

Parameters Wet bulb temperature, ºF Hot water temperature, ºF Cold water temperature, ºF

If the ground area is too limited or the ground area is a problem in the foundry shop, such as when a cooling tower is erected atop a building, the L/G ratio should be changed to meet the design requirement. In Table 6.1, the L/G ratio changing condition employs a relatively high liquid loading without increasing the air quantity. This is increased the slope of air operating line H1 – H2 to H1 – (H2) as illustrated in Figure 6.4. 70


65 60

H2'

55 50

(H2)

40

H2

H1'

45

L/G

(L/G)1

H1

Water operating line Air operating line (calculated design) Air operating line (L/G changing)

35 30 25 79.4 80.4 81.4 82.4 83.4 84.4 85.4 86.4 87.4 88.4 89.4 90.4 91.4 92.4 93.4

20

Water tem perature, ºF

Figure 6.4. Enthalpy-Temperature Diagram of Air and Water by changing L/G ratio

109 The slope H1 – H2 from the calculated design is changed to H1 – (H2) significantly. Since the area H 1 (H 2 )H ′2 H 1′ between the saturation and operating lines is smaller than the calculated designed area H 1 H 2 H ′2 H 1′ , the tower potential is decreased. As a result, the cross sectional area of the tower decrease, but the height of the tower is more required in design. The great amount of fill volume is also used in tower construction. Another condition is the close approach that uses the same L/G ratio (same slopes) and equal ranges of 9ºF for the removal of the same process heat load. This condition is shown in Figure 6.5. The air operating line is removed from H1 – H2 to (H1) – (H2) with the same inlet air but between the temperatures of 88.4ºF and 79.4ºF instead of 91.4ºF and 82.4ºF. It is occurred that the area between the saturation curve and the operating line is greatly decrease by (H1) – (H2). As a result, the tower characteristic value increases as 3.7656 and also decreases the tower potential. Therefore, the higher tower height is needed in the construction site but the requirement of fill volume is quite near to the calculated design fill. 70


65 60

(H2)'


55 50

(H2) (H1)'

H2

45

Air operating line (calculated design)

40 (H1) H 1 35

Air operating line (Close approach)

30 25 79.4 80.4 81.4 82.4 83.4 84.4 85.4 86.4 87.4 88.4 89.4 90.4 91.4 92.4 93.4

20

Water tem perature, ºF

Figure 6.5. Enthalpy-Temperature Diagram of Air and Water at the Close Approach Condition

110

CHAPTER 7 CONCLUSION, RECOMMENDATION AND FURTHER SUGGESTIONS

7.1. Conclusion Cooling ponds are one of the economically competitive alternatives for the dissipation of waste heat from electric induction furnaces, power plant, natural gas generating stations and other industries. Based on the conceptual studies, cooling pond system has been designed and constructed for 0.16 ton coreless induction furnace. As a result of the strong interaction among the several process variables involved, the cooling pond system is very complex. Besides, environmental heat transfer mechanisms that are simulated by the pond system include solar radiation heat gain, heat transfer due to evaporation, convection heat transfer to the atmosphere, long-wave radiation heat transfer and conductive to the surrounding soil. The solution scheme involves the calculation of cooling pond area, volume resulting from the overall energy balance, the internal flow rate passing through the induction coil and pump selection methods. As a result, the cooling pond surface area 1,000 ft2 and volume 6,000 ft3 can be given satisfactorily for the 0.16 ton melting capacity of two coreless induction furnaces during the six hours operating time without overheating and rising the water temperature in the pond. According to the consideration of flow velocity, the velocity about 2.8 m/sec of the maximum melting temperature of molten metal is obtained. The corresponded pump for cooling pond system is the centrifugal pump sized of 11 kW, pumping capacity 0.69 m3/min and total head 46 m. Performance curves can be used in the selection of required pond size and the designation of operating time corresponding to the furnace capacity. Results and calculations are based on different climatic conditions and time limitation. The most important influence on the cooling pond performance is pond surface area itself, but not its depth. The cooling system study specially pointed out that cooling pond can be accepted as the heat exchanger apparatus giving the maximizing evaporative cooling effect for 0.16 ton coreless induction furnace because it is inexpensive in construction,

111 simple in design and it can be used satisfactorily during warm weather (ambient temperatures higher than water temperatures). Although the algae growth and inwater barriers may occur in the pond, they can easily protect with the helps of bioaugmentation and biofiltration process or other chemical treatments. To achieve water temperatures at or near the wet bulb ambient temperature during the peak temperature month, cooling pond also offers the most satisfactory solution. Nevertheless, cooling pond system is suitable for the specified operating time in electric induction furnaces, but it is insufficient for fulfilling the continuous running time and mass production in melting process. Instead of cooling pond, cooling tower system presented in the design of induced draft counterflow type should be used efficiently and beneficially for coreless induction furnace. The design calculations have been presented to construct the cooling tower as well as the supplementary consideration of the use of cooling pond. Through the study, it is shown that tower characteristic is the primary important factor. Nowadays, the industrial sector is rapidly growing in Myanmar. To produce the high quality metal products, electric induction furnaces are great tools for the melting metals and casting applications. In electric induction furnaces, the cooling system is an essential part to prevent overheats and damage of the induction coil. So, the extensive emphasis of cooling system should be tried and considered to be more efficient and effective rather than the present pond system. Finally, the general trend studied on the cooling system is that the induction coil of electric melting furnace in foundry sectors is an important role for fulfilling the cooling cycle completely.

7.2. Recommendation The cooling system reviewed and overall assessment conducted to prevent the damaging of induction coil for the furnaces also yielded some work practice recommendations, which would assist foundries melting with coreless induction furnace. The following are some of these recommendations concerning the cooling system of induction melting furnaces. •

Cooling pond system can be used for producing small amount of molten metal such as 0.16 ton or 0.5 ton because of the low capital cost, but it requires the larger surface area to maximize the evaporative cooling effect of pond.

112 •

Cooling tower system should be used for the continuous operating time in induction furnace and mass production melting size of more than one ton capacity. The space requirement of cooling tower is less than the cooling pond although it has the higher constructing cost.

7.3. Further Suggestions Further research suggests in the following factors. •

The baffles with the different shapes should be applied to enhance the cooling rate and longer residence time. With the baffle in the pond, the efficiency of cooling pond can also be improved by introducing the proper slope under the bottom of the pond to reduce dilution and by replenishing the quality make-up water directly.

•

One of the heat exchangers such as shell and tube type heat exchanger, spiral heat exchanger and plate and fame heat exchanger should be used for the reduction in foundry complexity and space saving. The use of alternative exchanger technologies in the cooling system of induction furnace will give not only the superior thermal performance but also compact design and multiple capabilities.

113

REFERENCE LIST

[1]

Mortimer, J. 1994. “Automating the Melting Process”: A Key to Competing (June).

[2]

Horwath, J. et al. 1994 “Coreless Induction Melting Optimization Study Prepared for the EPRI Center for Material Production”.

[3]

American Cast Metals Association et al. 1990. “Foundry Energy Conservation Workbook”.

[4]

A.D. Little. 1992. “Alternative Melting Technologies in Scrap-based Steel making EPRI Center for Materials Production”.

[5]

Smith and Bullard. 1995. “Best Melting Practices in Medium Frequency Coreless Induction Furnace”. The Foundryman (July).

[6]

American Foundrymen's Soecity, Inc. 1982. “Foundry Energy Management”.

[7]

Booth, M. 1996. “Developments in Electric Ferrous Technology”. Foundryman (March).

[8]

Process Metallurgy International. 1998. “Personal communication”.

[9]

“Energy Conservation in Iron Casting Industry”. 1998. Hosted by Ministry of Industry Socialist Republic of Viet Nam and Organized by The Energy Conservation Center, Japan (ECC), Hanoi.

[10] Bailey, D. 1977. “Cooling-water Systems for use with Electric Induction Furnaces”, British Foundryman, v. 70, Pt. 2, Februry. [11] Robiette, A. G. E. 1955. “Electric Melting and Smelting Practice”, Charles Griffin and Company Limited, London. [12] Loew, E.A. 1954 “Direct and Alternating Currents Theory and Machinery”, 4th edt, McGraw Hill Book Company Inc.,New York. [13] Paschkis, V and Persson, J. 1960 “Industrial Electric Furnaces and Appliances”, 2nd etd, Interscience Publishers Inc., New York. [14] Marks, L. S. 1967. “Standard Handbook for Mechanical Engineers”, 7th edt, McGraw Hill Book Company Ltd, Tokyo. [15] Dwight, H. D. 1922. Trans. AIEE, 41, 203.

114 [16] Phelps, S. M. 1946. “American Refractories Institute”, Technical Department, Mellon Institute, Pittsburgh. [17] Kern, D Q. 1997. “Process Heat Transfer”, Tata McGraw-Hill Publishing Company, NewDelhi, India. [18] Incropera, F. P. and DeWitt, D. P. 1990. “Fundamentals of Heat and Mass Transfer”, 2nd edt. [19] Mott, R.L. 1979. “Applied Fluid Mechanics”, 2nd edt, Charles E. Merrill Publishing Company, Ohio. [20] Chopey, N. P. 1994. “Handbook of Chemical Engineering Calculations”, 2nd edt, McGraw Hill International Edition, Singapore. [21] Church, A. H. 1944. “Centrifugal Pumps and Blowers”, John Wiley & Sons, Inc., New York. [22] Peerless Pump, Brochure B-4003. 1979. “System Analysis for Pumping Equipment Selection”, U.S.A. [23] Dickinson, C. 1988. “Pumping Manual”, 8th edt. Surrey, England: Trade & Technical Press, Ltd. [24] Walker, R. 1972. “Pump Selection”. Ann Arbor, Mich.: Ann Arbor Science Publishers. [25] Stepanoff, A. J. 1965. “Pumps and Blowers”, Two-Phase Flow, John Wiley & Sons, Inc., New York. [26] Hicks, T. G., and Edwards, T. W. 1971. “Pump Application Engineering”, McGraw Hill Company, Inc., New York. [27] Daily, J. W. 1950. “Hydraulic Machinery”, in Rouse, H., ed., Engineering Hydraulic, New York: Wiley. [28] Fox, R. W. and McDonald, A. T. 1994. “Introduction to Fluid Mechanics”, 4th edt, SI version, John Wiley and Sons. Inc., New York. [29] Streeter, V. L.; Wylie, E. B. and Besford, K. W. 1998. “Fluid Mechanics”, 9th edt, WCB/ McGraw Hill Company, Singapore. [30] Evans, J. E. , Pentair Pump Group, “Pracific Liquid & Air System”, [31] Harleman, D. R. F.; Jirka, G.H. and Stolzenbach, K. D. 1977. “Theoretical and Experimental Investigation of Emergency Heat Releases from Floating Nuclear Power Plants”, R. M. Parsons Laboratory for Water Resources

115 and Hydrodynamics Technical Report No. 206, Department of Civil Engineering, M.I.T. [32] Cerco, C. F. 1977. “Experimental and Analytical Study of the Design of Shallow Cooling Ponds”, Department of Civil Engineering, M.I.T. [33] Jirka, G. H.; Abraham, G. and Harleman, D. R. F. 1975. “An Assessment of Technique for Hydrothermal Prediction”, R. M. Parsons Laboratory for Water Resources and Hydrodynamics Technical Report No. 203, Department of Civil Engineering, M.I.T. [34] Ryan, P. J. and Harleman, D. R. F. 1973. “An Analytical and Experimental Study of Transient Cooling Pond Behavior”, R. M. Parsons Laboratory for Water Resources and Hydrodynamics Technical Report

No.161,

Department of Civil Engineering, M.I.T. [35] Edinger, J. E.; Brady, D. K. and Geyer, J. C. 1974. “Heat Exchange and Transport in the Environment”, Report No. 14, Research Project RP-49, Electric Power Research Institute, Palo Alto, CA. [36] Shanahan, P 1984. “Water Temperature Modeling”: A Practical Guide, Environmental Research and Technology, Inc. Concord, Massachusetts. [37] TVA (Tennessee Valley Authority). 1972. “Heat and Mass Transfer Between a Water Surface and the Atmosphere”. Water Resources Research Laboratory Report No. 14. Tennessee Valley Authority, Division of Water Control Planning, Engineering Laboratory, Norris, TN. April. [38] Duffie, J. A. and Beckman, W.A. 1994. “Solar Engineering of Thermal Processes”, 2nd Edt, John Wiley, New York. [39] Rietveld, M. R. 1978. “A New Method for Estimating the Regression Coefficients in the Formula Relating Solar Radiation to Sunshine Agricultural Meteorology”, 19, 243-252. [40] Magal, B. S. 1999. “Solar Power Engineering”, Fourth reprint, TATA McGraw Hill Publishing Company Limited, Bombay. [41] Brunt, D. 1932. “Notes on Radiation in the Atmosphere. Quarterly Journal of the Royal Meteorological Society”, 58, pp. 389 – 418. [42] Perry, Green, Dom W et al. 1995. “Chemical Engineering Handbook”, 6th edt, McGraw Hill Company, New York. [43] Hensley, J. C. 1998. “Cooling Tower Fundamentals”, Marley Cooling Technologies, Inc. Overland Park, Kansas U.S.A.

116 [44] London, A. L.; Mason, W. E. and Boetter, L. M. K. 1940. “Performance Characteristics of a Mechanically Induced Draft Counterflow Packed Cooling Tower”, Trans. ASME, vol.63, p-41. [45] Merkel, F. 1925. Verdunstungskuchlung, VDI Forschungsarbeiten, No 275, Berlin. [46] Wang, S. K. 1993. “Handbook of Air Conditioning and Refrigeration”, McGraw Hill Company, U.S.A. [47] Baker, D. R. and Shryock, H. A. 1961. “A Comprehensive Approach to the Analysis of Cooling Tower Performance”, Technical Bulletin, R-61-P-13, Markey Company, U.S.A. [48] Moody, L. F. 1944. “Friction Factors for Pipe Flow”, Transactions of the ASME, 66, 8, November, pp, 671 – 684. [49] Crane Company. 1969 “Flow of Fluids through Valves, Fittings, and Pipe”, Technical paper No 410, Chicago. [50] Hunsaker, J. C. and Rightmire, B. G. 1947. 1st edt, McGraw Hill Company Inc., New York and London.

117

APPENDIX A PROGRAM

The followings are the Matlab program to evaluate the volume flow rate for each branch in pipes and friction losses calculation.

clc

%##### Volume Flow Rate Calculation of Branches of Pipe Line for 0.16 Ton Coreless Induction Furnace #####

fprintf('##### Volume Flow Rate Calculation of Branches of Pipe Line for 0.16 Ton Coreless Induction Furnace ##### \n'); QT=input('\n Sum of two flow rate , m3/min , QT = '); L1=input('Pipe length of one branch , m , L1 = '); L2=input('Pipe length of another branch , m , L2 = '); D1=input('Pipe diameter of one branch , m , D1 = '); D2=input('Pipe diameter of another branch , m , D2 = '); fprintf('\n We must be assumed the first volume flow rate of Q1 snd Q2. \n'); Q1=input('Volume flow rate on one branch , m , Q1 = '); Q2=input('Volume flow rate on another branch , m , Q2 = '); Nu1=input('Kinematic viscosity of one branch at -degree Celcius , m2/min , Nu1 = '); Nu2=input('Kinematic viscosity of another branch at -degree Celcius , m2/min , Nu2 = ');

Re1=(4*Q1)/(pi*Nu1*D1); Re2=(4*Q2)/(pi*Nu2*D2);

fprintf('\n Reynold number of one branch Re1 = %g ',Re1); fprintf('\n Reynold number of another branch Re2 = %g ',Re2); fprintf('\n Now we take the pipe wall roughness of the branches as GI or PVC. \n');

118 E1=input('Pipe wall roughness of one branch , mm , E1 = '); E2=input('Pipe wall roughness of another branch , mm , E2 = '); fprintf('\n By using these data. The relative roughness of the branches are: \n');

z=E1/(D1*1000); x=E2/(D2*1000);

fprintf('\n E1/D1 = %g ',z); fprintf('\n E2/D2 = %g ',x); fprintf('\n From Moody"s diagram, using Reynold number and Relative roughness, we have: \n'); f1=input('Friction factor of one branch , f1 = '); f2=input('Friction factor of one branch , f2 = ');

hm1=f1*L1/D1; hm2=f2*L2/D2;

a=input('\n Equivalent length of elbow according to the number of unit in one branch (Le/De)1 = '); b=input('\n Equivalent length of valve according to the number of unit in one branch (Le/Dv)1 = '); c=input('\n Equivalent length of Tee according to the number of unit in one branch (Le/Dt)1 = '); d=input('\n Equivalent length of sudden contraction according to the number of unit in one branch (Le/Dsc)1 = '); e=input('\n Equivalent length of sudden enlargement according to the number of unit in one branch (Le/Dse)1 = '); a2=input('\n Equivalent length of elbow according to the number of unit in one branch (Le/De)2 = '); b2=input('\n Equivalent length of valve according to the number of unit in one branch (Le/Dv)2 = '); c2=input('\n Equivalent length of Tee according to the number of unit in one branch (Le/Dt)2 = ');

119 d2=input('\n Equivalent length of sudden contraction according to the number of unit in one branch (Le/Dsc)2 = '); e2=input('\n Equivalent length of sudden enlargement according to the number of unit in one branch (Le/Dse)2 = '); hn1=f1*(a+b+c+d+e); hn2=f2*(a2+b2+c2+d2+e2); hT1=hm1+hn1; hT2=hm2+hn2;

fprintf('\n Total losses on one branch , m = %g \n',hT1); fprintf('\n Total losses on another branch , m = %g \n',hT2);

A1=pi*D1^2/4; A2=pi*D2^2/4; PA=1+hT1; PB=1+hT2; w=sqrt(PB/PA); A=[A1 A2;1 -w]; B=[QT;0]; X=inv(A)*B; V1=X(1); V2=X(2); QT1=V1*A1+V2*A2; Q11=V1*A1 Q22=V2*A2

while Q1 ~= Q11 ;

f1=input('Friction factor of one branch , f1 = '); f2=input('Friction factor of one branch , f2 = ');

Q1=Q11;

hm1=f1*L1/D1;

120 hm2=f2*L2/D2; hn1=f1*(a+b+c+d+e); hn2=f2*(a2+b2+c2+d2+e2); hT1=hm1+hn1; hT2=hm2+hn2; A1=pi*D1^2/4; A2=pi*D2^2/4; PA=1+hT1; PB=1+hT2; w=sqrt(PB/PA); A=[A1 A2;1 -w]; B=[QT;0]; X=inv(A)*B; V1=X(1); V2=X(2); Q11=V1*A1 Q22=V2*A2 A= abs(Q11-Q1);

if A < 0.00009;

V11=V1/60; V22=V2/60; fprintf('\n Flow velocity on one branch = %g m/min \n',V1); fprintf('\n Flow velocity on another branch = %g m/min \n',V2); fprintf('\n Flow velocity on one branch = %g m/sec \n',V11); fprintf('\n Flow velocity on another branch = %g m/sec \n',V22); fprintf('\n Volume flow rate on one branch = %g m3/min \n',Q1); fprintf('\n Volume flow rate on another branch = %g m3/min \n',Q2);

break end end

121

APPENDIX B GRAPHS

Figure B.1. Skin Effect in Isolated Rounded Copper Tubings Source: Loew (1954)

Source: Condensed from data of Gould pumps, Inc., SI values added by Handbook editor

Figure B.2. Composite Rating Chart for a Typical Centrifugal Pump

2“× 2“ - 7

Discharge connection size (nominal inch)

Suction connection size (nominal inch)

Casing class – Nominal size (in inches) of largest impeller

Form of pump designation:

122

Source: Moody (1944)

Figure B.3. Moody Diagram

123

124

Figure B.4. Nomograph of Cooling Tower Characteristics Source: Perry et al. (1995)

125

APPENDIX C TABLES

Table C.1. Pipe Roughness – Design Values Material

Roughness, ∈ (mm)

Roughness, ∈ (ft)

Glass, plastic

Smooth

Smooth

Riveted steel

0.9 – 9

0.003 – 0.03

Concrete

0.3 – 3

0.001 – 0.01

Wood stave

0.18 – 0.9

0.0006 – 0.003

Cast iron

0.25

0.00085

Galvanized iron

0.15

0.0005

Asphalted cast iron

0.12

0.0004

Commercial steel of wrought iron

0.046

0.00015

Drawn tubing (Copper, brass, lead tubing)

0.0015

0.000005

Source: Streeter and Wylie (1998)

Table C.2. Resistance in Valves and Fitting expressed as Equivalent Length in Pipe Diameters, Le/D Type

Equivalent length in pipe diameter, Le/D

Globe valve – fully open

340

Angle valve – fully open

145

Gate valve – fully open

5

Check valve – swing type

135

Check valve – ball type

150

Foot valve with strainer hinged disk

75

Butterfly – fully open

40

90º standard elbow

30

90º long radius elbow

20

90º street elbow

50

Ball valve

3

45º standard elbow

16

45º street elbow

26

Close return bend

50

Standard tee – with flow through run

20

– with flow through branch

60

Source: The Crane Co. Chicago, Ill (1969), Hunsaker and Rightmire (1947), and Mott (1979)

126 Table C.3. Properties of Water at Various Temperatures (Referred to water at 68ºF, weighting 62.32 lb/ft3) Vapor Specific pressure, ft gravity absolute 50 0.41 1.002 60 0.59 1.001 70 0.84 1 80 1.17 0.998 90 1.62 0.997 100 2.2 0.995 110 2.96 0.993 120 3.95 0.99 130 5.2 0.988 140 6.78 0.985 150 8.74 0.982 160 11.2 0.979 170 14.2 0.975 180 17.85 0.972 190 22.3 0.968 200 27.6 0.965 210 34 0.961 220 2.49 41.45 0.957 230 6.07 50.35 0.953 240 10.27 60.75 0.948 250 15.12 73 0.944 260 20.72 87.35 0.939 270 27.15 103.3 0.935 280 34.48 122 0.93 290 42.85 144 0.925 300 52.3 169 0.92 Source: http://www.pacificliquid.com/ pumpintro.pdf Temperature, ºF

Vapor pressure, psig

Table C.4. Comparison of Different Roofing Materials, Showing White (Cooler Option)

Roofing material Asphalt shingle Clay tile Concrete tile Cementitious shingle Metal sheet or shingle

Reflectance value 5 – 15 % 25 – 35 % 10 – 30 %

Reflectance for white (cooler option) 31 – 35 % 70 – 80 % 70 – 80 %

Cost increase for white (cooler option) <1% ~ 35 % ~ 20 %

10 – 30 %

60 – 80 %

None

70 %

70 – 80 %

None

Source: Green Seat, Energy Star® Homes Project

Steady 15 (4.6)

Increase Depends on specific speed

How decreased head affects: Capacity Power input

None Decrease

None Increase

Small to medium

Medium

Viscous, nonabrasive

Steady 22 (6.7)

Screw and gear

Rotary

Source: Handbook of Chemical Engineering Calculations, Chopey (1994)

Decrease Depends on specific speed

Small to largest available

Low to high

Clean, clear, dirty, abrasive; liquid with high solids content

Steady 15 (4.6)

Axial flow

How increased head affects: Capacity Power input

Usual capacity range

Discharge pressure range

Liquids handled

Discharge flow Usual maximum suction lift, ft (m)

Volute and diffuser

Centrifugal

Table C.5. Characteristics of Modern Pumps

Clean and clear

Pulsating 22 (6.7)

Double acting power

Pulsating 22 (6.7)

Triplex

None Increase None Decrease

Decrease Increase Small increase Decrease

Relatively small

None Decrease

None Increase

Low to highest produced

Pulsating 22 (6.7)

Direct acting steam

Reciprocating

127

Design and Construction of an Induction Furnace (Cooling Sys

Recommend Documents